Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.

fw001

The Foundations of Physical

Organic Chemistry: Fifty Years of
the James Flack Norris Award

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.fw001

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 ACS SYMPOSIUM SERIES 1209

The Foundations of Physical

Organic Chemistry: Fifty Years of
the James Flack Norris Award
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.fw001

E. Thomas Strom, Editor

University of Texas at Arlington
Arlington, Texas

Vera V. Mainz, Editor

University of Illinois at Urbana-Champaign
Urbana, Illinois

Sponsored by the
ACS Division of the History of Chemistry

American Chemical Society, Washington, DC

Distributed in print by Oxford University Press

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Library of Congress Cataloging-in-Publication Data

Names: Strom, E. Thomas, editor. | Mainz, Vera V., editor. | American

Chemical Society. Division of the History of Chemistry.
Title: The foundations of physical organic chemistry : fifty years of the
James Flack Norris Award / E. Thomas Strom, editor, University of Texas at
Arlington, Arlington, Texas, Vera V. Mainz, editor, University of Illinois
at Urbana-Champaign, Urbana, Illinois ; sponsored by the ACS Division of
the History of Chemistry.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.fw001

Description: Washington, DC : American Chemical Society, [2015] | [Oxford] :

Distributed in print by Oxford University Press | Includes bibliographical
references and index.
Identifiers: LCCN 2015041510 (print) | LCCN 2015046593 (ebook) | ISBN
9780841230712 | ISBN 9780841230729 ()
Subjects: LCSH: James Flack Norris Award in Physical Organic Chemistry. |
Chemistry, Organic--United States--History. | Chemistry, Organic--Awards.
| Chemistry--United States--History. | Chemistry--Awards.
Classification: LCC QD248.5.U6 F68 2015 (print) | LCC QD248.5.U6 (ebook) |
DDC 547/.13--dc23
LC record available at http://lccn.loc.gov/2015041510

The paper used in this publication meets the minimum requirements of American National
Standard for Information Sciences—Permanence of Paper for Printed Library Materials,
ANSI Z39.48n1984.

Copyright © 2015 American Chemical Society

Distributed in print by Oxford University Press

All Rights Reserved. Reprographic copying beyond that permitted by Sections 107 or 108
of the U.S. Copyright Act is allowed for internal use only, provided that a per-chapter fee of
$40.25 plus $0.75 per page is paid to the Copyright Clearance Center, Inc., 222 Rosewood
Drive, Danvers, MA 01923, USA. Republication or reproduction for sale of pages in this
book is permitted only under license from ACS. Direct these and other permission requests
to ACS Copyright Office, Publications Division, 1155 16th Street, N.W., Washington, DC
20036.

The citation of trade names and/or names of manufacturers in this publication is not to be
construed as an endorsement or as approval by ACS of the commercial products or services
referenced herein; nor should the mere reference herein to any drawing, specification,
chemical process, or other data be regarded as a license or as a conveyance of any right
or permission to the holder, reader, or any other person or corporation, to manufacture,
reproduce, use, or sell any patented invention or copyrighted work that may in any way be
related thereto. Registered names, trademarks, etc., used in this publication, even without
specific indication thereof, are not to be considered unprotected by law.

PRINTED IN THE UNITED STATES OF AMERICA

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Foreword
The ACS Symposium Series was first published in 1974 to provide a
mechanism for publishing symposia quickly in book form. The purpose of
the series is to publish timely, comprehensive books developed from the ACS
sponsored symposia based on current scientific research. Occasionally, books are
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.fw001

developed from symposia sponsored by other organizations when the topic is of

keen interest to the chemistry audience.

Before agreeing to publish a book, the proposed table of contents is reviewed

for appropriate and comprehensive coverage and for interest to the audience. Some
papers may be excluded to better focus the book; others may be added to provide
comprehensiveness. When appropriate, overview or introductory chapters are
added. Drafts of chapters are peer-reviewed prior to final acceptance or rejection,
and manuscripts are prepared in camera-ready format.

As a rule, only original research papers and original review papers are
included in the volumes. Verbatim reproductions of previous published papers
are not accepted.

ACS Books Department

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Preface
The symposium on which this book is based came about because one of
us (Tom Strom) became increasingly concerned about the status of his chosen
discipline, physical organic chemistry. With the recent flowering of organic
synthesis, physical organic chemistry seemed to be shrinking or perhaps it was
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.pr001

just being absorbed into the tool kit of the synthetic chemist. The only Nobel
Prize that can be reasonably attributed to a physical organic chemist is the 1994
award to George Olah, although Jeffrey I. Seeman has recently made a strong
case that R. B. Woodward was actually a physical organic chemist in disguise (1).
However, it is clear that Woodward’s 1965 Nobel Prize was awarded because of
his prowess in synthetic chemistry.
Tom decided to arrange a symposium for the ACS History Division that
examined the history and fundamentals of physical organic chemistry. He asked
his friend Jeffrey Seeman to be co-organizer. In examining themes for such
a symposium, Seeman noted that 2014 would see the awarding of the 50th
James Flack Norris Award in Physical Organic Chemistry. It became clear that
the research carried out by the winners of this award, sponsored by the ACS
Northeastern Section, gave insight into the fundamentals of the discipline. Jeff
and Tom decided on a two stage symposium: presentations by early winners of
the award and a panel discussion by recent award winners on the topic, “Whither
Physical Organic Chemistry.” The symposium then came together rather quickly
and was scheduled for the March, 2014, ACS National Meeting in Dallas, TX.
James Flack Norris was an early physical organic chemist, before the
discipline received its name. His picture is on the cover of this book, and
Arthur Greenberg led off the symposium with a talk on Norris. Then came talks
from Norris Award winners Ned Arnett, Ronald Breslow, Andrew Streitwieser,
John Brauman, Paul Schleyer, Kendall Houk, and Michael Wasielewski. The
participants for the closing panel discussion were John Baldwin, Ned Porter,
Matthew Platz, Hans Reich, John Roberts, and Michael Wasielewski.
ACS Books was very interested in issuing a volume on this historical
symposium. Co-organizer Jeff Seeman was unable to continue as a co-editor
because of the press of other activities. Tom asked his long-time friend and
chemical historian Vera Mainz to co-edit the volume, and she graciously agreed
to do so. Greenberg, Arnett, Breslow, Streitwieser, and Schleyer agreed to submit
book chapters, but Brauman, Houk, and Wasielewski were unable to do so.
However, Norris Award winners Ken Wiberg, Keith Ingold, and Wes Borden,
who were unable to participate in the original symposium, agreed to submit
chapters on their work.

ix
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The steady progress toward completion of this volume ended abruptly with
the death of Paul Schleyer in November, 2014. In his last message to us on Nov.
19, he said he would try to finish his chapter the very next week. He died just two
days later. Paul was 84 years old, but he was still strongly involved in quantum
chemistry research and had a great deal left to give to chemistry and also to the
history of chemistry. Your editors were faced with the problem of transforming
his oral presentation, which we had on tape, into a written chapter. Also, we felt
that this volume should not only contain a chapter by Paul, but a chapter about
Paul. Eventually we were able to persuade Andy Streitwieser to provide such a
chapter. Fortunately, Andy was able to obtain Paul’s unfinished memoir, which he
had started but never completed for Jeff Seeman’s series of books by renowned
organic chemists. We thank Paul’s widow Inge for permission to use this material.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.pr001

Consequently, Paul Schleyer is represented in two chapters in this volume.

With the precedent of having a chapter about a deceased Norris Award winner
in the volume, we asked ACS Books for permission to have two more chapters
about deceased winners. This permission was given. Therefore, the book was
completed with a chapter on William Doering written by Ron Magid and Maitland
Jones and a chapter on Glen A. Russell written by Kathleen Trahanovsky and E.
Thomas Strom.
Our first words of thanks must go to the authors of our book chapters. We
are grateful for the sharing of their recollections and research with the readers of
this book. For financial support of the original symposium, we want to thank
the ACS History of Chemistry Division, ACS President Tom Barton, and the
Northeastern ACS Section. The ACS Division of Organic Chemistry was also
a sponsor of the original symposium. We also thank the HIST Program Chair
Seth Rasmussen for his assistance in arranging the symposium. We appreciate the
important work done by Jeff Seeman in obtaining the original group of speakers.
We thank the many reviewers of these chapters for their comments and corrections.
For their invaluable assistance in completing this volume, we thank Tim Marney,
Bob Hauserman, Arlene Furman, Lindsey Watson, and Tara Urban of ACS Books.
Someday someone will write a history of physical organic chemistry. This
volume, with its contributions by so many noted physical organic chemists, will
be an invaluable starting point for such a history.

x
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 References
1. Seeman, J. I. R. B. Woodward, A Great Physical Organic Chemist. J. Phys.
Org. Chem. 2014, 27, 708–721.

E. Thomas Strom
Department of Chemistry and Biochemistry
University of Texas at Arlington
Box 19065
Arlington, TX 76019-0065
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.pr001

Vera V. Mainz
Department of Chemistry
University of Illinois at Urbana-Champaign
600 S. Mathews Avenue
Urbana, IL 61802

xi
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Editors’ Biographies
E. Thomas Strom
Dr. E. Thomas (Tom) Strom is an Adjunct Professor at the University of Texas
at Arlington (UTA), where he teaches organic and polymer chemistry. He came to
UTA after retiring from Mobil in Dallas, where he worked 32 years as a research
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ot001

chemist studying oil field chemistry. He was Chair of the ACS Division of the
History of Chemistry in 2011-2012. His research interests are in the history of
chemistry and the study of anion radicals by electron spin resonance spectroscopy.
He was one of the initial ACS Fellows and is a past winner of the Dallas-Fort
Worth ACS Section’s Doherty Award. He received his B.S.Chem degree from the
University of Iowa, his M.S.Chem degree in nuclear chemistry from UC-Berkeley,
and his Ph.D. in physical organic chemistry from Iowa State University working
under mentor Glen A. Russell. Previously he has co-edited two volumes in the
ACS Symposium Series: "100+ Years of Plastics. Leo Baekeland and Beyond"
(2011) and "Pioneers of Quantum Chemistry" (2013).

Vera V. Mainz
Dr. Vera Mainz is retired Director of the NMR Lab in the School of Chemical
Sciences at the University of Illinois at Urbana-Champaign. She received a B.S.
in Chemistry and Mathematics at Kansas Newman College (1976), a Ph.D. in
Inorganic Chemistry at the University of California Berkeley (1981, with R. A.
Andersen), spent 1-1/2 years working at Rohm and Haas in Springhouse, PA,
and had a postdoctoral position at the University of Illinois at Urbana-Champaign
(1983-1985) before becoming Director of the NMR Lab. She was elected to the
position of Secretary-Treasurer of the History of Chemistry Division (HIST) of
the ACS in 1995, and has served as Secretary-Treasurer since that time. Her
interest in the HIST Division was kindled when she presented her work on the
chemical genealogy of the University of Illinois (UI) Chemistry Department at
a HIST symposium on chemical genealogies in 1994. She has continued her
work in this area, posting her information to a website at http://www.scs.uiuc.edu/
~mainzv/Web_Genealogy/, and plans to update this project when her schedule
allows. Vera’s interest in the history of chemistry led her and her husband, Gregory
Girolami, to co-curate two exhibits at the Univ. of Illinois’ Rare Book Room:
1) From Alchemy to Chemistry: Five Hundred Years of Rare and Interesting
Books, http://www.scs.uiuc.edu/~mainzv/exhibit/; 2) Crystallography – Defining
the Shape of Our Modern World, found online at URL http://www.scs.illinois.edu/
xray_exhibit/. Vera was a member of the ACS Fellows Class of 2012, which

© 2015 American Chemical Society

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 honored her contributions to the ACS (HIST and local section service) and the
many students she has helped while working in the NMR Lab.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ot001

306
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Chapter 1

James Flack Norris: His Early Contributions

to Physical Organic Chemistry
Arthur Greenberg*
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

Department of Chemistry, University of New Hampshire, Durham,

New Hampshire 03824
*E-mail: [email protected]

In honor of the fiftieth anniversary of the James Flack Norris

Award in Physical Organic Chemistry, a brief biographical
sketch of Norris is presented along with discussion of his
research accomplishments in the contexts of the influence of
his doctoral mentor Dr. Ira Remsen and the early development
of physical organic chemistry. Dr. Norris had two distinct
academic careers, researcher and teacher at Massachusetts
Institute of Technology (MIT) and administrator and teacher
at Simmons College. A third career, scientific leader and
statesman, co-existed with these dual careers. This chapter
will briefly describe Norris’s researches in molecular addition
compounds, the triphenylmethyl system, Friedel-Crafts
chemistry as well as his twenty paper series on “The reactivity
of atoms and groups in organic compounds” (from 1920 through
1940).

The Early Days of Physical Organic Chemistry and Some

Modern Perspectives
The origins of the field recognized as “physical organic chemistry” may be
dated close to the start of the twentieth century. In 1895 Paul Walden demonstrated
(1) the stereochemical inversion that now bears his name. The concept of
steric hindrance had been introduced by Victor Meyer (2) one year earlier. In
1899, Julius Stieglitz made a solid proposal for the existence of carbocations
(3). In 1900 Moses Gomberg published his ground-breaking discovery of the
triphenylmethyl free radical (4). Explicit recognition of the field dates to the

© 2015 American Chemical Society

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 period between 1920 and 1940 (5). Christopher K. Ingold published “Principles
of an Electronic Theory of Organic Reactions” in Chemical Reviews in 1934 (6)
and this was later expanded into his classic textbook, Structure and Mechanism in
Organic Chemistry, published in 1953 (7). Louis P. Hammett formally anointed
the field with his publication of the 1940 textbook Physical Organic Chemistry
(8). Consistent with this early history, the American Chemical Society (ACS)
presented the first James Flack Norris Award in Physical Organic Chemistry to
Christopher Ingold in 1965 and the second award to Louis Hammett in 1966 (see
the Appendix).
Hammett’s Preface is brief and essentially defines physical organic chemistry
as “…the investigation of the phenomena of organic chemistry by quantitative
and mathematical methods (8).” The title of Ingold’s book explicitly describes its
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

theme and, whereas Hammett’s 1940 textbook emphasizes kinetics and equilibria,
the first 196 pages of Ingold’s large tome emphasizes structural properties and
includes resonance theory and Hückel molecular orbital theory (7). The late
1950s and early 1960s witnessed the widespread and fundamental introduction of
physical organic chemistry into the undergraduate chemistry curriculum. Many
from my own generation were first explicitly introduced to physical organic
chemistry through Edwin S. Gould’s Mechanism and Structure in Organic
Chemistry (9) (note the permutation of Ingold’s title). Other excellent text books,
varying slightly in emphasis, soon followed (10–13).
The 1950s, 1960s and 1970s might well be considered to be a heyday of
“classical” physical organic chemistry with the great non-classical carbonium
ion controversy as its centerpiece. The sometimes thunderous arguments and
disagreements during this period, among some of the most distinguished chemists
of the era, led to refinements in the interpretation of substituent effects, solvent
effects, steric effects and anchimeric assistance. Ultimately, the controversy
was settled through spectroscopy experiments at temperatures approaching 0
K, coupled with increasingly reliable computational chemistry. During this
period other “shoots” began to emerge including the beginnings of mechanistic
biochemistry and nanotechnology. Indeed, even as its methodologies became
widely employed in many fields of investigation, practitioners started to worry
over whether physical organic chemistry, as a field, was still recognizable and even
“viable” (read fundable) as such. This author recalls attending an evening talk
in 1976 by a distinguished long-time physical organic chemist who pronounced
chemistry “a dead field” and confidently laid out his new strategy which was
to do only “useful” chemistry. Queried at the end of the talk about the totally
counter-intuitive results derived from the emerging field of gas-phase organic
ion chemistry (e.g. toluene a stronger acid in the gas phase than methanol), he
described this and similar such observations as “artifacts.” Still, the evolution of
chemical science and changes in funding priorities and criteria have led to (useful)
periods of self-doubt and self-examination. Such concerns provided impetus for
an International Union of Pure and Applied Chemistry (IUPAC) Symposium in
print concerning the past, present and future of physical organic chemistry (14).
Reports of the demise of physical organic chemistry remain premature. This is
readily appreciated by simply viewing the sheer scope of the table of contents of
the 1100-page, folio-sized 2006 tome by Eric Anslyn and Dennis Dougherty (15).

2
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Early in the twenty-first century we find ourselves in a funding environment
increasingly demanding practical solutions to the problems of the day such as
“sustainability”, a term so widely and indiscriminately used as to become almost
as meaningless as it is highly fundable. Yet who would be so bold (or reckless) to
assert that they can predict the future outcomes of fundamental research in physical
organic chemistry? It was in 1959, that Richard Feynman first drew attention
to the concept of nanoscience (16). A year later Edel Wasserman reported the
first synthetic catenanes (17) and Gottfried Schill published Catenanes, Rotaxanes
and Knots in 1971 (18). Who could have envisioned that a generation later such
chemical exotica would play an important role in the design of nanodevices (19)?
Similarly, Robert Bergman provided strong experimental and theoretical evidence
for the esoteric, indeed evanescent, 1,4-benzyne (20). Some nineteen years later,
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

naturally-occurring anti-cancer agents including calicheamicin were discovered.

An enzyme “pulls a trigger” in this molecule, forming a 1,4-benzyne derivative that
attacks DNA in the cancer cell (21). Thus, while the principles and methodologies
of physical organic chemistry are widely adopted by many diverse scientific fields,
physical organic chemistry itself remains vital, recognizable and will undoubtedly
produce breakthroughs hard to anticipate today.
So where does the research of James Flack Norris fit into the development of
physical organic chemistry? He was, in fact, “present at the creation”, as Tidwell
(22) references Norris’s early preparation of carbocation salts (23). Norris’s
fundamental research was published even as he inveighed against tricoordinate
carbon (24) and Gomberg’s free radical. Moreover, his quantitative studies
during the 1920s and 1930s of the dependence of rates of reaction on substituents
and solvents certainly fit the Hammett definition of physical organic chemistry.
In the remainder of this chapter we will briefly explore biographical details of
Professor Norris’s life, the influence of his research mentor Professor Ira Remsen,
his contributions to the nascent field of physical organic chemistry, and the two
different James Norris Flack awards: the chemical education award, presented by
the Northeastern Section of the American Chemical Society (NESACS), starting
in 1951, and the award in physical organic chemistry presented by the American
Chemical Society at the national level starting in 1965.

Brief Biography of James Flack Norris

James Flack Norris (Figure 1) can be said to have had two nearly-distinct
academic careers. Superimposed upon these was another career as a chemistry
leader and scientist-statesman. His academic career started and finished at
Massachusetts Institute of Technology (MIT) during which his synthetic and
quantitative studies, very much congruent with Hammett’s definition, added
substantially to a burgeoning new field. This influence was recognized explicitly
by the James Flack Norris Award in Physical Organic Chemistry celebrated in
the present symposium. Norris spent a twelve-year hiatus period forming the
chemistry program at Simmons College, with one year on sabbatical at Karlsruhe.
This experience, complemented by important textbooks he authored, combined
3
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 with his teaching accomplishments at MIT and Simmons provided impetus
for the earlier endowed award, the James Flack Norris Award for Outstanding
Achievement in Teaching of Chemistry.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

Figure 1. Dr. James Flack Norris, Professor of Chemistry at MIT. (Courtesy

Department of Chemistry, MIT)

4
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 A number of sources provide details of the life and career of Professor James
Flack Norris (25–30), the most extensive being the 1951 piece by Blake (26),
reprinted in the special commemorative issue of The NUCLEUS in 1965 (27).
While the 1974 National Academy of Sciences Biographical Memoir compiled
and reformatted material from the earlier Tenney L. Davis article (25), its author
John D. Roberts also contributed insights about the significance of Norris’s
research (28).
James Flack Norris was born in Baltimore, Maryland on January 20, 1871.
He was the fifth of nine children born to Methodist Reverend Richard Norris
(1835-1897) and Sarah Amanda (Baker) Norris (1849-1917). Young Norris
graduated from Central High School in Washington, D.C. and attended Johns
Hopkins University and received his A.B. degree in chemistry, coming under the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

influence of Professor Ira Remsen with whom he would complete his Ph.D. in
1895. His research involved study of “molecular addition compounds” including
those formed between amines and molecular halogens His interest in the analytical
chemistries of selenium and tellurium also developed during this period. Norris
was appointed Instructor at MIT in 1895 and promoted to Assistant Professor
in 1900 where he served until 1904. During 1904 Norris served as Chairman
of the Northeastern Section of the ACS. Clearly, his leadership qualities were
recognized early by his peers. On February 4, 1902 Norris married Anne Bent
Chamberlin (1876-1948). She was a student at the Museum of Fine Arts in
Boston and later became a pioneer in social work for the deaf. In 1903 Norris
was honored with a lectureship at Simmons College in Boston and in 1904 was
appointed the first Professor of Chemistry at Simmons. Simmons College was
established in 1899 with the aspiration of becoming the “MIT for women.” Norris
would spend his next twelve years at Simmons organizing its course of chemistry
instruction as well as the laboratories. From 1907 through 1910 he served as
Director of the School of Science. Sensing the need to apply physical chemistry to
his interests in organic chemistry, Norris spent part of the 1910-1911 period with
Fritz Haber in Karlsruhe. (The irony in this association would become apparent
almost a decade later following the end of World War I). Returning to Simmons,
he was appointed Chairman of the Department of General Science. Norris was
granted lectureships at Harvard during 1912 and 1913 where one of his students
was Louis P. Hammett (26). He remained at Simmons until 1915, then spent one
year as professor at Vanderbilt University before returning to MIT as professor
in 1916. Upon leaving Vanderbilt he was presented a silver cigarette case with
the inscription “Sunny Jim” (31), an apt nickname for this dapper, well-liked man
and one that would remain with him for the rest of his life.
In 1906, James and Anne Norris built a summer house on Long Lake
in Western Maine that they christened “Good Cheer.” It was in this relaxing
environment that Norris wrote the four books and subsequent editions that would
eventually seed the Norris Fund after Anne Norris’s passing. The first of these,
titled The Principles of Organic Chemistry, was published in 1912. The second,
Experimental Organic Chemistry, was published in 1915. By the time third
editions of these books appeared in 1933, each had sold a total, in all editions, of
over 70,000 copies (26).

5
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 In October, 1917 MIT granted Norris a leave of absence to work for the U.S.
Government on gas-related problems at the Bureau of Mines in Washington, D.C.
He was subsequently placed in charge of “Offence Chemical Research” at the
Bureau. In early 1918 he was commissioned a Lieutenant Colonel in the Chemical
Warfare Service, based in London. After World War I (1919), Norris was placed
in charge of investigating the German manufacture of poison gas including the
frequency of illness and death among workers at Bayer Company and Höchst (32).
It is not apparent from my brief perusal whether Norris had interviewed Haber in
1919. Haber’s tragic contribution to the war effort, supervising the employment
of 6000 steel cylinders filled with 160 tons of chlorine gas at Ypres in 1915, and
his researches and applications of poison gas, are well known (32). A patriotic
German, to the consternation of many, Haber was awarded the Nobel Prize in
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

Chemistry in 1918 for his discovery of the fixation of atmospheric nitrogen to

manufacture fertilizers that ultimately saved millions from starvation. Born Jewish
he converted to Christianity. This did not mollify the Nazis and when they came to
power in 1933, he lost his positions in Germany, fled to England and died in 1934.
In 1919 Norris was honorably discharged, returned to MIT and resumed a
research career that focused initially on the Friedel-Crafts reaction. James Mason
Crafts was on the MIT faculty from 1870 through 1874, spent the next seventeen
years in Paris working with Charles Friedel, returning in 1892 as Professor. Crafts
was President of MIT from 1897 to 1900 and continued his research for over a
decade afterward. Thus, Norris had early access to Crafts as a faculty colleague as
well as during Craft’s later years in Boston. During the 1920s, Norris’s research
moved into areas easily recognized almost a century later as physical organic
chemistry. These studies involved the effects of substituents and solvents upon
reaction rates determined quantitatively and they will be very briefly outlined
below.
In 1924, James Flack Norris was appointed Chair of the National Research
Council Section on Chemistry and Technology. He was the last ACS President
elected to serve multiple terms (1925, 1926; ten previous ACS presidents had
served multiple terms consecutively or non-consecutively). On December 29,
1925, Norris gave the opening talk at the inaugural National Organic Chemistry
Symposium (33). Titled “The Opportunities for Research in Organic Chemistry”,
Norris had originally planned a talk titled “Quantitative Measurement of Organic
Reactivity of Organic Compounds (33).” This latter topic was, in all likelihood,
the basis of his subsequent guest presentation, “Chemical Reactivity,” delivered
at Columbia University during summer, 1926 (34). During the period 1925-1928
Norris was Vice President of IUPAC. He had been elected an Honorary Member of
the Royal Institution of Great Britain in 1925 and was a member of an educational
delegation to the USSR in 1928. In 1937 Norris received the Gold Medal of
the American Institute of Chemists honoring is accomplishments as “teacher and
investigator.” He served as Chairman of the Faculty of MIT during 1937-1939.
James Flack Norris was elected to membership in various august groups including
the American Academy of Arts and Sciences (serving as Vice President during
1936), the National Academy of Sciences (elected 1934), and as a fellow of the
American Association for the Advancement of Science (Chairman, Chemistry
Section in 1930). But it is almost irresistible to mention his membership in the

6
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 rather lesser known “Ouroboros.” This club, founded in 1916 (35), evolved from
the “Chemistry Round Table,” convened by members of NESACS in 1910. The
ouroboros is a mythical alchemical legged serpent that regenerates itself even as
it consumes its own tail. The membership of Ouroboros would include Theodore
Richards, Charles James, James Conant, Arthur Noyes, Louis Fieser and Kenneth
Wiberg among many luminaries. The first female member, Anna J. Harrison,
Professor at Mount Holyoke, was admitted to Ouroboros in 1970, eight years
before she became the first female President of ACS. Ouroboros seems to have
become extinct during the first decade of the twenty-first century (35).
Over the course of a four-decade-long academic career, James Flack
Norris authored roughly seventy publications (28) and mentored twenty-one
Master degree students (including three women) and thirty-eight Ph.D. students
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

(including three women) (27). It must be noted here that as a junior faculty
member at MIT and during his period at Simmons, Norris had no graduate
assistants. The first two Norris-trained graduates received masters (S.M.) degrees
in 1920 and 1921 and his first Ph.D. student graduated in 1922 (27). One of his
masters students, Charles J. Pederson (S.M. 1927) would spend his subsequent
career at Du Pont and share the 1987 Nobel Prize in Chemistry with Donald
J. Cram and Jean-Marie Lehn. The title page of the 1937 Ph.D. thesis of 1965
Nobel Laureate Robert Burns Woodward is signed by Norris as Chairman of
the Department Committee of Graduate Students (36). Woodward famously
completed his B.S. (1936) and Ph.D. (1937) degrees at MIT in a total of four
years. Norris is revealingly quoted in the Boston Globe in 1937:

…we saw we had a person who possessed a very unusual mind and we
wanted it to function at its best. If the red tape necessary for less brilliant
minds had to be cut, we let it go. We did for Woodward what we have
done for no other person like him in our department. We think he will
make a name for himself in the scientific world (37).

In his brief dissertation acknowledgement Woodward thanked “…Professors

James F. Norris and Avery A. Morton, who have taken a particular interest in this
work…” (36). Norris was also Morton’s research mentor at MIT (Ph.D. 1924)
prior to the latter’s appointment to the MIT faculty.
During summer, 1940 successful cataract surgery in Norris’s right eye (June)
was followed by phlebitis in July and hospitalization for blood transfusions some
two weeks later. He died on August 4, 1940 survived by his wife Anne. The
devoted couple had no children. Anne Bent (Chamberlin) Norris (born May 29,
1879) (38) passed away on May 29, 1948. She had been elected in 1933 the first
female President of the American Society for the Hard of Hearing and continued
to serve this and other related organizations for many years to come. In her will
Mrs. Norris provided the Northeast Section of the ACS with a gift of $10,000,
also sharing the remainder of the estate equally with MIT. The will stated:

It is my wish that the Directors of said Society shall use the money in any
way they may see fit to perpetuate the memory of my said husband James
F. Norris.

7
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 This fund would soon form the basis of the Norris award in teaching and over
a decade later the Norris award in physical organic chemistry. These will be
discussed later.

The Influence of Ira Remsen upon the Career of James Flack

Norris
Although it might at first appear to be a routine academic exercise, the
influence of Norris’s Ph.D. advisor Ira Remsen (39) on his future career as
scientist, scientific leader and statesman was quite profound and worth examining
briefly. He left his mark on the education of American Ph.D. chemists from the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

late nineteenth century into modern times. Remsen saw no sharp dividing line
between inorganic and organic chemistries and made noteworthy contributions in
both fields (he lectured and authored important textbooks in both areas). Remsen
explored structure-reactivity relationships in organic chemistry during the final
quarter of the nineteenth century.
Ira Remsen (1846-1927) was born in New York City, received his
undergraduate education at the Free Academy (now the City College of New
York) and, obeying his father’s wishes, obtained his M.D. degree in 1867 at
the College of Physicians and Surgeons of Columbia University. However,
Remsen developed an interest in chemistry and, exercising his free will, travelled
to Munich hoping to work with Justus Liebig. Liebig was no longer taking
students so young Remsen worked with Privatdocent Jacob Volhard, learning
techniques in analytical chemistry. During this period, Friedrich Wöhler was an
occasional visitor to the Munich laboratory and in fall 1868 he inspired Remsen
to move to Göttingen. There Remsen pursued organic chemistry research under
the direction of Rudolf Fittig, receiving his Ph.D. in 1870. The twenty-four
year old Remsen followed Fittig to Tübingen where “he drank in the spirit of
the German laboratories” (39). Returning to the United States in 1872 with a
faculty appointment at the liberal arts institution Williams College, Remsen was
severely disappointed by the lack of interest and support for teaching experimental
chemistry (39). He did complete his masterful textbook Theoretical Chemistry,
an extraordinarily useful work that made the principles and breakthroughs of the
period widely accessible.
Despite the serious limitations at Williams, Remsen conducted experimental
research including studies of the oxidation of carbon monoxide and reactions of
phosphorus compounds. The concept of valence was still quite new and many
chemists held the view that each element exhibits a single valence. Thus, reaction
of PCl3 with chlorine yields what was commonly regarded as the “molecular
addition compound” PCl3.Cl2. Remsen, aware of Wurtz’s recent determination
of vapor density consistent with the formula PCl5, concluded that the reaction
of PCl3 and chlorine did in fact produce pentavalent phosphorus. Similarly, his
studies of the reaction of ozone with PCl3 led Remsen to interpret the product as
containing pentavalent phosphorus: O=PCl3 (39).
Ira Remsen was one of a galaxy of stars recruited to Johns Hopkins University
in 1876 (39). There he established a Ph.D. program based upon the German
8
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 model and revolutionized graduate education in chemistry in the United States.
At Hopkins much of his research focused on organic chemistry. An early interest
in structural organic chemistry led to “Remsen’s Law”- e.g. a methyl group ortho
to a sulfamido group (SO2NH2) is protected from certain oxidation reactions.
Remsen’s researches into what one might term “primordial physical organic
chemistry” is exemplified by an 1884 paper in which he demonstrated that
isopropyl bromide is less stable than n-propyl bromide in zinc and an acid but
more stable in alcoholic sodium hydroxide (39). Remsen’s student Norris would
eventually make such structure-activity comparisons quantitative (“physical
organic”) rather than simply qualitative. In 1879 Remsen and Russian chemist
Constantin Fahlberg, working in Remsen’s laboratory, published the discovery
of saccharin (39). Fahlberg would file patents in 1884 and 1886 claiming sole
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

credit for the discovery and starting a business manufacturing and selling the first
commercial synthetic sweetener.
Ira Remsen was a scientific and academic leader and scientific statesman,
a path traversed by James Flack Norris decades later. Shortly after arriving at
Johns Hopkins, Remsen inaugurated The American Chemical Journal. After
publication of fifty volumes, Remsen agreed to incorporate it with The Journal
of the American Chemical Society (JACS). Remsen was elected President of the
American Chemical Society for 1902. He served as Acting President of Johns
Hopkins (1889-90) and President from 1901 through 1913. In 1909 Ira Remsen
was appointed by President Theodore Roosevelt to Chair a Board studying
administration of the pure food law (39). Based upon this very brief biographical
outline, it is reasonable to conclude that Ira Remsen exerted major influence upon
James Flack Norris in both scientific and public venues.

Early Scientific Research on Selenium, Tellurium and

Molecular Addition Compounds
The research of James Flack Norris under the mentorship of Ira Remsen at
Johns Hopkins University included the chemistry, notably analytical chemistry,
of selenium and tellurium. The earliest paper with Norris’s name on it was
co-authored with Remsen (40). Starting with a salt assigned the formula
SeBr4.SeOBr2.2(CH3)3N.HBr, reaction with bromine formed (CH3)3N.Br2. The
same product was obtained through reaction of trimethylamine with bromine.
Similar chemistry was observed between trimethylamine and iodine. Such
“molecular addition compounds” were poorly understood at the time. The
physical methods for testing their nature were quite limited- formation of unique
crystals upon reactions of molecular substances, new melting points, and percent
composition of constituent elements were among relatively few experimental
criteria. Theories about the bonding in these complexes included notions of
“residual affinities” of atoms in the molecules.
Henry Fay would co-author Norris’s next paper, which involved
determination of selenious and selenic acids (41). These two young chemists
first met as assistants at MIT. Three additional papers, also co-authored with Fay,
9
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 would follow between 1898 and 1900 (28). In 1902, Fay was best man at Norris’s
wedding.
Although first investigated at MIT, a decade after Norris finished his doctoral
studies at Johns Hopkins, it is worthwhile mentioning Norris’s continued interest
in selenium and tellurium. Among significant problems in the late nineteenth
and very early twentieth century were the “anomalies” in the ordering in the
periodic table of some elements according to atomic weight. One example
involves tellurium, heavier than iodine, rather than lighter than iodine as would
be expected by their placement:
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

One simple explanation could be very minor contamination of tellurium samples

with another, heavier chemical element. However, in 1906 Norris confirmed the
purity and atomic weight of tellurium (42), so the apparent anomaly remained.
Today, we understand that this “reversal” is due to the presence of heavy isotopes
of tellurium (128Te; 130Te) in significant (each >30%) abundance. However, it was
only in 1907 that Bertram Boltwood characterized “ionium”, soon to be recognized
as an isotope of thorium (the term coined by Frederick Soddy in 1913). Also in
1907, Joseph John (“J.J.”) Thomson discovered canal rays associated with neon
corresponding to atomic masses 20 and 22. These measurements were confirmed
by Francis William Aston in 1919 using his early mass spectrograph.

Debating Tricoordinate Carbon with Moses Gomberg

Although the term “free radical” is in common (chemical) usage, the word
“radical” is understood by most non-chemists as referring to “extreme” conditions
(as in radical surgery) or in an “extreme” political context. However “radical”
can be defined as “of or from the root” or “foundation or source of something.”
In the late eighteenth century, Louis Bernard Guyton de Morveau introduced the
term “radical”, later adopted by Antoine Laurent Lavoisier, to describe RX + Y →
RY + X. Here radical R is exchanged between X and Y. How R moved from X to
Y was not imaginable. However, nineteenth-century chemists, especially organic
chemists, did pose such questions and tried to obtain or at least infer the presence of
the “free radical” (e.g. R). By the 1860s, the tetravalent nature of carbon was well
accepted. During the latter part of the nineteenth century a number of excellent
chemists claimed to have isolated free radicals but were quickly proven incorrect.
In 1900 Moses Gomberg reported his attempt to synthesize hexaphenylethane
through reaction of triphenylmethyl chloride with zinc metal (Wurtz reaction) in a
non-polar solvent free of oxygen (4). Zinc chloride precipitated leaving a yellow
solution. This solution reacted instantly with added iodine to form triphenylmethyl
iodide. The yellow solution also reacted instantly, when exposed to air, to form a
peroxide:
10
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The observation of the stable triphenylmethyl free radical was indeed a radical
claim but was well accepted within a few years. At the end of his paper Gomberg
famously placed this injunction:

This work will be continued and I wish to reserve the field for myself.

At this time, Norris was investigating the Friedel-Crafts reaction and the possibility
of employing carbon tetrachloride and benzene to synthesize tetraphenylmethane.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

The reaction yielded triphenylmethyl chloride and therein did Norris collide with
Gomberg (23) who had just reported the compound even as the Norris manuscript
was ready for submission. Responding to Gomberg’s expansive territorial claim,
Norris countered (23):

He wishes to reserve this field for further work. As we have been working
with triphenylchloromethane in this laboratory for about a year, with a
view of preparing tetraphenylmethane from it, it seems advisable to give
a preliminary account of the work that has been accomplished, and to
describe briefly what is now in progress.

When Norris combined triphenylmethyl chloride, benzene and aluminum

trichloride in carbon disulfide solution, he failed to obtain tetraphenylmethane
but found instead a solid that was insoluble in the non-polar solvents benzene,
carbon disulfide, carbon tetrachloride and petroleum ether. The solid was
soluble in nitrobenzene but decomposed in alcohol, ether, acetone, chloroform
and acetic acid. The solid was reported as an addition compound of formula:
Al2Cl6.2(C6H5)3C.Cl (43) [later modified to (C6H5)3CCl.AlCl3)]. Norris realized
that under typical Friedel-Crafts reaction conditions, after addition of the third
benzene ring, a significant solubility change would occur resulting in precipitation
and no further reaction. The Friedel-Crafts conditions employed by Norris were,
of course, very different from the reaction conditions employed by Gomberg.
Indeed, as intimated by Tidwell (22), Norris was in fact exploring the chemistry
of the stable triphenylmethyl carbocation. Norris then reported reactions of
triphenylmethyl chloride and benzene in the presence of sodium metal (23).
This was, of course, uncomfortably close to Gomberg territory. In studying
the chemistry of triphenylmethyl chloride in concentrated sulfuric acid, HCl
was given off and the final product was triphenylcarbinol (24). Ironically, this
paper (24) was titled by Norris “On the Non-Existence of Trivalent Carbon.”
He concluded that the highly-unsaturated hydrocarbon prepared by Gomberg
was best represented as (C6H5)3C : C6H4 and implied that the structure included
the central carbon attached to two carbons on one of the benzene rings (24).
Gomberg, however, sharply criticized Norris’s studies on Wurtz-type chemistry
and made a convincing case that experimental divergences from his own results
were largely the result of incomplete purification by Norris of solvents (43).
11
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Gomberg was totally unconvinced by the hydrocarbon formula advanced by
Norris. Indeed, he was irritated by Norris’s very presence in the field (43):

It seems, therefore, strange that Norris should at this stage give up

his original problem, the synthesis of tetraphenylmethane, and take up
the study of the action of metals upon triphenylchlormethane. I have
published only a preliminary paper upon this subject, asking that the
field might be reserved for myself.

Gomberg and Norris did agree on the experimental conversion of triphenylmethyl

chloride to the alcohol in concentrated sulfuric acid (43). While Norris lost
the triphenylmethyl radical debate to Gomberg, it is clear that he was an early
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

contributor to stable carbocation chemistry. During the first decade of the

twentieth century, conductivity studies in high dielectric constant solvents would
demonstrate the reality of stable carbocation salts including triphenylmethyl
carbocation.

Contributions to Quantitative Organic Chemistry (1920s and

1930s)

In 1904 James Flack Norris left MIT to establish the chemistry program
at Simmons College. He would return to MIT in 1916, with war assignments
occupying his attention from 1917 through 1919. As a consequence, Norris
published one research paper (42) in 1906 and one each in 1907, 1910 and 1916.
Not until 1920 did Norris research papers reappear (three in the Journal of the
American Chemical Society) (27). One series of papers (1907, 1916 and 1920)
dealt with reactions of alcohols in aqueous solutions of HCl and HBr. Primary
alcohols were found to be very unreactive whereas benzyl alcohol and allyl
alcohol were highly reactive. The theme of exploring organic reactivities through
kinetics became the major thrust of his research program at MIT starting in 1916
(34). This is one area in which Norris departed markedly from his doctoral mentor
since Remsen was judged to be somewhat indifferent to physical chemistry
(44). In his 1926 presentation at Columbia University (34), Norris nicely laid
out his rationale for quantitative studies of organic reactions. In describing the
understanding at that time of the chemical reactions of inorganic substances,
Norris declared (34):

The modern conception of the changes involved in inorganic chemical

reactions are based on the quantitative study of the energy changes of
these reactions.

and further (34):

12
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The facts summed up in Faraday’s law and in the electromotive series of
the elements made possible the broadest generalizations in regard to the
chemical behavior of the elements and their compounds.

In other words, the reactions of inorganic substances were, in Norris’s view,

determined by thermodynamics. While the situation in organic chemistry was
deemed far more complex (34):

Organic chemistry lacks such broad generalizations. The radicals in

these compounds correspond to the elements in inorganic chemistry. Can
these radicals be studied in some way to give results of a general nature
that are applicable in this field of interpretations of reactions, as is the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

electromotive series of the elements in inorganic chemistry?

When Norris resumed full-time faculty duties at MIT he entered a new

landscape in which chemical theories of reactivity of organic molecules were
being tested by experiment (5). Notably Arthur Lapworth had introduced a theory
of “alternative polarities” as early as 1905. Robert Robinson became friendly
with Lapworth in Manchester in 1909 and interested in the applications of
Lapworth’s theory. Combining Lapworth’s views and Johannes Thiele’s concept
of “residual valencies”, Robinson began developing a coherent theory of reactions
of unsaturated molecules including substituted benzenes. In 1925, Robinson
modified Thiele’s 1899 representation of benzene to provide the structure we
now commonly employ. Whereas residual valencies at the ends of 1,3-butadiene
explained the reactivity of this molecule (as well as its 1,4-addition chemistry), in
benzene, the ends meet in a cycle, neutralizing residual valencies and accounting
for its stability. Robinson also introduced the “curvy arrow” to illustrate the
movement of chemical bonds. He was an early supporter for applying electrons
(Lewis pairs) toward understanding mechanisms, a position advocated by Thomas
Martin Lowry (5).
An opposing school of thought grew in the early twenties around a theory of
“affinity demand” proposed by Bernard Flürscheim, related to work by Thiele and
Arthur Michael (5). Modifications, emphasizing electrical charges on atoms, were
introduced by Daniel Vorländer. This view deemphasized the need for electron
pairs. Christopher Ingold was, in fact, more in accord with this viewpoint.
The early differences in chemical theory between Robinson and Ingold
intensified in 1923 and 1924 and reached a head in 1926 when Robinson came
close to accusing Ingold of plagiarism. However, from 1926 onward, Ingold
proceeded to define the developing field of mechanistic organic chemistry,
introducing, for example, the terms “nucleophile” and “electrophile.” The
chemistry historian William H. Brock terms Ingold “…the great systematizer of
twentieth-century organic chemistry who firmly established the importance of
physical chemistry in understanding the subject” (45). Robinson’s efforts would
henceforth be increasingly directed to the studies of structure and synthesis of
natural products that would eventually earn for him the 1947 Nobel Prize in
Chemistry. Still, he made one large last effort in publishing in 1932 a pamphlet
detailing his interpretation of organic reaction theory (46). However, this

13
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 work was quickly overshadowed by Ingold’s extensive 1934 review article
(6). In 1930 one of the greatest partnerships in twentieth century chemistry
began when Edward David Hughes was appointed by Ingold (initially) as a
post-doctoral fellow at University College in London. Hughes was a master of
experimental kinetics and, with Ingold, soon combined kinetics, stereochemistry
and mechanism and developed the nomenclature of nucleophilic substitution. The
Robinson-Ingold rivalry would continue for decades. As Brock notes (47): “For
although the documentary evidence is lacking, Ingold and Hughes had seriously
believed in the 1930s that Robinson, as Britain’s most senior and respected
chemist, lay behind the opposition that their views provoked.”
A 1920 paper, co-authored with Dorothy M. Tibbetts (48), was built upon
two earlier research areas: molecular addition compounds and the triphenylmethyl
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

system. In this paper, the authors included a succinct statement characterizing the
early philosophy of physical organic chemistry:

…especially in the field of organic chemistry…the chemist has it in his

power to change by the use of the proper substituents the nature of the
molecule and thus develop or reduce its ability to form addition products
with other substances.

Triphenylmethyl chloride was said by Norris to have an “active chlorine”

which is readily hydrolyzed and its high “residual affinity” or “latent affinity”
(i.e. high latent energy) is also manifested in its ability to form molecular
addition compounds such as (C6H5)3CCl.AlCl3. In contrast the “firmly-bound”
chlorine in methyl chloride does not form molecular addition compounds.
Diphenylmethyl chloride does not form analogous addition compounds although
p-phenyldiphenylmethyl chloride does form a molecular addition compound (48).
No mention was made in this paper of carbocation salts.
In 1925, the first of a series of twenty papers titled “The Reactivity of Atoms
and Groups in Organic Compounds” was co-authored with Avery A. Ashdown
and appeared in the Journal of the American Chemical Society (49). In it Norris
explored the relationship between reactivity and structure. In reactions between
p-nitrobenzoyl chloride and eleven alcohols, second-order rate constants varied by
a factor of almost seventy [relative values for RO-H: R = CH3, 100; R = CH3CH2,
45.7; R = (CH3)2CH, 5.4; R = (CH3)3C, 1.46].
In his presentation at Columbia University in summer, 1926 (34), Norris
summarized his on-going research on the reactivities of alcohols, alkyl chlorides
and acyl chlorides. In this presentation Norris emphasized the critical importance
of the quantitative measurement of chemical reactivities. In describing the
dependence of rates of reaction of derivatives of C6H5COCl with ethanol, he
reported data for o-, m- and p-substituted compounds. Table 1 includes a small
selection of data from this paper (acid chloride reactions were run at 0 °C and
alkyl chloride reactions at 25 °C).

14
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Table 1. Rates (relative to unsubstituted, X = H) of ethanolysis for a series of
substituted benzoyl chlorides and diphenylmethyl chlorides selected from
Table 1 in reference (34).
Substituent (X) C6H5COCl derivatives (C6H5)2CHCl derivatives
H 1 (assigned) 1 (assigned)
m-Cl 3.17 0.0451
p-Cl 1.74 0.421
p-CH3 0.567 18.76
p-CH3O 0.79 3200
p-NO2 17.7 -------
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

In describing the inverse influences of these substituents on the rates of these

two sets of reactions, Norris did not postulate an underlying explanation. At the
time he considered the differentiation of acyl halides from alkyl halides, based
upon the typically far greater reactivities of the former group, to be “an entirely
arbitrary one” since suitable substitution could produce alkyl halides more reactive
than acyl halides. Although different mechanisms were undoubtedly suspected,
there was no discussion of mechanism nor was there specific discussion of the
enormous rate enhancement noted for the p-methoxy substituent in Table 1. It is
likely that thirty-two year old Louis P. Hammett, appointed to the Columbia faculty
three years earlier, was present at this lecture and participated in discussions with
Norris. Two years earlier, J.N. Brønsted had published the first example of a linear
free energy relationship (50):

This refers to the dependency of reaction rate (k) on ionization constant (Ka) in
general acid-catalyzed reactions. Hammett would publish his famous linear free-
energy relationship some thirteen years later (51):

This relationship refers to the dependence of equilibrium (K) or rate (k) constants
on the electronic nature of a substituent on a benzene ring (σ) and the sensitivity of
a reaction type to electronic effects (ρ). This work would eventually systematize
the types of kinetic findings published by Norris.

15
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The twentieth and final paper in the series “Reactivity of Atoms and Groups
in Organic Compounds was co-authored with Arthur E. Bearse, and published
in 1940 (52). This paper was received in January, 1940. Hammett’s famous
textbook was published later that year. Although Norris’s paper considers both the
acidities of substituted benzoic acids and their rates of conversion to the chlorides
with SOCl2, there is no reference to Hammett’s earlier work. Three other papers
authored by Norris, all focusing on Friedel-Crafts chemistry, were submitted and
published in JACS early in 1940. Later that summer Norris died.
In his biographical sketch of James Flack Norris (28), John D. Roberts
provided the follow summary of Norris’s scientific career:

Norris was rather less successful in his work on reactivity than was
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

his younger counterpart at Harvard, James B. Conant, who displayed

an almost unerring instinct for choosing reactions for study of greater
simplicity and involving wider ranges of reactivity, along with an
excellent feel for the basic physical chemistry involved. Nonetheless,
Norris had the prescience to be at the forefront of the still-developing
area of making comparisons of organic reactivity under controlled
conditions, and some of the reactions he was first to study are among the
most important in preparative chemistry.

The James Flack Norris Awards

Anne Chamberlin Norris died in 1948, and her will provided a sizeable
endowment, to be administered by the Northeastern Section of the American
Chemical Society, to honor the memory of her late husband (53). A committee was
established to decide how best to use the bequest. The decision was announced
in January, 1950 to establish the James Flack Norris Award for Outstanding
Achievement in Teaching of Chemistry. The first awardee (1951) was George
Shannon Forbe, Professor Emeritus, Harvard University. Following retirement,
Dr. Forbe was a lecturer at Northeastern University in Boston. Initially, awards
were presented in alternate years (1951, 1953 and 1955), then one in 1956, spring
and fall awards in 1957 and 1959 and a single annual award starting in 1960.
The initiative that led to the James Flack Norris Award in Physical Organic
Chemistry came from Henry A. Hill, who was Chairman of NESACS in 1963
(54). Hill, a graduate of Johnson C. Smith University in Charlotte, North Carolina,
would later become the first African American President of ACS (1977). He
obtained his Ph.D. at MIT in 1942 and had overlapped briefly with Professor
Norris. Hill learned that the Norris Fund had “…grown both in book value and
income, and is now in excess of the fondest expectations of ten years ago” (54).
An ad hoc committee was formed that included Avery A. Ashdown, a Norris
Ph.D. graduate, and four other MIT alumni, met in February 1963. The decision
was made to fund a physical organic chemistry award that would be administered
16
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 by the American Chemical Society. The committee also authorized funding to
assist in the furnishing of a James Flack Norris room in the new MIT chemistry
building (54). The national ACS accepted responsibility in June of that year and
the first award was presented to Christopher K. Ingold in 1965 (see the Appendix).
A Symposium on Physical Organic Chemistry, to celebrate the first award, was
held on September 14, 1965 at the National ACS Meeting in Atlantic City,
New Jersey. Joseph F. Bunnett (Brown University) chaired the meeting. Other
speakers in addition to the honoree, Christopher K. Ingold (University College,
London), were Robert B. Woodward (Harvard University), Andrew Streitwieser,
Jr. (University of California at Berkeley), C. Gardner Swain (Massachusetts
Institute of Technology), Charles A. Bunton (University of California at Santa
Barbara), Maurice M. Kreevoy (University of Minnesota), and Frederic P. Olsen
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

(University of California at Riverside). Happily and most fittingly, Professor

Emeritus Streitwieser made a presentation at the March 17, 2014 symposium
celebrating the fiftieth anniversary of the award.
In 1972 ACS granted $10,000 to Simmons College to furnish a James
Flack Norris room. It served as collegial interaction space for faculty, staff and
students until 2000 when the building was totally renovated (55). A photograph
of Dr. Norris hangs proudly in the Department of Chemistry and Physics main
office. Indeed, a chemistry major was a member of the first graduating class of
Simmons College (55). It is also worthwhile to mention the James Flack Norris
and Theodore William Richards Undergraduate Summer Scholarships awarded
by NESACS. Richards, a professor at Harvard, was the first American to win the
Nobel Prize in Chemistry (1914).
There are some interesting and noteworthy points in revisiting the James Flack
Award in Physical Organic Chemistry. Two awardees also received the Norris
Award in Teaching: Louis P. Hammett (1960) and Paul D. Bartlett (1978). There
is a single case in which parent (Christopher K. Ingold in 1965) and offspring
(Keith U. Ingold in 1993) each received this award. None of the awardees were
or subsequently became Nobel laureates. This explicitly raises a question my
own Ph.D. advisor Pierre Laszlo rhetorically posed to me in the 1960s: Why had
no Nobel Prize in Chemistry been yet awarded to a physical organic chemist?
Far be it from me to plunge into the treacherous waters of controversy in which
this question has often bobbed through the years. It is likewise interesting that
Nobel laureates who have made important contributions to physical organic
chemistry, such as Donald J. Cram (1987) and George A. Olah (1994), were not
Norris awardees. Olah, in particular, discovered conditions for making transient
carbocation intermediates relatively long-lived and observable by NMR. And
for that matter, what about Robert Burns Woodward? In a paper titled “R.B.
Woodward, A Great Physical Organic Chemist” (56), Jeffrey Seeman reminds
readers of the 1965 Nobelist’s impact well beyond organic synthesis. Here
one also sees some parallels (and differences) between Woodward and Robert
Robinson, who similarly received his Nobel Prize for work in synthetic organic
chemistry but certainly left an early mark upon physical organic chemistry. In his
paper (56), Seeman also notes that in April, 1957 B.S. Hutchinson, the Chairman
of NESACS, informed Woodward that “…the committee has unanimously chosen
you as the nominee” (i.e. for the James Flack Norris Award for Outstanding

17
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Achievement in the Teaching of Chemistry). Seeman could locate no record of
Woodward’s response. Thus, for reasons not readily discernable, R.B. Woodward,
received neither of the two Norris awards despite being a very strong case for
the physical organic award and actually having been chosen unanimously for the
teaching award.
Today physical organic chemistry remains recognizable as such even as it
forms the foundation for research in very diverse fields. These contributions
continue to be recognized by the James Flack Norris Award in Physical Organic
Chemistry in areas diverse enough that predictions of the future (14) will almost
certainly miss surprises that are unimaginable at the present time.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

Dedication

In memory of Dr. Paul von Ragué Schleyer (February 27, 1930 – November
21, 2014).

Acknowledgments
The author is grateful to Thomas Strom who extended the invitation to make
a presentation at the symposium in Dallas on March 17, 2014. For initiating my
interest in physical organic chemistry, I wish to acknowledge my Ph.D. advisor
Pierre Laszlo and three professors, each a physical organic Norris awardee, from
whom I took classes in the 1960s: Kurt Mislow, Paul von Ragué Schleyer and
Peter Stang. I was also very fortunate during this formative period to have been
stimulated in seminars by Kenneth Wiberg, Michael Dewar and Ronald Breslow,
all physical organic Norris awardees. I am grateful to Joel Liebman, Richard
Johnson, Glen Miller and Gary Weisman who have been long time friends,
colleagues, teachers and sounding boards. I also wish to express my gratitude
to Jeffrey Seeman who provided me with ideas, critiques and early access to
interesting archival materials. Special thanks to Susan J. Greenberg who helped
me to organize the March 17, 2014 presentation.

Appendix
Winners of the James Flack Norris Award in Physical Organic Chemistry
[For the list of winners of the James Flack Norris Award for College Chemistry
Teaching visit the Northeast Section of the American Chemical Society (NESACS)
site: www.nesacs.org/awards_norris.html]

18
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

References
1. Walden, P. Über die gegenseitige umwandlung optischer antipoden. Ber.
Dtsch. Chem. Ges. 1896, 29, 133–138.
2. Meyer, V. Über ein seltsames gesetz bei der esterbildung aromatischer säuren.
Ber. Dtsch. Chem. Ges. 1894, 27, 510–512.
3. Stieglitz, J. On the constitution of the salts of imido ethers and other
carbamide derivatives. Am. Chem. J. 1899, 21, 101–111.
4. Gomberg, M. An instance of trivalent carbon: triphenylmethyl. J. Am. Chem.
Soc. 1900, 22, 757–771.
5. Brock, W. H. The Norton History of Chemistry; W.W. Norton & Co.: New
York, NY, 1993; pp 506−569.
6. Ingold, C. K. Principles of an electronic theory of organic reactions. Chem.
Rev. 1934, 34, 225–274.
7. Ingold, C. K. Structure and Mechanism in Organic Chemistry; Cornell
University Press: Ithaca, NY, 1953.
8. Hammett, L. P. Physical Organic Chemistry; McGraw-Hill Book Co., Ltd.:
New York, NY, 1940; p v.
9. Gould, E. S. Mechanism and Structure in Organic Chemistry; Holt, Rinehart
and Winston, Inc.: New York, NY, 1959.
10. Hine, J. Physical Organic Chemistry; McGraw-Hill Book Co., Inc.: New
York, NY, 1962.
19
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 11. Leffler, J. E.; Grunwald, E. Rates and Equilibria of Organic Reactions; John
Wiley and Sons, Inc.: New York, NY, 1963.
12. Wiberg, K. B. Physical Organic Chemistry; John Wiley & Sons: New York,
NY, 1964.
13. Kosower, E. M. An Introduction to Physical Organic Chemistry; John Wiley
& Sons: New York, NY, 1968.
14. Tidwell, T. T., Rappoport, Z., Perrin, C. L., Eds. Physical Organic Chemistry
for the 21st Century: A Symposium in Print. Pure Appl. Chem. 1997, 69 (2).
15. Anslyn, E. V.; Dougherty, D. A. Modern Physical Organic Chemistry;
University Science Books: Sausolito, CA, 2006.
16. Feynman, R. P. There’s Plenty of Room at the Bottom. Presented at American
Physical Society Meeting, Pasadena, CA, December 29, 1959.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

17. Wasserman, E. The preparation of interlocking rings: a catenane. J. Am.

Chem. Soc. 1960, 82, 4433–4434.
18. Schill, G. Catenanes, Rotaxanes and Knots; Academic Press: New York, NY,
1971.
19. Stoddart, J. F. Putting mechanically interlocked molecules (MIMs) to work
in tomorrow’s world. Angew. Chem., Int. Ed. 2014, 53, 11102–11104.
20. Jones, R. R.; Bergman, R. G. p-Benzyne. Generation as an intermediate in a
thermal isomerization reaction and trapping evidence for the 1,4-benzenediyl
structure. J. Am. Chem. Soc. 1972, 94, 660–661.
21. Nicolaou, K. C.; Dai, W.-M. Chemistry and biology of the enediyne
anticancer antibiotics. Angew. Chem., Int. Ed. 1991, 30, 1387–1416.
22. Tidwell, T. T. The first century of physical organic chemistry: a prologue.
Pure Appl. Chem. 1997, 69, 211–213.
23. Norris, J. F.; Sanders, W. W. On triphenylchlormethane. Am. Chem. J. 1901,
25, 54–62.
24. Norris, J. F. On the nonexistence of trivalent carbon. Am. Chem. J. 1901, 25,
117–122.
25. Davis, T. L. James Flack Norris. Ind. Eng. Chem. 1936, 14, 325–326.
26. Blake, J. T. The first James Flack Norris Award for Excellence in the Teaching
of Chemistry (reprinted from the NUCLEUS, June, 1951). The NUCLEUS
1965, 42, 307–309.
27. Special Issue Commemorating the First Presentation of the James Norris
Flack Award in Physical Organic Chemistry. The NUCLEUS, Newsletter of
the Northeastern Section of the American Chemical Society; August, 1965;
Vol. 42, No. 10.
28. Roberts, J. D. James Flack Norris 1871-1940. Biographical Memoir;
National Academy of Sciences: Washington, DC, 1974; pp 413−426.
29. Onofrio, J. James Flack Norris. Maryland Biographical Dictionary;
Somerset Publishers, Inc.: 1999; pp 401−402.
30. Simon, M. S. The James Flack Norris Award for Outstanding Achievement
in the Teaching of Chemistry. The NUCLEUS 2002, 81 (3), 4.
31. Wikipedia (accessed October 6, 2014) seems as good a source as any for
researching “Sunny Jim.” He is the dapper, jaunty fellow depicted on the front
of Force Toasted Whole Wheat Flakes. Force cereal was originally produced
in the U.S. in 1902, discontinued here in 1983 but is apparently still available

20
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 in the United Kingdom. Its bit of advertising doggerel (just in case you were
curious):
Jim Dumps was a most unfriendly man,
Who lived his life on the hermit plan;
In his gloomy way he’d gone through life,
And made the most of woe and strife,
Till Force one day was served to him.
Since then they’ve called him “Sunny Jim.”
32. Fitzgerald, G. J. Chemical warfare and medical response during World War
I. Am. J. Public Health 2008, 98, 611–625.
33. Fenlon, E. E.; Myers, B. J. Profiles in chemistry: a historical perspective on
the National Organic Symposium. J. Org. Chem. 2013, 78, 5817–5831.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

34. Norris, J. F. Chemical Reactivity; Columbia University Press: New York,

NY, 1927.
35. Jones, P. R.; Atkinson, E. R. Ouroboros- A New England Chemists Club.
Presented at American Chemical Society Meeting, Boston, MA, August 19,
2002.
36. The author thanks Jeffrey Seeman for disclosing and providing this
information.
37. Moore, P. J. T.; Bowden, M. E. In Robert Burns Woodward- Architect
and Artist in the World of Molecules; Benfey, O. T., Morris, P. J. T., Eds.;
Chemical Heritage Foundation: Philadelphia, PA, 2001; pp 3−10.
38. Moore, M. C. Anne Chamberlin Norris. The NUCLEUS 1965, 42, 315–316.
39. Noyes, W. A.; Norris, J. F. Ira Remsen 1846-1927. Biographical Memoir;
National Academy of Sciences: Washington, DC, 1931; Vol. 14, pp
205−257.
40. Remsen, I.; Norris, J. F. The action of the halogens on the methylamines. Am.
Chem. J. 1896, 18, 90–95.
41. Norris, J. F.; Fay, H. Iodometric determination of selenious and selenic acids.
Am. Chem. J. 1896, 18, 703–706.
42. Norris, J. F. On the elementary nature and atomic weight of tellurium. J. Am.
Chem. Soc. 1906, 28, 1675–1684.
43. Gomberg, M. On trivalent carbon. Am. Chem. J. 1901, 25, 317–335.
44. Servos, J. W. Physical Chemistry from Ostwald to Pauling: The Making of
a Science in America; Princeton University Press: Princeton, NJ, 1990; pp
93−94.
45. Brock, W. H. The Norton History of Chemistry; W.W. Norton & Co.: New
York, NY, 1993; p 507.
46. Robinson, R. Outline of an Electrochemical (Electronic) Theory of the
Course of Organic Reactions; Institute of Chemistry: London, U.K., 1932.
47. Brock, W. H. The Norton History of Chemistry; W.W. Norton & Co.: New
York, NY, 1993; p 531.
48. Norris, J. F.; Tibbetts, D. M. Organic molecular compounds. J. Am. Chem.
Soc. 1920, 42, 2085–2092.
49. Norris, J. F.; Ashdown, A. A. The reactivity of atoms and groups in organic
compounds. I. the relative reactivities of the hydroxyl-hydrogen atoms in
certain alcohols. J. Am. Chem. Soc. 1925, 47, 837–846.

21
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 50. Brønsted, J. N.; Pedersen, K. J. Die katalytische zerzetzung des nitramids
und ihre physilalische-chemische bedeutung. Z. Phys. Chem. 1924, 108,
185–235.
51. Hammett, L. P. The effect of structure upon the reactions of organic
compounds. Benzene derivatives. J. Am. Chem. Soc. 1937, 59, 96–103.
52. Norris, J. F.; Bearse, A. E. The reactivity of atoms and groups in organic
compounds. XX. The effect of substituents on the relative reactivities of the
hydroxyl group in derivatives of benzoic acid. J. Am. Chem. Soc. 1940, 62,
953–956.
53. The Northeastern Section of the American Chemical Society (NESACS):
James Flack Norris Award. www.nescas.org/awards/_norris.html (accessed
October 18, 2014).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch001

54. Hill, H. A. The James Flack Norris Award in Physical Organic Chemistry.
The NUCLEUS 1965, 42, 305–30.
55. Personal communication from Dr. Rich Gurney, Chair, Department of
Chemistry and Physics, Simmons College, to A. Greenberg on February 17,
2014.
56. Seeman, J. I. R. B. Woodward, a great physical organic chemist. J. Phys.
Org. Chem. 2014, 29, 708–721.

22
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Chapter 2

My Study of Optical Activity - From the Distant

Past to the Present with Stops in-Between
Kenneth B. Wiberg1,*
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Department of Chemistry, Yale University, New Haven, Connecticut 06520

1Present address: 865 Central Ave., Apt A404, Needham,

Massachusetts 02492
*E-mail: [email protected]

Physical organic studies encompass a variety of subjects. Our

work included studies of small ring compounds where the
higher energies lead to increased reactivity, studies of the
mechanisms of reactions including oxidations, the application
of ab initio theoretical calculations to studies of reactivity and
properties, as well as spectroscopic studies. More recent work
has concentrated on the relationship between structure and
conformations on specific rotations.

Introduction
In the late 1940’s when I was a graduate student at Columbia University
working with Bill Doering, I was given the problem of looking at the
stereochemistry of t-alkyl group migration in electron deficient molecular
rearrangements. It was known that s-alkyl groups migrated in a seemingly
concerted process, but t-alkyl was sufficiently different that it might involve bond
cleavage to a t-alkyl cation followed by recombination. Starting with a chiral
t-alkyl group this might lead to some loss of chirality. I proceeded to prepare
methylethylisobutylacetic acid and to resolve it via a salt with brucine (1). A
number of its reactions were studied and it appeared that chirality was normally
conserved.
In the course of this work I became interested in the connection (if any)
between structure and the sign and magnitude of the optical rotation. Not much
was known about the subject at that time, and so my interest lay mainly dormant

© 2015 American Chemical Society

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 for the next 50 years, except for noting the development of the quadrant and
octant rules (2), and Brewster’s rules (3).
In the meantime, I carried out a number of studies that had a physical-organic
context and I wrote a book (4) on the subject while enjoying a sabbatical leave
(1962) in Rudolf Criegee’s Institute at Karlsruhe, Germany.

Small Ring Compounds

Some of our work dealt with the synthesis and properties of
small ring compounds such as the remarkably strained and reactive
hydrocarbons, bicyclo[1.1.0]butane (1) (5), [1.1.1]propellane (2) (6), and
Δ1,4-bicyclo[2.2.0]hexane (3) (7). They have calculated strain energies of 64, 98
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

and 87 kcal/mol respectively (8). These values are large considering that the C-C
bond strength is 88 kcal/mol.

Despite its strain, 1 and its derivatives are readily formed via 1,3 elimination
reactions (5). The central bond is easily cleaved. The propellane, 2, also may
be formed via a 1,3-elimination reaction of the bridgehead dihalides via reaction
with t-butyllithium (6), and an elimination reaction from a tetrachloride provides
a convenient way in which to prepare it in some quantity (9). The structure is
interesting in that it has an "inverted" tetrahedral geometry at the bridgehead
carbons (10). It adds iodine to form the diiodide, and reducing agents readily
convert it back to 2. It is very reactive toward free radicals and can lead to
polymers (6). It’s thermal stability is probably related to the high energy of any
species formed by breaking a C-C bond. The alkene, 3, can be prepared in dilute
solutions, is a very reactive dienophile and these reactions lead to propellanes.
It readily dimerizes to form a pentacyclic propellane that opens to form an
interesting diene (7).
It was important to have experimental information about some of these
small ring compounds. This led us to determine the heats of formation of
cyclopropene, 1-methy-cyclopropene, methylenecyclopropane, bicyclobutane
and cyclobutene using a flame calorimeter (11). The heat of formation of the
"aromatic" cyclopropenyl cation could be derived from the heat of formation of
cyclopropene and the results of a mass spectrometric study (12). Our interest in
energies led us to make use of a precision solution calorimeter that we designed
to be completely computer controlled including data collection and analysis (13).
It was used to study steric interactions in the hydrolysis of orthoesters (14) and
lactones (15). It was also used in determining the heats of hydration of alkenes
(16), heats of reduction of ketones (17) as well as in a conformational study (18).
Solvolytic studies of exo and endo-5-bicyclo[2.1.0]pentyl tosylates
(4,5) gave a 106:1 relative reactivity (19) (much greater than that for the
24
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 corresponding norbornyl derivatives), and related studies of cis- and trans-fused
bicyclo[3.1.0]hexyl derivatives gave further information on the role of orbital
interactions on reactivity (20). The interest in these and related compounds led to
a study of what is now known as "molecular mechanics" and to develop the first
general computer program for strain energy minimization (21).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Mechanisms of Oxidation
Other studies dealt with the mechanisms of oxidation of organic compounds
by hydrogen peroxide (22), potassium permanganate (23) and chromic acid
(24). The reactions we studied included the oxidation of aldehydes, alkenes
and alkanes. Both MnO4- and HCrO4- usually undergo an initial two electron
reduction in their reactions with organic compounds, commonly leading to a
Mn(5) or Cr(4) intermediate. Using a stopped flow reactor and UV detection, it
was possible to observe the formation and subsequent reaction of the inorganic
intermediates, and to observe their rates of reaction (25). The studies of oxidation
reactions led to an interest in deuterium kinetic isotope effects (26) and to writing
a review on the subject (27).

Computational Studies
We became interested in using ab initio theory to supplement our experimental
studies. A group equivalent scheme was developed to easily estimate heats of
formation from the calculated total energies (28). One problem in making use of
this theory results from the fact that the calculations are for the gas phase whereas
most reactions are carried out in solution. We studied the use of Onsager’s reaction
field model (29) and methods derived from it (30) to estimate solvent effects, and
applied it to several problems (31).
Chemists have often been interested in electron density distributions and
assigning charges to atoms in molecules. We applied Bader’s theory of atoms
in molecules (32) and electron density difference maps to the question of why
the energies of fluorinated methanes are linearly dependent with the number of
fluorines, whereas this is not true with other substituents (33). We found that each
fluorine in carbon tetrafluoride has the same electron distribution and charge as
the fluorine in methyl fluoride. Thus, the carbon charge increases linearly with
the number of fluorines resulting in increasing C+-F- Coulombic attraction (34).
We have reviewed the differences in the charge that result from using the several
available models (35), and we have examined the relationship between bond
dipoles and the intensities of IR bands (36).
25
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 As a result of a question from G. B. Ellison, we became interested in
substituent effects on X-H bond dissociation energies (BDE). In the case of
RO-H, experimental results showed that the BDE decreased from 119 in HOH to
105 in MeO-H and 87 kcal/mol in HOO-H (37). We found that it was possible
to accurately reproduce these and other experimental results computationally
using CBS-APNO (38). Structural effects, calculations of spin densities, and
studies of electron density distributions showed that the main effect was a two
center-three electron conjugative interaction that stabilized the radical center (39).
A subsequent more extended study included second row elements and found in
interesting difference between P and S in their ability to interact with a radical
center (40). The differences between compounds with C=O and C=S also were
examined (41).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Spectroscopic Studies
NMR spectroscopy has become the most important structural tool for organic
chemists. It was known that hydrogen chemical shifts are normally upfield from a
hydrogen nucleus, whereas carbon chemical shifts are normally downfield from a
bare carbon nucleus (42). Thus, hydrogens experience a diamagnetic interaction
that depends on the electron density about it, whereas carbons experience mainly
a paramagnetic interaction that results from the coupling between the ground state
and the excited states of the molecule in the presence of a magnetic field. We have
examined the contribution of the excited states and the effect of their orientation
and thereby gained a better understanding of the carbon chemical shifts (43).
Ultraviolet spectroscopy provides information about some electronically
excited states of molecules. However, studies in solution lead to broadening of
vibrational structure in the UV bands. We have made use of vacuum ultraviolet
(VUV) spectroscopy for gas phase measurements including spectral regions
where atmospheric oxygen absorbs. Bicyclobutane is an example where the
vibrational structure of the bands could be observed, allowing the transitions to be
assigned (44). Calculations clarified its photochemical transformation leading to
butadiene and cyclobutene (45). Other studies made use of calculations to assist
in gaining an understanding of the experimental results (46).
Infrared spectroscopy provides information on the inner workings of
molecules. We have carried out extended studies of many molecules such
as cyclopropene (47), cyclohexane (36), bicyclo[.1.10]pentane (48) and
[1.1.1]propellane (49).

Back to Optical Activity

I "retired" in 1997 when I turned 70, and in that year Polavarapu published
the first ab initio optical rotation calculation (50). It made use of the 1928 report
by Rosenfeld (51) of his derivation for an equation for optical rotation based on
the newly developed wave mechanics:

26
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 where M is the molar mass in g/mol, v is the energy of the polarized light, νba is the
transition energy to exited state b, and Rba is the rotatory strength for the transition.
The rotatory strength, Rba is given by

and couples the ground state a to excited state b under the simultaneous action of
the electric (µE1) and magnetic (µM1) dipole transition operators. The summation is
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

over all of the excited states of the molecule. Thus, the observed rotation depends
on accurately calculating both the electric and magnetic transition moments for a
given transition and taking the product of them for each excited state, followed by
a summation. This becomes a formidable computational effort, and could not be
done until there was both improvements in ab initio theory and the development
of fast computers.
The electric transition moment, such as for the π→π* transition of an alkene
lies along the bond coordinate, whereas a magnetic transition moment, such as for
an n→π* transition of a ketone goes about the axis. A product of the two would
describe a helix that can be either right or left handed. This in turn can lead to
an interaction with polarized light leading to either positive or negative sign of
rotation.

Rather than using Rosenfeld’s equation given above, Polavarapu (50) recast it
in a form in which one would not have to do a sum over excited states, and allowing
a more direct calculation. This was followed by a number of related studies. One
that caught my attention was by Stephens, Devlin, Cheeseman and Frisch (52) that
made use of density functional theory (DFT) and showed that reasonably accuracy
(±35°) could be achieved.
This reinvigorated my interest in optical activity, especially when the needed
code became available in a development version of the Gaussian program (53), I
returned to thinking about my old interest in the relationship between structure and
optical rotation. In the following "optical rotations" (OR) will be used as a short
hand for specific rotations that are defined as

27
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 where the cell length is given in dm (usually 1.0) and the concentration is given in
g/ml or as the density with neat liquids. It has the units of deg. dm-1(g/mL)-1 The
wavelength of the polarized light is given as λ.

Experimental Studies of Optical Activity

When the rotation for a compound is measured, one observes the weighted
average over its possible conformations. Here was a place where calculations
could provide information that could not be obtained by direct measurement,
and the calculated values could be compared against experiments by applying
the mole fractions of the conformers derived from the energy differences among
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

conformers using the Boltzmann equation. Since for many compounds, these
energies are not experimentally available, a calculation of the conformer energies
allows the rotations to be estimated, providing a double check on the accuracy of
the calculations.
In our first study, we examined one of the simplest of chiral compounds, 3-
choro-1-butene (54). This was prepared in chiral form and the specific rotations
were obtained in several solvents. This compound has three conformers resulting
from rotation about the CH2=CH-CH2 single bond (Figure 1). The relative energies
of the conformers were obtained from a scan of the energy as a function of torsion
angle (Fig 2), and the rotation also was calculated as a function of the torsion angle
(Figure 3) The relative free energies of the conformers were obtained using G2 (55)
allowing them to be estimated at 298K giving 1.05 and 1.10 kcal/mol for the two
gauche conformers. The calculated results are in good accord with experimental
meaurements of the conformer energies (56).
It was surprising to find the wild swings of the rotation with a range of -526 at
τ = 180′° to +435 at τ = 320° when the compound has a maximum rotation of only
[α]D = ±57.3°. Using the populations of the two conformers obtained from the G2
energy differences, and the calculated rotations shown in Figure 3, an estimated
rotation is on the order of -99.7 whereas the observed rotation was [α]D = -57.3.
Thus the rotation was estimated to be twice as large as observed. Since this was
one of the earliest studies of its type and led to remarkably large changes in OR, it
seemed important to obtain further information. It might be noted that a later study
by Tam, Abrams and Crawford (57) found that density functional theory was not
adequate for calculations for this compound, and that a coupled-cluster method
(CCSD) was able to give a much more satisfactory result.
In order to see if the double bond were responsible for the deviation between
calculated and observed rotations, we examined 2-substututed butanes where X =
F, Cl, CN and C=CH (58). These substituents are convenient in that they do not
add to the number of conformers. In addition they would be expected to lead to
significant differences in electronic levels that are involved in the calculation of
optical rotation, and this was confirmed by measurement of their VUV spectra. It
is well known that compounds such as these have three conformations, trans and
two gauche. The energies of the conformers were obtained by using G3.

28
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Figure 1. 3-Chloro-1-butene conformers

Figure 2. Calculated conformational energies

29
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Figure 3. Calculated Optical rotation profile for 3-chloro-2-butene

The OR of these compounds were calculated using B3LYP/aug-cc-pVDZ, and

the results are summarized in Figure 4. Despite the obvious differences among
the substituents there is remarkable similarity between the curves for the four
compounds. This suggests that in this case, substituents have a small effect other
than creating the chiral center. This was a surprising conclusion.
The observed specific rotation is the weighted average for the three
conformers. The averaged rotations were calculated using G3 (59) calculated
conformational energies allowing a comparison with the experimental rotations
(Table 1).
With these compounds there is fairly good agreement between experiment
and theory. Here, one must remember that the calculations refer to the gas phase
whereas the observed values are for the condensed phase. In some cases these
differences can become significant.
The agreement between observation and calculations could be fortuitous, and
it remained to be seen if the calculated rotations for the individual conformers of 2-
chlorobutane are in agreement with the calculated values. Some useful information
may be obtained from the temperature dependence of the optical rotation that
results from the effect of temperature on the conformer ratio. The conformational
energies have been studied experimentally giving an average value of ΔG ≈ 700
cal/mol. The data from the temperature dependence study should be given by

where αA is the rotation for the low energy conformer and α′ is the average of
that for the higher energy conformers. A least squares treatment gave a specific
rotations of -77.4° for the low energy conformer and 37.4° for the average of the
30
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 higher energy conformers. This may be compared with the calculated values of
-77.6° and 34.6°. The remarkably good agreement is probably in part fortuitous,
but the result confirm that the DFT-OR calculations for 2-chloro- butane are
quite good. A similar study for 2-cyanobutane (2-methylbutyronitrile) found
slightly poorer agreement with the calculated rotations about 25% larger than the
experimental values.
These results suggest that the calculations for saturated compounds may be
significantly more satisfactory than for their unsaturated analogs. This may be a
result of the double bond π→π* transition not being treated well using DFT.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Figure 4. Relationship between calculated optical rotations and the C2-C3

torsion angle

31
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Table 1. Calculated and observed specific rotations at 589 nm for
conformational mixtures at 25° C.
X Calculated Observed,neat Observed,soln
F -17.2 na
Cl -37.1 -33.8 -33.8
CN -45.1 -36.3 -34.1
HCC -57.8 -46.8
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Twisted Achiral Compounds

The large conformational effects suggest that achiral compounds may give
significant optical rotations when twisted from planarity. This was examined
for a number of hydrocarbons including 1-butene and n-butane (Figure 5) (60),
and showed that these compounds can give large rotations in their twisted forms.
However, there is a significant effect of theoretical method on the observed values.

Figure 5. Calculated optical rotations for twisted butane.

32
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The calculated optical rotation as a function of C-C torsion energy for butane
has a simple maximum at 90° with the 6-311++G** basis set, but it become much
different using the aug-cc basis sets. This is seen more dramatically in a study of
30° rotated ethane (i.e. halfway between staggered and eclipsed). Some calculated
rotations are given in Table 2.

Table 2. Effect of basis set on B3LYP calculated specific rotations at 589

nm for 30° rotated ethane.
Basis set bf [α]D
6-311++Gd 86 -495.1
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

6-311(2+,2+)Gdf 100 -505.1

6-311++(2d,2f) 114 -226.3
6-311++G(3df,3pd) 186 -19.5
cc-pVDZ 58 -208.4
cc-pVTZ 144 -215.3
aug-cc-pVDZ 100 +41.2
aug-cc-pVTZ 230 +53.6
aug-cc-pVQZ 436 +57.8
UGBSIP 766 +59.1

Most of the details concerning the basis functions used in the calculations
are not of importance in the present context. The column bf gives the number of
functions (s,p,d..) that were used in the calculation, and generally the larger the
number the better the result. The first 4 basis functions are of the "Pople" type
and they are very satisfactory for most purposes such as geometry optimizations
and vibrational frequency calculations, whereas the latter basis functions are of
the "Dunning" type that also are normally very satisfactory. But, as seen from
the calculated results, especially the change in sign, in the present case there is a
large difference with the Dunning functions giving the correct sign as shown by
the largest basis set calculations
These basis sets include extra functions beyond the basic group. It is possible
to improve on the simple s, p, d .. set by allowing these nucleus centered functions
to move in response to the electric fields in the molecule. This happens when
a small amount of a p-function is added to s, and when some d is added to p,
etc. In addition, it sometimes is useful to include "diffuse" functions that are
much larger than those in the basic group. They are indicated by + in the Pople
sets and by aug in the Dunning sets and are frequently needed in calculations
for anions (61). Hydrogens diffuse functions are rarely important for geometry
optimizations and so only diffuse s functions are included in the "Pople" sets but p

33
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 diffuse functions also are included to the "Dunning" sets. It is these latter functions
on hydrogens that lead to the large differences noted in Table 2. The addition of
p-diffuse functions to the former set corrects the problem.
Why should these seldom needed functions be so important for optical activity
calculations? A clue is given by the density difference plot in Figure 6 that shows
the difference resulting from the addition of these functions (60). At a relatively
far distance from the nucleus, there are some helical differences that result from
the directional nature of the diffuse p functions. Optical activity results from
the helical character of the wave function, and switching stereoisomers results in
switching the sense of helicity and the sign of rotation. This effect of helicity is
strikingly seen in the optical rotations of helicenes (6) (62) and triangulenes (7)
(spiro-linked cyclopropanes) (63) that give very large rotations.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Figure 6. Difference density plot between calculations with and without diffuse p
functions on hydrogens of twisted ethane.

34
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Use of Minimal Basis Sets

Calculations of optical rotation have come to using increasingly large basis
sets with many polarization functions and diffuse functions. The observations
given above made us think that much of this complexity might not be essential for
the calculations. Would it be possible to reproduce their results using a minimal
basis set (STO-3G or 3-21G) to represent the core and valence electrons and
then add a full set of diffuse functions (64)? Could a minimal basis set "steer"
the diffuse functions that are relatively far from the nuclei so that they adopt
the essential helical character of the molecule? In a non-resonant situation, the
polarized light used to interrogate the molecule might interact mainly with the
outer diffuse electron density leading to rotation of the plane of polarization.
This seemed something of a wild idea, but we decided to explore it. The study
was concerned with how well the large basis set B3LYP calculations could be
reproduced.
We used both STO-3G and 3-21G as the minimal basis sets and added
diffuse functions from the aug-cc-pVDZ and aug-cc-pVTZ basis set. The object
was to see if the results from B3LYP calculations using large basis sets could
be reproduced, A number of compounds were examined including α-pinene,
3-carene, norbornenone and 2,3-pentadiene (64). The results were quite good,
and an example from the carenes is shown in Figure 7.
Here the rotation increases as the number of basis functions is increased. They
reach an asymptote as shown by the horizontal lines and the points are numbered.
The first two points are for STO-3G and 3-21G that are very small basis sets, and
it is clear that they do not give satisfactory computational results. Point 5 is for
cc-pVDZ (no diffuse functions). Point 3 involves just the addition of diffuse s, p,
d functions to STO-3G and gives a calculated rotation that is essentially the same
as that derived from the use of a very large basis set . Even with large basis sets
such as for points 14(cc-pVTZ) and 20 (cc-pVQZ), diffuse functions are needed
giving points 19 and 21. The remaining points are identified in ref. (64).

35
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Figure 7. Effect of basis set on the calculated optical rotation of 3-carene at 355
nm (upper), 589 nm (center) and 633 nm (lower). The x axis gives the number of
basis functions on a logarithmic scale. The open symbols represent unaugmented
basis sets and the solid symbols are for the augmented sets.

Gas Phase Measurements

Until recently studies of optical rotation could not properly compare the
calculated values with the experimental results because they were obtained in the
condensed phase, either neat or in solution whereas the theoretical calculations
36
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 refer to the gas phase. Although a heroic effort by Biot (65) in 1817 to study
the optical rotation of turpentine in the gas phase showed that the vapor had a
rotation, the experiment ended rather disastrously when the vessel containing
the boiling turpentine broke open and was set on fire. There were some other
attempts to carry out gas phase studies but they were hampered by the low density
of molecules in the gas phase and the small rotations for most molecules.
This problem was solved by my colleague Prof. Patrick Vaccaro who made
use of a cavity ring-down cell that allow multiple passes through the cell (Figure
8) (66).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Figure 8. Cavity ring-down polarimeter

Here, the sample is placed in an evacuated cell having high reflexance mirrors
at each end. A laser pulse is prepared and sent into the cell though one mirror. The
light that gets into the cell will be reflected by the other mirror, and at each pass,
some light will leak through the second mirror and will be analyzed. This provides
the opportunity to have the effect of a very long path length. But, this alone would
not lead to the observation of optical rotation since the effect of light moving in
the forward direct would be reversed in the back direction. This can be avoided
by inserting two quarter-wave plates, and the region between them will respond to
the optical rotation of the sample.
Chloropropionitrile is one of the simplest chiral compounds, and we have
examined the (S)-(-) form derived from (S)-(+)-alanine in some detail (67). It has a
rotation of [α]D = -14.5 as a neat liquid. A determination of the gas-phase rotation
gave [α]D = -8.3. It can be seen that there is a significant effect of phase. The OR
in solution was obtained in a number of solvents, giving a range of [α] from -11.6
in acetonitrile, to -17.4 in cyclohexane and -34.6 in benzene.
We have studied solvent effects on equilibria and reaction rates and found that
free energies in cyclohexane, di-n-butyl ether, acetone and acetonitrile usually
are linearly related to the Onsager dielectric constant function, (ε+1)/(ε+2) (29).
The function takes a zero value for the gas phase and goes to 0.5 for the highest
dielectric constant solvents. The correlation would be expected based on the
reaction field model of solvent effects (29). Extrapolation usually yields a value
close to that observed in the gas phase when this is available (31).
37
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Application of this model to chloropropionitrile OR gave the result shown in
Fig 9. There is a linear correlation, but extrapolation to the gas phase led to a value
of -22, quite far from the observed value of -8.3. We have found that this is the
usual case for optical rotation. Chemists generally think of cyclohexane solutions
to be similar to the gas phase, but for optical rotation, acetone and acetonitrile
solutions usually give the closest OR to the gas phase.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Figure 9. Relationship between optical rotation and solvent dielectric constant

The effect of solvent on optical rotations has been studied computationally

at the DFT level by using a polarizable continuum model (PCM) (68). It was
unfortunately not able to give a very satisfactory result, but better results have
been obtained by using discrete solvent molecules in the calculations (69).
However, this involves averaging over solvent conformations, and the added
molecules greatly lengthen the execution times.

Sum-over States Calculations

It is clear that excited states are intimately connected with optical rotation. It
would be of interest to know which states are the most important. For example,
it is commonly thought that the carbonyl group is very important in the OR
of ketones (2). It is possible to obtain information on this subject from a
sum-over-states calculation of OR making direct use of the Rosenfeld equation.
Electronically excited states can be obtained by calculations using time dependent
38
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 density functional theory (TDDFT) and this code in Gaussian also gives the
Rotatory strength for each transition. Thus, it is now relatively easy to carry out
the calculation. We have examined this for a number of molecules (70).
Chloropropionitrile gives a typical result. As with most molecules, the first
excited states are not the main source of the observed OR as may be seen in Fig 10.
The rotation continues to depend on the number of excited states beyond 100. It is
necessary to examine a very large number of states (Figure 11) in order to come to
a constant value for this quantity. It is fortunate that linear response methods have
been developed that do not need to deal with excited states directly.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Figure 10. Sum-over-States Study for Chloropropionitrile including the first

125 excited states.

Figure 11. Sum-over-States for Chloropropionitrile including >1500 excited

states.
39
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

Figure 12. Relationship between specific rotations for a norbornenone model

and its C=C-C-C torsion angle

Norbornenone
Of the compounds that we have examined, there is one, (S)-(-)-norbornenone
(6), that behaves differently. This compound was first prepared in chiral form by
Mislow and Berger (71). They found a specific rotation in isooctane solution of
[α]D = -1160 that is very large for such a small and relatively simple molecule.
The saturated derivative, norbonanone (7) has a rotation of ~20. A B3LYP/6-
311+G* calculation for 6 gave a rotation of -1260 (52) that was in remarkably
good agreement with the experiment. A sum-over-states study found that the first
electronic transition was largely responsible for the rotation (70). This is one of
the few molecules for which this is the case.

40
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The large rotation clearly involved an interaction between the n→π* transition
for the carbonyl group that leads to a large magnetic transition dipole and the
π→π* transition of the carbon-carbon double bond that lead to an electric transition
dipole. One way to examine the interaction of these groups is to delete the remote
bridgehead carbon and then examine the effect of rotating the C=CH2 with respect
to the C=O (54). The results of this study are shown in Fig 12. The largest negative
rotation is found at (O=)C-C-C=C angle similar to that found in norbornenone
itself. This orientation is the one that allows the maximum interaction between
the C=C and C=O. This model has been examined in more detail by Caricato, et.
al. (72) who examined how the relative orientations of the electric and magnetic
transition dipoles change with torsion angle. The magnitude of the optical rotation
in norbornenone has also been examined by Moore, Srebro and Autschbach (73).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

It was noted above that the DFT calculated rotation of (S)-(-)-norbornenone

is in very good agreement with that reported for an isooctane solution However, a
later calculation by Mach and Crawford at the CCSD level gave a much smaller
predicted rotation (-561 ) (74) whereas for a DFT calculation at the same level,
[α]D = -1260. If that were correct, there must be a large solvent effect or a problem
with the CCSD calculations. In order to determine which was correct, I prepared
a sample of (R)-(+)-norbornenone, and Priyanka Lahiri measured its rotation in
the gas phase (75). giving [α]D = 755, about half-way between the two calculated
values. Further theoretical developments are needed in order to reproduce the
experimental optical rotations.
Norbornenone also shows a large solvent effect on its optical rotation. As
compared to the gas phase [α]D = 755, in acetonitrile it was 950, and in cyclohexane
it was 981 (72). This shows the importance of referring the solution phase optical
rotations to the gas phase rotations.

Acknowledgments
I would like to thank the Chemistry faculties at the University of Washington
and Yale University for the help and encouragement they have given with regard
to my scientific work. I would specially thank Patrick Vaccaro, James McBride,
Harry Wasserman, Jerry Berson, William Jorgensen and Kurt Zilm at Yale and in
addition I would like to recognize G.B. Ellison (U. Colorado); George Petersson
(Wesleyan), and Jim Cheesman, Marco Caricato, Gary Trucks and Mike Frisch
(Gaussian). Most of my work was supported by the National Science Foundation.
I would also like to extend my appreciation to the 100+ graduate student and
many postdoctoral workers without whom little could be done. I wish I could have
noted all of their work but that would require much more space and time than was
available.

References
1. Doering, W. v. E.; Wiberg, K. B. The Synthesis and Resolution of
Methylethylisobutylacetic Acid. J. Am. Chem. Soc. 1950, 72, 2608–2610.
41
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 2. Cf. Eliel, E. L.; Willen, S. H. Stereochemistry of Organic Compounds;
Wiley-Interscience: New York, 1994; p 1022ff.
3. Brewster, J. H. Some Applications of Conformational Dissymmetry Rule.
Tetrahedron 1961, 13, 106–122.
4. Wiberg, K. B. Physical Organic Chemistry; Wiley: New York, 1965.
5. Wiberg, K. B.; Ciula, R. P. Ethyl Bicyclo[1.1.0]butane-1-carboxylate. J.
Am. Chem. Soc. 1959, 81, 5261–5262. Wiberg, K. B.; Lampman, G. M.;
Ciula, R. P.; Connor, D. S.; Schertler, P.; Lavanish, J. Bicyclo[1.1.0]butane.
Tetrahedron 1965, 21, 2749–2769.
6. Wiberg, K. B.; Walker, F. H. [1.1.1]Propellane. J. Am. Chem. Soc. 1982,
104, 5239. Wiberg, K. B.; Waddell, S. T. Reactions of [1.1.1]Propellane. J.
Am. Chem. Soc. 1990, 112, 2194–2216.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

7. Wiberg, K. B.; Burgmaier, G. J.; Warner, P. Δ1,4-Bicyclo[2.2.0]hexene. J.

Am. Chem. Soc. 1971, 93, 246–247. Wiberg, K. B.; Matturro, M. G.;
Okarma, P. J.; Jason, M. E.; Dailey, W. P.; Burgmaier, G. J.; Bailey, W. F.;
Warner, P. Bicyclo[2.2.0]hex-1(4)-ene. Tetrahedron 1986, 42, 1895–1902.
8. Wiberg, K. B. The Concept of Strain in Organic Chemistry. Angew. Chem.,
Int. Ed. 1986, 23, 312–322.
9. Semmler, K.; Szeimies, G.; Belzner, J. Tetracyclo[5.1.0.01,8.02,7]octane, A
[1.1.1]Propellane Derivative and a New Route to the Parent Hydrocarbon. J.
Am. Chem. Soc. 1985, 107, 6410–6411.
10. Wiberg, K. B. Inverted Tetrahedral Geometries at Carbon. Acc. Chem. Res.
1985, 26, 379–386.
11. Wiberg, K. B.; Fenoglio, R. A. Heats of Formation of C4H6 Hydrocarbons.
J. Am. Chem. Soc. 1968, 90, 3395–3397.
12. Wiberg, K. B.; Bartley, W. J.; Lossing, F. P. On the Strain Energy in
Cyclopropene and the Heat of Formation of the C3H3+ Ion. J. Am. Chem.
Soc. 1962, 84, 3980–3981.
13. Wiberg, K. B.; Squires, R. R. A Microprocessor-controlled System for
Precise Measurement of Temperature Changes. Determination of the
Enthalpies of Hydrolysis of some Polyoxygenated Hydrocarbons. J. Chem.
Thermodyn. 1979, 11, 773–786.
14. Wiberg, K. B.; Martin, E. J.; Squires, R. R. Thermochemical Studies of
Carbonyl Compounds. 3. Enthalpies of Hydrolysis of Orthoesters. J. Org.
Chem. 1985, 50, 4717–4720.
15. Wiberg, K. B.; Waldron, R. F. Lactones. 2. Enthalpies of Hydrolysis,
Reduction and Formation of the C4-C13 Monocyclic Lactones. Strain
Energies and Conformations. J. Am. Chem. Soc. 1991, 113, 7697–7705.
16. Wiberg, K. B.; Wasserman, D. J.; Martin, E. Enthalpies of Hydration of
Alkenes 2. The n-Heptenes and n-Pentenes. J. Phys. Chem. 1984, 88,
3684–3688.
17. Wiberg, K. B.; Crocker, L. S.; Morgan, K. M. Thermochemical Studies of
Carbonyl Compounds. 5. Enthalpies of Reduction of Carbonyl Groups. J
Am. Chem. Soc. 1991, 113, 3447–3450.
18. Bailey, W. F.; Connon, H.; Eliel, E. L.; Wiberg, K. B. Calorimetric
Determination of the Conformational Enthalpy and Entropy of 2-Phenyl-

42
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 1,3-Dioxane. Effectively Free Rotation of an Equatorial 2-Phenyl Group. J.
Am. Chem. Soc 1978, 100, 2202–2209.
19. Wiberg, K. B.; Lowry, B. R.; Colby, T. H. Bicyclo[2.1.1] Derivatives. J. Am.
Chem. Soc. 1961, 83, 3998–4006.
20. Wiberg, K. B.; Pfeiffer, J. Orbital Factors in Cyclobutyl Solvolysis. J.
Am. Chem. Soc. 1968, 90, 5324–5325. Wiberg, K. B.; Williams, V. Z.;
Friedrich, L. E. Solvolysis of Bicyclo[2.1.0]-2-pentyl Derivatives. J. Am.
Chem. Soc. 1968, 90, 5338–5339.
21. Wiberg, K. B. A Scheme for Strain Energy Minimization. Application to the
Cycloalkanes. J. Am. Chem. Soc. 1965, 87, 1070–1078. Wiberg, K. B.;
Boyd, R. H. Application of Strain Energy Minimization to the Dynamics of
Conformational Changes. J. Am. Chem. Soc. 1972, 94, 8426–8430.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

22. Wiberg, K. B. The Mechanisms of Hydrogen Peroxide Reactions. 1. The

Conversion of Benzonitrile to Benzamide. J. Am. Chem. Soc. 1953, 75,
3961–3964.
23. Wiberg, K. B.; Stewart, R. The Mechanisms of Permanganate Reactions. 1.
The Oxidation of Some Aromatic Aldehydes. J. Am. Chem. Soc. 1955, 77,
1159–1163.
24. Wiberg, K. B.; Mill, T. The Kinetics of the Chromic Acid Oxidation of
Benzaldehyde. J. Am. Chem. Soc. 1958, 80, 3022–3029.
25. Wiberg, K. B.; Schaefer, H. Direct Observation of Intermediates in the
Chromic Acid Oxidation of Secondary Alcohols. J. Am. Chem. Soc.
1967, 89, 455–457. Wiberg, K. B.; Deutsch, C. J.; Rocek, J. Permanganate
Oxidation of Crotonic Acid, Spectrometric Detection of an Intermediate. J.
Am. Chem. Soc. 1973, 95, 3034–3035.
26. Wiberg, K. B.; Slaugh, L. H. The Deuterium Isotope Effect in the Side-Chain
Halogenation of Toluene. J. Am. Chem. Soc. 1958, 80, 3033–3039.
27. Wiberg, K. B. The Deuterium Isotope Effect. Chem. Rev. 1955, 55, 713–743.
28. Wiberg, K. B. Group Equivalents for Converting ab initio Energies to
Enthalpies of Formation. J. Comput. Chem. 1984, 5, 197–199.
29. Onsager, L. Electric Moments of Molecules in Liquids. J. Am. Chem. Soc .
1936, 58, 1486–1493.
30. Wong, M. W.; Frisch, M. J.; Wiberg, K. B. Solvent Effects. 1. The
Mediation of Electrostatic Effects by Solvents. J. Am. Chem. Soc. 1991,
113, 4776–4783. Foresman, J. B; Keith, T. A.; Wiberg, K. B.; Snoonian, J.;
Frisch, M. J. Solvent Effects. 5. The Influence of Cavity Shape, Truncation
of Electrostatics and Electron Correlation on Ab Initio Reaction Field
Calculations. J. Phys. Chem. 1996, 100, 16098–16104.
31. Wiberg, K. B.; Keith, T. A.; Frisch, M. J.; Murcko, M. Solvent Effects
on 1,2-Dihaloethane Gauche/Trans Ratios. J. Phys. Chem. 1995, 99,
9072–9079. Wiberg, K. B.; Rablen, P. R.; Rush, D. J.; Keith, T. A.
Amides. 3. Experimental and Theoretical Studies of the Effect of
Medium on the Rotational Barriers for N,N-Dimethylformamide and
N,N-Dimethylacetamide. J. Am. Chem. Soc. 1995, 117, 4261–4270.
32. Bader, R. F. W. Atoms in Molecules, A Quantum Theory; Oxford University
Press: 1990.

43
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 33. Wiberg, K. B.; Rablen, P. R. Origin of the Stability of Carbon Tetrafluoride.
Negative Hyperconjugation Reexamined. J. Am. Chem. Soc. 1993, 115,
614–625.
34. Wiberg, K. B. J. Chem. Educ. 1996, 73, 1089–1095.
35. Wiberg, K. B.; Rablen, P. R. Comparison of Atomic Charges Derived via
Different Procedures. J. Comput. Chem. 1993, 14, 1504–1518.
36. Wiberg, K. B.; Wendoloski, J. J. Charge Redistribution in the Molecular
Vibrations of Acetylene, Ethylene, Ethane, Methane, Silane and Ammonium
Ion. Signs of the M-H Bond Moments. J. Phys. Chem. 1984, 88, 586–593.
Wiberg, K. B.; Walters, V. A.; Dailey, W. P Infrared Intensities: Cyclohexane.
A Molecular Force Field and Dipole Moment Derivatives. J. Am. Chem. Soc
1985, 107, 4860–4867.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

37. Blanksby, S. J.; Ellison, G. B. Bond Dissociation Energies of Organic

Molecules. Acc. Chem. Res. 2003, 36, 255–263.
38. Montgomery, J. A, Jr.; Ochterski, J. W.; Petersson, G. A. A Complete Basis
Set Model Chemistry. J. Chem. Phys. 1994, 101, 5900–5909.
39. Wiberg, K. B.; Ellison, G. B.; McBride, J. M.; Petersson, G. A. Substituent
Effects on O-H Bond Dissociation Enthalpies: A Computational Study. J.
Phys. Chem. A 2013, 117, 213–218.
40. Wiberg, K. B.; Petersson, G. A. A Computational Study of RXHn X-H Bond
Dissociation Enthalpies. J. Phys. Chem. A 2014, 118, 2353–2359.
41. Wiberg, K. B.; Wang, Y.-g.; Miller, S. J.; Puchlopeck, A. I. A;
Bailey, W. F.; Fair, J. D. Disparate Behavior of Carbonyl and Thicarbonyl
Compounds: Acyl Chlorides vs Thiocarbonyl Chlorides and Isocyanates
vs Isothiocyanates. J. Org. Chem. 2009, 74, 3659–3664. Wiberg, K. B.;
Bartelson, A. L.; Bailey, W. F. Disparate behavior of ketones and thioketones
on reaction with organolithiums. Terahedron Lett. 2011, 52, 2169–2171.
Wiberg, K. B.; Bailey, W. F.; Petersson, G. A. Computational Study of the
Properties of Small Molecules Containing O, S and Se. J. Phys. Chem. A
2011, 115, 12624–12630.
42. Pople, J. A.; Schneider, W. G.; Bernstein, H. J. High Resolution Nuclear
Magnetic Resonance Spectroscopy; McGraw-Hill: New York, 1959.
43. Wiberg, K. B; Hammer, J. D.; Zilm, K.CheesemanJ. R.; Keith, T. A. NMR
Chemical Shifts. 1. The Role of Atomic Orbital Phase in Determining the
Sign and Magnitude of the Paramagnetic Terms: ClF, CH3F, CH3NH3+,
FNH3+, and HC≡CF. J. Phys. Chem. 1998, 102, 8766–8773. Wiberg, K.
B.; Hammer, J. D.; Keith, T. A.; Zilm, K. NMR Chemical Shifts. 2.
Interpretation of the Carbon Chemical Shifts in Monocyclic Aromatic
Compounds and Carbenes. J. Phys. Chem. 1999, 103, 21–27.
44. Wiberg, K. B.; Ellison, G. B.; Peters, K. S Ultraviolet Spectrum of
Bicyclobutane. J. Am. Chem. Soc. 1977, 99, 3941–3946. Wiberg, K. B.;
Peters, R. S.; Ellison, G. B.; Alberti, F. High Resolution Spectrum of the A
State of Bicyclobutane. J. Am. Chem. Soc. 1977, 99, 3946–3951.
45. Rossi, A. R.; Wang, Y.; Wiberg, K. B. Excited States and Photochemistry of
Bicyclo[1.1.0]butane. J. Phys. Chem. A 2009, 113, 1686–1695.
46. Wiberg, K. B.; Hadad, C. M.; Foresman, J. B.; Chupka, W. A. Electronically
Excited States of Ethylene. J. Phys. Chem. 1992, 96, 10756–10768.

44
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Wiberg, K. B.; DeOliveira, A. E.; Trucks, G. A Comparison of Electronic
Transition Energies for Ethene, Isobutene, Formaldehyde and Acetone
Calculated Using RPA, TDDFT, and EOM-CCSD. Effect of Basis Sets. J.
Phys. Chem. 2002, 67, 4436–4440.
47. Eggers, D. F.; Schultz, J. W.; Wiberg, K. B.; Wagner, E. L.; Jackman, L. M.;
Erskine, R. L. Vibrational Spectrum of Cyclopropene. J. Chem. Phys. 1967,
47, 946–955.
48. Wiberg, K. B.; Rosenberg, R. E.; Waddell, S. T. Infrared Intensities.
Bicyclo[1.1.1]pentane. A Normal Coordinate Analysis and Comparison
with [1.1.1]Propellane. J. Phys. Chem. 1992, 96, 8293–8303.
49. Wiberg, K. B.; Daley, W. P.; Walker, F. H.; Waddell, S. T.; Crocker, L.
S.; Newton, M. Vibrational Spectrum, Structure and Energy of
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

[1.1.1]Propellane. J. Am. Chem. Soc. 1985, 107, 7247–7257.

50. Polavarapu, P. L. Ab Initio Molecular Optical Rotations and Absolute
Configurations. Mol. Phys. 1997, 91, 551–554.
51. Rosenfeld, L. Z Quantenmechanische Theorie der Natürlichen Optischen
Aktivität von Flüssigkeiten und Gasen. Physik 1928, 52, 161–175.
52. Cheeseman, J. R.; Frisch, M. J.; Devlin, F. J.; Stephens, P. J. Hartree-Fock and
Density Functional Theory ab initio Calculation of Optical Rotation Using
GIAOs: Basis Set Dependence. J. Phys. Chem. A 2000, 104, 1039–1046.
53. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.;
Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino,
J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda,
R.; Hasegawa, J.; Isheda, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Vreven, T; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark,
M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi,
R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S.
S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.;
Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann,
R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ő.; Foresman, J. B.;
Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision C.01; Gaussian,
Inc.: Wallingford, CT, 2009.
54. Wiberg, K. B.; Vaccaro, P. H.; Cheeseman, J. R. Conformational Effects on
Optical Rotation. 3-Substituted 1-Butenes. J. Phys. Chem. A 2003, 125,
1888–1896.
55. Curtiss, L. A.; Raghavachari, K.; Trucks, G. W.; Pople, J. A. Gaussian-2
Theory for Molecular Energies of 1st-Row and 2nd-Row Compounds. J.
Chem. Phys. 1991, 94, 7221–7230.
56. Schei, S. H. 3-Chloro-1-butene - Gas Phase Molecular Structure and
Conformation as Determined by Electron-Diffraction and by Molecular
Mechanics and ab Initio Calculations. J. Mol. Struct. 1984, 118, 319–332.
57. Tam, M. C.; Abrams, M. I.; Crawford, T. D. Chiroptical Properties of (R)-
3-Choro-1-butene and (R)-2-Chlorobutane. J. Phys. Chem. A 2007, 111,
11232–11241.

45
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 58. Wiberg, K B.; Wang, Y.; Vaccaro, P. H.; Cheeseman, J. R.; Luderer, M. R.
Conformational Effects on Optical Rotation. 2-Substituted Butanes. J. Phys.
Chem. A 2005, 109, 3405–3410.
59. Curtiss, L. A.; Raghavicharan, K. Gaussian-3 and Related Methods for
Accurate Thermochemistry. Theor. Chem. Acc. 2002, 108, 61–70.
60. Wiberg, K. B.; Wang, Y-g.; Vaccaro, P. H.; Cheeseman, J. R.; Trucks, G.;
Frisch, M. J. Optical Activity of 1-Butene, Butane and Related
Hydrocarbons. J. Phys. Chem. A 2004, 108, 32–38.
61. Spitznagel, G. W.; Clark, T.; Chandrasekhar, J.; Schleyer, P. v. R.
Stabilization of Methyl Anions by First-Row Substituents. The Superiority
of Diffuse Function Supplemented Basis Sets for Anion Calculations. J.
Comput. Chem. 2004, 3, 363–371.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

62. Newman, M. S.; Lednicer, D. The Synthesis and Resolution of Hexahelicene.

J. Am. Chem. Soc. 1956, 78, 4765–4770.
63. de Meijere, A.; Kozhushkov, S. I.; Schill, H. Three-Membered Ring Based
Molecular Architectures. Chem. Rev. 2006, 106, 4926–4996.
64. Wiberg, K. B.; Caricato, M.; Wang, Y.-g.; Vaccaro, P. H. Towards the
Accurate and Efficient Calculation of Optical Rotatory Dispersion Using
Augmented Minimal Basis Sets. Chirality 2013, 25, 606–616.
65. Biot, J. B. Memoire sur une Modification Remarquable Qu’Eprouvent les
Rayons Lumineux dans Leur Passage a Travers Certains Corps Diaphanes,
et Sur Quelques Autres Nouveaux Phenomenes d’Optique. Mem. Acad. Sci.
1817, 2, 114–119.
66. Müller, T.; Wiberg, K. B.; Vaccaro, P. H. Cavity Ring-Down Polarimetry
(CRDP). A New Scheme for Probing Circular Birefringence and Circular
Dichroism in the Gas Phase. J. Phys. Chem. A 2000, 104, 5959–5968.
67. Wiberg, K. B.; Wang, Y.-g.; Wilson, S. M.; Vaccaro, P. H.; Cheeseman, J.
R. Chiroptical Properties of 2-Chloropropionitrile. J. Phys. Chem. A 2005,
109, 3448–3453.
68. Mennucci, B.; Tomasi, J.; Cammi, R.; Cheeseman, J. R.; Frisch, M. J.;
Devlin, F J.; Stephens, P. J. Polarizable Continuum Model (PCM) of Solvent
Effects on Optical Rotations of Chiral Molecules. J. Phys. Chem. A 2002,
106, 6102–6113.
69. Cf. Lipparini, F.; Egidu, F.; Chaappelli, C.; Barone, V. The Optical Rotation
of Methyloxirane in Aqueous Solution; A Never Ending Story? J. Chem.
Theory Comput. 2013, 9, 1880–1884.
70. Wiberg, K. B.; Wang, Y.; Wilson, S. M.; Vaccaro, P. H.; Cheeseman, J. R.
Sum-over-States Calculation of the Specific Rotation of Some Substituted
Oxiranes, Chloropropionitrile, Ethane and Norbornenone. J. Phys. Chem. A
2006, 110, 13995–14003.
71. Mislow, K.; Berger, J. D. Dissymmetic Non-Conjugated Chromophores-
Optical Rotatory Dispersion of Bicyclo[2.2.1]hept-5-ene-2-one and
Bicyclo[2.2.2]oct-5-en-2-one. J. Am. Chem. Soc. 1962, 84, 1956–1961.
Sandman, D. J.MislowK. Optical Purity and Optical Rotatory Power of
Dehydronorcamphor. J. Org. Chem. 1968, 33, 2924–2496.

46
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 72. Caricato, M.; Vacaro, P. H.; Crawford, T. D.; Wiberg, K. B.; Lahiri, P.
Insights on the Origin of the Unusually Large Specific Rotation of
(1S,4S)-Norbornenone. J. Phys. Chem. A 2014, 116, 4863–4871.
73. Moore, B., II; Srebro, M.; Autschbach, J. Analysis of Optical Activity in
Terms of Bonds and Lone-Pairs: The Exceptionally Large Optical Rotation
of Norbornenone. J. Chem. Theory Comput. 2012, 8, 4336–4346.
74. Mach, T. J.; Crawford, T. D. Basis Set Dependence of Coupled Cluster
Optical Rotation Calculations. J. Phys. Chem. A 2011, 115, 10045–10051.
75. Lahiri, P.; Wiberg, K. B.; Vaccaro, P H.; Caricato, M.; Crawford, T. D.
Large Solvation Effect on the Optical Rotatory Dispersion of Norbornenone.
Angew. Chem., Int. Ed. 2014, 53, 1386–1389.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch002

47
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Chapter 3

Some Thermochemical Studies of Solvation

from the 1960s and 1970s
Edward M. Arnett*
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch003

Duke University, Durham, North Carolina 27705

*E-mail: [email protected]

Initial commentary refers to the stressful emergence of physical

organic chemistry as a post-WWII interdisciplinary field
benefitting greatly from instrument development. Design of a
simple solution calorimeter in the author’s laboratory allowed
the measurement of solvent-solute interactions over the entire
range from weak hydrogen bonding to heats of formation
of simple carbocations in superacids or of simple, reactive
carbanions in superbases.

The Beginnings of Physical Organic Chemistry as a Discipline

in the 1940s
Without doubt Louis Hammett’s classic monograph, Physical Organic
Chemistry (1), named the field. So authoritatively and elegantly was it written
that seventy four years after its publication in 1940 Hammett’s book maintains its
rigor and clarity. It contains essentially no errors although, of course, most of the
areas in Hammett’s original discussion have been enormously expanded and any
significant errors of interpretation or prediction on Hammett’s part would have
long since been revealed.
During the early 1940s most of the world’s chemical activity was dedicated
to the second World War in which chemistry played a major role. Some of the war
effort was applied to mechanistic studies of the reactions, such as nitration, most
vital to the conflict. When peace returned there was a cohort of organic chemists
eager to apply the principles in Hammett’s book to the study of electronic reaction
mechanisms as they had been developing since the 1920s, especially by Ingold’s
“English School” , and which they had been using to solve problems during the
war years.

© 2015 American Chemical Society

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Of particular significance in gaining recognition for physical organic
chemistry as a clearly identifiable discipline was the organization of the first
Conference on Reaction Mechanisms by Charles C. Price of Notre Dame and
Paul D. Bartlett of Harvard, held in September of 1946 at Notre Dame University.
As a vital means of communication in the Physical Organic-Reaction Mechanism
Community, this conference has continued steadily through the past seven
decades, meeting every two years without pause. During most of this period there
have also been regular meetings of the Physical Organic Gordon Conferences as
well as major participation in the Organic Symposia at ACS National Meetings
and numerous international symposia.
A major contributor to the early growth of the field was the establishmentof
the James Flack Norris Award in Physical Organic Chemistry in 1963 by the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch003

Northeastern Section of ACS which celebrated the work of the leading physical
organic chemists of the time: Christopher Ingold (1965), Louis Hammett
(1966),Saul Winstein (1967), George Hammond (1968), Paul Bartlett (1969),
Frank Westheimer (1970).
It is significant that although Louis Hammett was a physical chemist,
and taught physical chemistry, the great majority of those who have identified
themselves as physical organic chemists have been organic chemists, that is, they
were trained in and teach organic chemistry. With a few notable exceptions,
relatively few academics whose backgrounds lie in physical chemistry, and teach
it, have identified themselves as physical organic chemists.
Perhaps the most important contributor to providing a balanced organic
and physical approach to the field was Paul Bartlett, who emphasized the role
of synthetic chemistry as a means to testing physical organic theories. (For an
example of this combined approach see Project 4 below involving the synthesis
of a strained molecule, o-di-t-butyl benzene, and measurement of its strain
energy). Bartlett was a student of James Bryant Conant (later president of Harvard
University) whose approach to organic chemical problems was that of physical
organic chemistry decades before publication of Louis Hammett’s book. No other
teacher of physical organic chemistry was as influential as Bartlett in terms of the
number of successful teachers who were produced in his laboratory at Harvard.
In the 1930s and 40s, it was by no means obvious to the average academic
organic chemist that the occasional sallies of their colleagues into chemical
kinetics or other traditionally physical disciplines would eventually lead to the
fundamental theoretical principles on which the teaching and practice of organic
chemistry would depend, even in introductory textbooks. In those times the
leading organic chemistry textbooks were devoted to describing synthetically
important reactions through the inter-conversion of important functional groups
. Little if any consideration was given to the electronic redistributions through
which reactants must pass on their way to products. There was little recognition
that such a detailed understanding of electron bonding could help to integrate and
rationalize organic chemistry and predict new and unexpected reactions.
The increasing attraction of organic chemists towards the elucidation of
organic reactions in terms of modern physics followed the well-precedented
strategy of reductionism, that is, the clarifying of complex systems in terms of
“simpler” ones . Elucidating biological systems by breaking them down into

50
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 chemical reactions is a classic example of reductionism as is the reduction of
molecular structures in terms of electrons: that is by explaining chemistry in
terms of physics. Note however reference (2).
Despite the value of understanding the wonderland of organic chemistry
through electronic mechanisms, the establishment of physical organic
chemistry was accompanied by the stresses which are almost inevitable as an
interdisciplinary field is introduced in academe. Intradisciplinary stresses arose
within the Organic Divisions of many Chemistry Departments. Often the full
professors held the power to discourage younger colleagues from their flirtations
with physical chemistry instead of by making a name for themselves (and
their department) by doing “real” organic chemistry. For example the Organic
Chemistry Division of the University of Illinois , was presided over by several
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch003

of the leading figures in American chemistry who were staunchly opposed to

electronic mechanism study as an intruder into the field. It was not until 1955
that Illinois hired a physical organic chemist as an Assistant Professor. Also
interdisciplinary stresses between the Organic and Physical Divisions of many
departments were deeply ingrained in the all too common parochialism of the day.
In the early 1940s while I was an undergraduate chemistry major at the
University of Pennsylvania, my research director was Allen Day who was so
impressed by the mechanistic insights of the British School that he wrote a
textbook for teaching the sophomore pre-med course based on the new principles.
Although his work was roundly dismissed by some of the elder Professoriat as
“a fad” I believe it was the first undergraduate text to use electronic mechanisms,
with curved arrows, as a rational basis for elucidating organic chemical reactions.
Also, John Miller, a physical chemist, began teaching a course based on Louis
Hammett’s book soon after it was published. It was my good fortune to do my
doctoral research under the joint direction of Day and Miller.
After several years in a small Philadelphia industrial firm and two years
teaching in a small liberal arts college I joined Paul Bartlett’s research group at
Harvard as a postdoctoral Research Fellow. After two years of total immersion in
the superb scientific environment of Bartlett’s laboratory, I obtained an Assistant
Professorship at the University of Pittsburgh and began years of study on the most
general and most important reaction in chemistry, proton transfer.
In the 1950s there was a dearth of data by which the relative strengths
of organic compounds as proton acceptors or donors was known. I set about
developing this data base building on work that I had begun at Harvard. A natural
extension of the study of organic compounds as acids and bases is the study of
their behavior as solvents. Considerable work had been done on various aspects
of the acid-base behavior of organic compounds, and corresponding solvent
effects, by physical chemists but relatively little from the viewpoint of organic
chemists looking for quantitative comparisons of the systems most useful to their
own work.
Against this background, I was especially gratified to receive the 1977 James
Flack Norris award: “for outstanding contributions to our understanding of the
profound role of solvation in organic chemistry”.
The period following World War II in the 1950s and ‘60s was a good one
for American science and physical organic chemistry benefitted enormously from

51
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 the new generation of organic chemists who were eager to unravel the details of
the classical organic reactions in terms of electronic reaction mechanisms. At the
same time rapid development of chemical instrumentation was taking place and,
of particular importance, the federal government began a program of investment
in academic scientific research through a system of financial support especially
through the National Science Foundation and the National Institute of Health.
Public support was strong through appreciation of the role that science had played
in winning WWII . The Cold War with Russia was beginning and American science
got a powerful boost when sputnik , the first man made orbital satellite, appeared
in the evening skies. The American public recognized at once that a competition
of great importance was taking place between American scientists and their Soviet
counterparts.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch003

The discovery on December 23, 1947 of semi-conductors in the Bell

Laboratories had an immediate impact on two areas from which my research
was to grow. Most immediately, the thermistor provided a remarkably delicate
sensor for detecting small temperature changes and became the essential element
in the simple solution calorimeters that my research group learned to make.
Of more general application was the use of transistors to replace vacuum tubes
in computers. Without semiconductors the development of computers to their
present ubiquitous applications would have been impossible. Significantly, the
IBM 360 which filled a large room and was one of the first workhorse institutional
computers had 25KB of memory while the tiny iPhone 4 in your pocket boasts
524,300 KB . The difference in capabilities is staggering
As my research group learned how to make simple solution calorimeters (3)
capable of measuring accurate, precise heats of reaction to the nearest hundredth
of a kilocalorie it was clear that we had a tool which would allow us to measure the
many interactions of strong acids with weak bases and of weak acids with strong
bases which had been our long-term interest, see for example Figure 1. The same
tool obviously could be applied to measuring solvent-solute interactions.
On Figure 1 note the tiny thermistor for detecting heat changes in the
calorimeter and the heaters used for injecting a measured amount of electrical heat
to calibrate each measurement of a heat of reaction. Both of these heat changes
were displayed for measurement on a strip chart recorder.
Our calorimeter gave us entrée to a virtually unoccupied niche in physical
organic chemistry. Most of the energy changes being studied at the time were
free energies in terms of rates or equilibrium constants. Our calorimeter provided
immediate access to the corresponding heat (enthalpy) and heat capacity changes
which are only obtainable from equilibrium constants by application of the van’t
Hoff equation to their temperature coefficients. This is a much more difficult
operation than the simple measurement of the heat change in a calorimeter.
Entropies of reaction were available immediately from the heat of reaction and the
Gibbs free energy i.e. the equilibrium constant. It was soon clear from our work
and that of many others that there was often an excellent correlation between the
heats of reaction for a series of compounds and their corresponding free energies
of reaction. Thus, if one measured the heats of reaction for a series of bases in a
strong acid, or a series of acids in a strong base , the scale of relative basicities

52
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 (or acidities) obtained was a very close approximation to the scale that would be
obtained from the equilibrium constants for protolysis.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch003

Figure 1. A Typical Solution Calorimeter developed by the Arnett group

Application of Solution Calorimetry to Four Physical Organic

Chemical Problems
Problem 1 - Complete Thermodynamic Analysis of the “Anomalous Basicity
Order” of Aliphatic Amines in Aqueous Solution (4, 5)
During the 1960s and ’70s an invaluable breakthrough for the study of proton
transfer and solvation studies arrived in the gas phase . Ion cyclotron resonance ,
high pressure mass spectroscopy and the “flowing afterglow” techniques provided
reliable values for the protolysis reaction in the gas phase from which the role of
solvent for the corresponding reactions has been removed. A complete description
of the new opportunities afforded by this combination of techniques is described in
(4). The quantitative relations between protolysis in the gas phase and in aqueous
53
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 solution allows calculation of the solvation terms through a cycle and the following
equation:

where P refers to the property: ΔG, ΔH, ΔS, g refers to the property in the gas
phase, i refers to the ionization process and l is the relevant property in solution and
s refers to solvation , the transfer from the gas phase to solution. Thus ΔHs(BH+)
means “the heat of solvation of the ‘onium ion from the gas phase to solution in
the liquid under discussion” and ΔHi(g)(B) refers to the heat of ionization of the
base B in the gas phase.
The first solvation problem that we attacked using the newfound ability to
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch003

study proton transfer in the gas phase was the curious basicity order for amines in
aqueous solution: ammonia < primary < secondary > tertiary (5) .
In the gas phase the order was quickly found to be: ammonia < primary <
secondary < tertiary as would be expected on the basis of simple inductive effects.
Obviously, the inverted order for ionization of the tertiary amines in solution
is a solvation phenomenon. Accordingly, a complete tabulation of all twelve
properties for ΔG, ΔH, ΔS implied by equation 1 was developed for several sets
of primary, secondary and tertiary amines and scrutinized for trends that might
localize the anomaly to free energies, enthalpies or entropies of solvation of the
amines or their ‘onium ions. It was concluded that “complete analysis does not
lead to a simple interpretation. The inverted order arises from slight difference
in the rates of changes of several thermodynamic properties in response to
progressive alkyl substitution”.
It had been exciting to think that some simple solvent-solute interaction was
clearly identifiable as responsible for the amine anomaly, but such was not the
case.

Problem 2 - Concerning the Ionic Nature of Alkyl Halide Solvolysis

Transition States

By the 1960s extensive kinetic studies of the substitution reactions of many

tertiary alkyl halides had provided overwhelming evidence for a carbocation
intermediate. Similar results for a number of bicyclic halides prompted deeply
nuanced investigations of their structure-reactivity relations. Several major
conferences featured vigorous debates between leaders in the physical organic
community about a special kind of “nonclassical” bonding which might lie at the
bottom of the observed behavior. During this period research in the laboratories
of Professor George Olah at the University of Southern California showed
how to generate stable solutions of the proposed ionic intermediates in various
superacidic media (magic acids) employing mixtures of SbFs with SO2 ClF and
SO2 F2 of great acidity thanks to anions of very low nucleophilicity. In such
solutions at low temperatures t-butyl cation and many others could be examined
at leisure by NMR and other means.
54
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 In our laboratory, Craig Petro adapted the calorimeter shown in Figure 1 to
measure heats of reaction of alkyl halides in magic acid at -80 to -100 deg.C
for nearly twenty primary and secondary alkyl and cycloalkyl halides in several
superacidic media at temperatures down to -100 deg.C (6). Agreement between the
results for different solvents and many control experiments implied that indeed the
halides had been dissociated to carbocations. These heats of ionization in magic
acid were compared with a half dozen gas phase heats of ionization and reference
(7) describes the correlation between the heats of ionization and the free energies
of activation for ethanolyisis of eleven tertiary and secondary chlorides. The slope
of the linear relationship was 0.88 with a coefficient of correlation of 0.987 if the
free energies of ethanolysis for the secondary halides were corrected suitably for
nucleophilic assistance from the solvent.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch003

Finally, a group of carbinols (mostly diaryl and triaryl) whose pKR+’s had
been measured by Hammett’s indicator method were subjected to ionization in
the calorimeter at -40deg.C. Plots of heat of reaction in magic acid versus pKR+
were linear with coefficients of correlation of 0.95 or better. Such plots allowed
interpolation to estimate the pKR+ for t-butyl, norbornyl and adamantyl carbinols
which could not have been determined by any other method. These correlations
of the heats of ionization in superacids with gas phase ionization, free energies of
activation for ethanolysis and pKR+ show that the same process is being measured
in all cases, i.e. conversion of the neutral precursor to an ionic type of product (or
intermediate).

Problem 3 - Dissection of the Bizarre Behavior of the Activation Parameters

for t-Butyl Chloride Solvolysis in Highly Aqueous Ethanol-Water Solutions

During the mid-1950s Winstein and Fainberg published their monumental

investigation of solvent effects on the solvolysis of various alkyl halides in pure
and mixed solvents at several temperatures to discover the effects of solvent
variation on the activation parameters (8).
One can imagine the Winstein laboratories set up to measure rate constants on
a factory production basis with row after row of temperature control baths! The
results did not follow any simple physical treatment and were often bizarre. See
for example the following plot of the three activation parameters for solvolysis of
t-butyl chloride in a series of aqueous ethanol
Note in articular in Figure 2 the painstaking effort that was made to study
a large series of solutions in the neighborhood of 0.8 mole fraction water where
the fantastic effects on the heats and entropies of activation take place but their
combined effect compensates each other so well that it scarcely affects the free
energy of activation. Winstein realized that any consideration of the effect of
solvent change on the solvolysis transition state could only be isolated if the
corresponding solvent effect on the initial state, that is on the free energy, heat,
and entropy of solution for t-butyl chloride itself were subtracted out. In fact it
might be possible that all of the weird behavior of the activation parameters was
due to solvent effects on the t-butyl chloride ground state.

55
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch003

Figure 2. Plot of ΔH≠, ΔF≠ and - T ΔS≠ vs. mole fraction water in aqueous-ethanol
solvents

Winstein obtained such data for the free energies of solution for t-butyl
chloride in some aqueous alcohol solutions, but Figure 2 shows that the more
interesting results are in the solvent effects on the heats and entropies of activation.
The corresponding heats and entropies of solution for the ground state of t-butyl
chloride were not accessible to Winstein but we could measure them easily using
our calorimeter. In fact some of the strangest behavior of the Winstein-Fainberg
data was seen in the isokinetic plot of ΔH≠ vs ΔS≠. Was this primarily due to the
changing solvation of the presumably salt-like character of the transition state, or
rather to a reflection of the changing heat and entropy of solvation of ground state
t-butyl chloride itself?
We obtained (9) appropriate heats and entropies of solution for t-butyl chloride
in the same series of aqueous ethanol solutions which Winstein and Fainberg had
studied and combined them with the free energies of solution that they had obtained
from published articles on solvent effects on the Henry’s Law constants. Our
combined data for the free energy, heat and entropy of activation, and of solution
for t-butyl chloride in seven of the aqueous ethanol system were tabulated. Finally,
the three thermodynamic properties for solvation of the transition state itself were
determined by subtraction of the ground state contribution from the three activation
parameters in each aqueous ethanol solution.
56
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 It turned out that virtually all of the strange behavior of the solvent effects on
activation parameters was attributable to varying solvation of the ground state and
a simple linear isokinetic plot for the behavior of the transition state was strikingly
similar to that for a variety of simple salts in aqueous methanol solutions. Once
again the overwhelming evidence indicated that the t-butyl chloride solvolysis
transition state is strikingly ion-like.

Project 4 - The Synthesis and Strain Energy in o-Di-t-butyl Benzene

As part of an extensive study of steric effects, Professor H.C. Brown and his
students investigated strain energies in ortho di-substituted aromatic compounds
The strain present in homomorphs of o-di-t-butylbenzene implied an estimated
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch003

strain (in o-di-t-butylbenzene) to be a minimum of 25 kcal/mole. and finally

Brown (10) submitted the challenging statement: “Strains of this magnitude
suggest that the preparation of such homomorphs should be a difficult matter.”
Interest in Brown’s studies were so great that at once o-di-t-butyl benzene became
one of the “molecules of the year” whose synthesis would assure a path to
recognition for aspiring young organic chemists . If o-di-t-butyl benzene could
be made it would provide a classical example of a theoretical problem which
could be tested by the synthesis of a compound presenting a structural feature
which theory predicted would be limiting. Fortunately for the writer, who was
at the time an Assistant Professor at the University of Pittsburgh, the reactions
of acetylenes with metal carbonyls was being heavily investigated nearby at the
U.S.Bureau of Mines. Disubstituted acetylenes were shown to give stable adducts
with dicobaIt octacarbonyl (11)

Figure 3. The presumed structure of the adduct formed from di-t-butyl acetylene
and dicobalt octacarbonyl based on reference (11)

We were encouraged to believe that if such an adduct could be made between

Co2(CO)8 and di-t-butyl acetylene (Figure 3) that treatment of the adduct with
gaseous acetylene under pressure might yield o-di-t-butyl benzene. Indeed, in the
hands of Mike Strem the target product was obtained (12, 13) and the first question
57
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 raised by H.C. Brown was answered; the existence theorem was proved, the strain
was not too great to preclude the molecule’s existence
Thanks to our reaction calorimeter we were well positioned to answer
Professor Brown’s second prediction: if o-di-t-butyl benzene could be made,
the strain energy from forcing the two bulky groups closely together should be
about about 25 kcal/mole. This problem was attacked at once using a strongly
acidic AIBr3 solution in CS2 at 30deg.C to convert o-di-t-butyl benzene into its
meta and para isomers (14). Many control experiments were done to demonstrate
that conversion of the strained ortho isomer to its presumably unstrained isomers
was entirely responsible for the large sudden release of heat when reaction was
initiated (Figure 4). Many replications anchored the heat of isomerization, and by
this reasoning the strain energy to be 22.3±0.5 kcal/mole.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch003

Figure 4. Schematic for preparation of o-di-t-butyl benzene and measurement of

its strain energy.

Mike Strem was also able to make 1,2,4-tri-t-butyl benzene, but all attempts
to make l,2,3,4-tetra-t-butylbenzene or the ultimately crowded l,2,3,4,5,6-hexa
t-butylbenzene failed. The ortho disubstituted benzenes were examined by a
variety of spectral and other methods for any evidence that the benzene ring was
being seriously warped into a “nonaromatic” benzene. Nothing more of particular
interest was found. High reactivity to relieve the strain from crowding two bullky
groups together was the only unusual property worth noting.

58
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 References
1. Hammett, L. P. Physical Organic Chemistry; McGraw Hill: New York, New
York, 1940.
2. Reductionism cannot be “run backwards”. The vast complex world of
chemistry is not predicable from physics by emergence any more than the
details of biology are predictable from chemistry.
3. Arnett, E. M.; Petro, C. J. Am. Chem. Soc. 1978, 100, 5402–5407.
4. Arnett, E, M. Acc. Chem. Res. 1973, 6, 404–409.
5. Arnett, E. M.; Jones, F. M., III; Taagepera, M.; Henderson, W. G.;
Beauchamp, J. L.; Holtz, D.; Taft, R. W. J. Am. Chem. Soc. 1972, 94,
4724–4726.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch003

6. Arnett, E. M.; Petro, C. J.AmChem.Soc. 1978, 100, 5408–5416.

7. Arnett, E. M.; Petro, C.; von, R; Schleyer, P. J. Am. Chem. Soc. 1979, 101,
522–526.
8. Winstein, S.; Fainberg, A. H. J. Am. Chem. Soc. 1957, 79, 5937–5950.
9. Arnett, E. M.; Bentrude, W. G.; Duggleby, P. McC. J. Am .Chem .Soc. 1965,
87, 2048–2050.
10. Brown, H. C.; Nelson, K. L J. Am. Chem. Soc. 1953, 75, 24–28.
11. Greenfield, H.; Sternberg, H. W.; Friedel, R. A.; Wotiz, J. H.; Markby, R.;
Wender, I. J. Am. Chem. Soc. 1956, 78, 120–124.
12. Arnett, E. M.; Strem, M. E. Chem. Ind. 1961, 2008–2009.
13. Following receipt of his PhD, Mike Strem decided to start a small business
aimed towards being a reliable source of organometallic catalysis. Since then
Strem Chemicals has gained an eminent position in American chemistry.
14. Arnett, E. M.; Sanda, J. C.; Bollinger, J. M.; Barber, M. J. Am. Chem. Soc.
1967, 89, 5389–5400.

59
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Chapter 4

A Life in Physical Organic Chemistry

Ronald Breslow*

Department of Chemistry, Columbia University, New York, New York 10027

Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

*E-mail: [email protected]

My life in science has pretty much focused on physical

organic chemistry. The topics have included novel aromatic
molecules--the cyclopropenyl cations and cyclopropenone--and
also antiaromaticity in the 4 pi-electron species including
cyclopropenyl anions, cyclobutadiene, and cyclopentadienyl
cation. Electrochemistry was an important tool, and esr
spectroscopy to detect triplet ground states in symmetrical
antiaromatics. We showed that aromaticity added resistance
to molecular wires, so non-aromatic wires have higher
conductivities. We showed the mechanism by which thiamine
pyrophosphate catalyzes important biochemical processes with
an intermediate nucleophilic carbene. We devised artificial
enzymes and explored hydrophobicity; this furnished a tool to
discover the geometries of transition states. We determined
some enzyme mechanisms, and experimentally showed a
scheme for the way that prebiotic chemistry could have formed
L-amino acids and D-sugars. We devised a group of novel
catalytic polymers, and also a new anticancer compound with
a novel mechanism.

The Importance of Physical Organic Chemistry

Physical organic chemistry combines the most important aspects of both
physical chemistry and organic chemistry. It asks about the properties of organic
compounds, and it also asks how the transformations of organic compounds occur
in chemical reactions, in detail. Organic chemistry itself is concerned with the
likely properties of new molecules and how to make them, and of course also with
the properties of naturally occuring compounds and their chemical structures. The
tools of physical organic chemistry tell us which compounds to make for desired

© 2015 American Chemical Society

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 properties, including biological and electronic properties. The tools created by
physical organic chemists also help synthetic chemists create new reactions and
strategies to solve the synthesis problems, and the tools and procedures needed to
deduce the structures of both the new compounds and also those found in nature.
Physical organic tools also help us understand the ways in which natural enzymes
perform their syntheses.

My Studies in Physical Organic Chemistry

As I will describe, my research programs have ranged from experiments to
help understand how some important enzymes work, to experiments to see how
their reaction mechanisms can be imitated in synthetic enzyme mimics, to the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

creation of new molecules that modulate biogical differentiation and showing

how they work, to showing how the phenomenon of hydrophobicity can be
used in artificial enzymes but also how it can be used to deduce the geometries
of transition state, to experiments to create and understand the properties of
new aromatic systems, to understand and demonstrate the new phenomenon of
antiaromaticity, to using electrochemistry to create and measure the properties
of antiaromatic systems, and to understand how the electrical conductivities of
molecular wires is related to the aromaticity of their components.
The results of this work are described in over 400 publications, too many
for a sensible bibliography for this chapter. Thus I will generally cite the first
publications in a given area and then either a published review or a recent
publication with its own bibliography that has generous citations of the other
publications. This plan shortchanges my coworkers who did all the intervening
work, but their names can be found in the citations that are in the most recent
publications.

Aromaticity
I have been interested in aromaticity since I was 12 years old. I had the
good fortune that a close family friend was Max Tishler, director of chemistry
at Merck. He gave me a copy of Conant and Blatt, Organic Chemistry, when I
was seven, and in it I learned about the aromaticity of benzene. I decided that an
analog with silicon should also have the special stability of aromatic compounds.
I set out to synthesize it--still a challenge for chemistry today--but my efforts
helped me become a finalist in the Westinghouse contest for high school students,
where I explained why hexasilylbenzene would be an interesting compound. This
has to some extent been my style: create an unknown compound predicted to
have interesting properties, then study their properties including their chemical
reactivities and even their biological properties
I joined the department of chemistry of Columbia University in 1956, as
a lowly instructor. The department was very strong, with Louis Hammett (not
retired) and Cheves Walling in physical organic chemistry, William Reinmuth in
electrochemistry, and Gilbert Stork in synthesis. When I started my independent
career, aromaticity in single rings was known to involve 6 delocalized pi electrons,
62
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 as in benzene, or cyclopentadienyl anion, or cycloheptatrienyl cation. However,
the theory of aromaticity indicated that there would be other magic numbers,
with 4n + 2 pi electrons and n an integer. The simplest aromatic system would
be the cyclopropenyl cation, with n = 0, so we set out to synthesize it and see
its properties. This is the part of physical organic chemistry relating properties
to structures, including those of non-natural synthetic species. We were able to
synthesize the 1,2,3-triphenylcyclopropenyl cation in 1957 (1), and saw that it was
reasonably stable in spite of its tremendous ring strain that would favor addition
to the cation. In 1967 we were able to synthesize the unsubstituted cyclopropenyl
cation 1 (Figure 1), again surprisingly stable considering its ring strain (2).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

Figure 1. Cyclopropenyl cation and cyclopropenone.

In general carbonyl groups can replace the cationic carbons in such species,
and still share the aromaticity. For example, cycloheptatrienyl ketone, tropone,
is also an aromatic 6-pi-electron nucleus. Thus we synthesized substituted
cyclopropenones (3) and the unsubstituted ketone itself 2 (Figure 1) (4), and saw
that it was strongly stabilized by its 2 pi-electrons and a carbonyl dipole. In fact,
in water it existed as the ketone while cyclopropanone, without the double bond,
was completely hydrated as a geminal diol to relieve strain. Subsequently others
(5) have shown that with amino group substituents and an imine in place of the
carbonyl group the cyclopropene system is very stable and interesting as a basic
unit in novel synthetic procedures. A review of this work on aromaticity is part of
a recent publication on the role of aromaticity in molecular electrical conductivity
(6).

Antiaromaticity
While it was clear that aromaticity in a single ring required 4n + 2 delocalized
pi electrons, it was not clear when I started my independent career what the
properties would be of monocycles with 4n pi electrons. It was clear that they
would not be aromatic, but would they simply be non-aromatic or could they be
worse than that? Cyclobutadiene was clearly very reactive, and defied isolation,
but could that simply reflect its strain and not a specific destabilization by
the delocalization of 4 pi electrons? We set out to examine this question. As
our several review articles describe (7–11) we concentrated on three species
with 4 pi electrons: the cyclopropenyl anion 3, the cyclobutadiene ring 4,
and the cyclopentadienyl cation ring 5 (Figure 2). We used kinetic studies,
and also electrochemistry, to explore these systems. Rate studies showed that
63
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 cyclopropenyl anion derivatives were very slow to form compared with analogs
without the double bond, even though the double bond added additional places to
delocalize charge into stabilizing substituents (6).

Figure 2. Parent antiaromatic compounds.

Electrochemistry gave us an important way to generate cyclopropenyl anion

Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

derivatives, and the parent anion, by adding two electrons to the corresponding
stable cations. With modern techniques we could get reversible reduction
potentials for the process, giving us solid information about the thermodynamic
energy of the anions and thus of the formal pKa’s of the cyclopropenes (12).
We also used these electrochemical methods to determine the pKa’s of simple
hydrocarbons, getting a good estimate of the pKa of saturated hydrocarbons such
as methane for the first time (13). Needless to say, with pKa’s in the 60’s for the
cyclopropenes of saturated hydrocarbons there was no way to generate the anions
by proton removal with base. We also extended our electrochemical methods
to generate and measure the energy content of anions to one-electron reduction
processes of cations, or one electron oxidations of anions, that could determine
the bond dissociation energies of hydrocarbons to form carbon radicals (11, 14).
This technique was then adopted by others, not always with full attribution.
The bottom line of our cyclopropenyl anion studies was clear evidence that it
was antiaromatic, especially destabilized by the delocalization of its 4 pi electrons.
I coined this word, and Michael Dewar may also have had it in mind as a possible
property.
We also used electrochemistry to convert a molecule with only partial
cyclobutadiene character (6) to one (7) where the cyclobutadiene component was
greatly increased (Figure 3), and saw that this required a higher (more positive)
oxidation potential than did a suitable reference compound (15). Since there was
no increase in angle strain in this case, it indicated that cyclobutadiene was also
antiaromatic, although the extra energy this added to the molecule was not clearly
defined. Again this is discussed in our reviews (10).

Figure 3. Increasing the antiaromatic cyclobutadiene character.

64
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The evidence of antiaromaticity was particularly clear in the cyclopentadienyl
cation. For one thing, silver cation could not generate it from cyclopentadienyl
chloride while cyclopentenyl chloride readily generated the cyclopentenyl cation
under the same conditions (10). As we will descibe momentarily, cyclopentadienyl
cation could be generated from the bromide with powerful Lewis acids, and its
structure was very special.
Theory predicted that 4-pi-electron molecules could exist as triplet states,
with two unpaired electrons, if they had regular symmetric structures (three-fold,
four-fold, five-fold, or later six-fold symmetries). Cyclopropenyl anion was not
completely flat but had one proton out of the plane to diminish its antiaromaticity,
so it did not meet the requirement of full symmetry and should not be a triplet.
Cyclobutadiene is not a square but an elongated rectangle, again to diminish
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

its antiaromaticity, so it should not be a triplet species. However, we thought

that symmetrical derivatives of cyclopentadienyl cation could have five-fold
symmetric structures, and we knew how to generate them.
We synthesized pentaphenylcyclopentadienyl cation, a reasonably stable
symmetric species, and saw an ESR spectrum indicating the presence of a triplet
(Figure 4) (16). However, when we lowered the temperature the signal weakened,
indicating that the triplet was a low-lying excited state of the molecule. Some
distortion might explain why it was not the ground state. We then prepared
pentachlorocyclopentadienyl cation from hexachlorocyclopentadiene with SnF5,
and saw by temperature studies that it was a ground-state triplet (17). Then with
Martin Saunders and Mike McBride of Yale we treated 5-bromocyclopentadiene
with SbF5 to create the unsubstituted cyclopentadienyl cation, and it was also a
ground-state triplet (18). We wrote several reviews of this triplet area (19–21),
including its possible use in organic ferromagnets. In later work we also created
the dication of a fused ring hexaaminobenzene and saw that it was also a potential
ground-state triplet (22) with possible use in organic ferromagnets. The six-fold
symmetry with 4-pi-electrons was what did it.

Figure 4. Triplet ground states of the symmetric cations.

It is important to note that the formation of triplets does not explain why
the molecules are antiaromatic, of higher than normal energy. Unpairing the two
electrons in triplets lowers the energies of the species, by the explanation behind
Hundt’s Rule.

65
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The Electrical Conductivity of Aromatic and Antiaromatic
Molecules
I became the scientific co-director of the Columbia University Nanoscience
initiative, and set out to create the two important components of nanoscale
molecular electronics--devices such as molecular transistors, and molecular-scale
wires to conduct electricity and connect the devices. The conductivities of
various molecules and corresponding wires were measured in collaboration with
Professor Latha Venkataraman with a gold break-junction device. Briefly, a gold
wire was stretched to breaking in a solution of the molecule or wire with contacts
such as amino groups or thioethers, and the conducting species spontaneously
attached across the break in the wire to permit the conductance determination.
We made two significant discoveries in this area. In one, we discovered that
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

with trimethyltin groups attached to the molecular wires the gold broken wires
inserted between the tin groups and the molecules, eliminating the resistances
of the amino groups or thioether groups and yielding high conductivities (23,
24). The other finding was that the aromaticity of such common molecular wires
as polythiophenes introduced extra resistance, and that non-aromatic molecules
were the best conductors (25). In work underway, we are making antiaromatic
molecular wires that we predict to be even better conductors.

The Mechanism of Catalysis by Thiamine Pyrophosphate in

Biochemistry
At the same time that the early aromaticity work was underway I also
published two papers in 1957 on the mechanism of catalysis by thiamine
pyrophosphate and related thiazolium cations (26, 27). In this work I showed that
the proton on a thiazolium cation ring rapidly exchanges with D2O using base
catalysis. I described the reason for this in a general full paper in 1958 that pointed
out that the thiazolium zwitterion from loss of the proton on C-2 has a carbene
resonance form, as shown in Figure 5 (28). I pointed out that such ion/carbene
resonance is quite common. The carbene has the interesting feature that the
resonance makes it a nucleophile while electron-deficient carbenes without such
resonance are electrophiles. Thus the field of "nucleophilic carbenes" was born,
and many examples are now in use as metal ligands and catalysts by others.

Figure 5. Thiazolium zwitterions and their carbene resonance forms

This mechanism for thiamine catalysis is widely accepted for both the
chemical catalyses and the biochemical ones, but more recently a paper in
Tetrahedron Letters asserted that it did not fit the observed and plotted data for
66
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 the chemical catalytic system (29). We showed that the data had been incorrectly
plotted, and that with a correct plot the thiamine mechanism described above is
fully supported (30).

Biomimetic Chemistry and Enzyme Mimics

The work to understand the mechanism by which Nature uses thiamine
pyrophosphate was only an early example of an overwhelming interest of mine:
can we use our insights into how enzymes work to create new chemistry with
the same general high selectivity and catalytic effectiveness typical of enzyme
reactions? If there is one general principle that differs between synthetic chemistry
and enzyme-catalyzed chemistry, it is that synthetic chemistry is dominated by
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

selective reactivity, while enzymes use selective geometry. Of course enzymes

still have to carry out favorable reactions, but in biochemical oxidations of
steroids, for example, the oxidation chemistry is not special, but where it occurs
does not reflect the intrinsic reactivity of various positions on the steroids, it
reflects the geometries of the enzyme/substrate complexes. I set out to use such
geometric control as a general principle.
Of course other chemists were aware of this idea, but they had generally used
the geometry of intramolecular reactions to carry out selective functionalizations.
A hydroxyl group of the steroid would be functionalized so it could oxidize a
nearby saturated carbon. for instance. I wanted the catalysts or reagents to impose
the geometry, so positions could be oxidized in steroids that were remote from any
functional groups. I called the results "remote oxidation." We were inventing new
reactions guided by the mechanistic ideas of physical organic chemistry, and some
of the detailed mechanisms of photochemical reactions and free radical processes.

Reaction of Steroids with Enzyme Mimics

The best place to start was with steroids. They are rigid with well-defined
geometries, and there are some syntheses that Nature performs, such as inserting
an oxygen into a saturated carbon in the middle of the molecule to make
corticosteroids, far from other functional groups, that would be of practical
interest since this was not possible with existing chemistry. We started with
photochemistry. We attached benzophenone units to steroids with an ester chain
and then photolyzed them. The geometry directed the excited state carbonyl
group to particular steroid hydrogens, and with this result we could introduce
oxygen atoms and double bonds into the steroid, several rings away from the
attachment point because of the length and rigidity of the benzophenone system
and of the steroid skeleton (31, 32).
We then turned to directed steroid chlorination, since chlorine atoms can easily
remove hydrogens from saturated carbons, and unselectively--the trick was to steer
them to a desired spot. We found that a general approach works. We attached a
long template to the steroid on carbon 3 on the alpha face at one end and performed
a free radical chlorination in which a chlorine atom bound to the template, often by
a three-electron weak bond, removed the hydrogen that it could reach. The binding
67
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 of the chlorine made it less reactive, so more selective for tertiary hydrogens rather
than secondary ones, and in steroids tertiary hydrogens on the alpha face have
well-defined distances from C-3. Varying the length of the template let us select
any one of those tertiary hydrogens and insert a chlorine in its place (33–36).
Of course an enzyme would not attach something to a substrate to steer
a reaction, it would simply bind to the substrate and hold its functionalizing
component where it could reach within the complex. In more recent work we have
figured out how to do that. We used cyclodextrins, doughnut shaped molecules
made up of glucose units in a ring, to bind hydrocarbon species into the hole in the
doughnut in water solution, using hydrophobicity as the driving force. With two
cyclodextrins linked to a metalloporphyrin and steroids with some phenyl rings at
the two ends of the molecule we saw spontaneous binding of the steroid into the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

cyclodextrin dimer in water. Then the metal ion, iron or manganese, picked up
an oxgen atom from hydrogen peroxide and the oxygen atom then inserted into a
nearby C-H bond, leaving a hydroxyl group selectively on a single position in the
steroid (37–39). This has a lot of analogy to the biological mechanism by which
steroids are hydroxylated.

Cyclodextrins in Enzyme Mimics

Binding substrates into the cavities of cyclodextrins can steer and promote
many other reactions in water solution. As our earliest example, we showed that
the chlorination of anisole (methoxybenzene) by HOCl is selectively directed
to the para position with a cyclodextrin present, while without it both the ortho
and para positions are chlorinated (40–42). The simplest idea is that the ortho
positions are blocked by anisole binding into the cyclodextrin cavity, but our
mechanistic studies showed that there was an additional factor: the chlorine
atom was catalytically delivered to the bound anisole by a hydroxyl group of the
cyclodextrin. This was an extremely simple analog of an enzyme, with binding
and selective catalysis by the cyclodextrin.
We have used cyclodextrin for binding substrates where there is a catalyst
attached to the cyclodextrin. For instance, we have attached a pyridoxamine
species to a cyclodextrin and found that it would bind an α-keto acid in water
and then synthesize an α-amino acid (43). Perhaps most interesting was the case
where we attached imidazole groups to a cyclodextrin and then used them as
catalysts for the hydrolysis of phosphate esters that bound into the cyclodextrin
cavity. With two attached imidazoles we were imitating the enzyme ribonuclease
in which two imidazoles of histidine units cooperate in cleaving RNA (44). We
could attach two imidazoles on neighboring glucose units of the cyclodextrin or
on units further apart. The classical accepted mechanism for the hydrolysis of
RNA by the enzyme needs the imidazoles far apart, but we saw that the isomers
with imidazoles on neighboring glucose units were the best (45). This indicated
a new mechanism for the hydrolysis, with a phosphorane intermediate, that we
proposed would also be used by the enzyme itself with RNA. Our subsequent
work showed that the enzyme in fact used the new mechanism (46).
68
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 This is a new tool for determining reaction mechanisms in special cases. When
there is bifunctional catalysis, with the cooperation of two catalytic species, fixing
their relative geometries can help decide between different mechanisms.

Catalysis by Polymers
While many of our studies used single molecules as enzyme mimics, in
Nature the catalysts are polymers, polypeptides. Thus we have explored the
use of various polymers as artificial enzymes. In one series of studies we have
used polyoxazolines, which can produce polyaziridine derivatives with high
stereoselectivity (47, 48). More recently we have created and studied a new
class of polymers derived by reducing the carbonyl groups of polypeptides
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

to methylene groups (47, 49) . These are again polyaziridines, but with the
sidechains and stereochemistry of their parent polypeptides, and chemical and
biological stability in contrast with the polypeptides themselves. They are also of
completely designed and homogenous length. We introduced this novel class of
polymers (we call them DOPPs, deoxygenated polypeptides) in two papers (47,
48) and recently described using them in mimics of ribonuclease (49). They have
a brilliant unexplored future.

Determining Transition State Geometries Using Hydrophobic

Effects
As described above, we have used the hydrophobic effect in enzyme mimics
to let substrates bind to catalysts in water. We also saw the hydrophobic effect
in simple reactions in water. For example, when we carried out Diels-Alder
additions, such as the Diels-Alder dimerization of cyclopentadiene, we saw
very high reaction rates with saturated solutions in water, compared with the
results in methanol or ethanol (50). One way to describe this is to say that
water somehow pushes the two molecules together, but another point is that the
very low concentration of cyclopentadiene in water has the same free energy as
pure cyclopentadiene, since they would be in equilibrium. The free energy of
cyclopentadiene in water at saturation is much higher than its free energy would
be with the same low concentration of cyclopentadiene in ethanol, for instance.
Hydrophobic effects can be modified with additions to the water. For
instance, with added ethanol the water can dissolve more cyclopentadiene before
saturation. Previous studies have shown that water solubilities of non-polar
molecules can be decreased with added lithium chloride, which increases the
hydrophobic effect, but solubilities are increased with guanidinium chloride
which decreases the hydrophobic effect. Thus in the Diels-Alder reaction of
cyclopentadiene with butenone in water we saw that the rate was increased with
the addition of lithium chloride but slowed with the addition of guanidinium
chloride (51). We examined many different related cases. Then we reasoned that
the rate effects could reflect the difference in exposed hydrophobic regions of
the transition state relative to that of the substrates. We showed this was true by
seeing that it accounted for the observed rate effects in terms of how much of
69
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 the reactant non-polar surface was "hidden" in the transition state TS (52). For
example, in the Diels-Alder dimerization of cyclopentadiene almost one face of
each reactant is hidden from solvent in the TS, but in the final product they have
partially opened up. With extensive studies we showed that indeed this was a
powerful new tool to determine something important about TS’s: not just their
energies, as rates reveal, but also their geometries.

Insights into the Mechanisms of Some Enzyme Reactions

Although we concentrated on artificial enzymes and their mechanisms,
we did investigate the mechanisms of two natural enzymes. The enzyme
carboxypeptidase catalyzes the cleavage of peptides, and the X-ray structure left
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

an ambiguity. It was clear that a zinc cation bound to the carbonyl oxygens of the
peptides while a carboxylate of the enzyme then promoted the cleavage, but it was
not clear whether it attacked the peptide carbonyl directly, forming an anhydride
intermediate that then hydrolyzed, or whether it delivered a water molecule to the
peptide carbonyl group without forming an intermediate anhydride. Analogs for
both mechanisms were known, but we were concerned that Ginodman claimed
that an anhydride was clearly formed by evidence that seemed to us ambiguous
(53). Thus we performed a series of experiments on the enzyme that clearly
showed that the Ginodman finding had a different cause, and that the reaction
was a straight hydrolysis with the enzyme carboxylate acting as a simple base,
not a nucleophile (54, 55). Since many other peptide cleavage reactions go via
covalent intermediates, but not this one, this was an important finding.
The other enzyme whose mechanism was of interest was ribonuclease,
which hydrolyzes RNA. We had done extensive work on mimics of this enzyme,
including a very recent study of a novel class of catalysts, and had concluded
that the enzyme mimics formed a phosphorane interrmediate, a 5-coordinate
phosphorus, in the course of the process. In other words, hydrolyzing the
phosphate ester in that case went through an intermediate, the opposite of our
conclusion in the carboxypeptidase case. However, the mechanism commonly
invoked for ribonuclease, in all the biochemistry textbooks, went with direct
cleavage of the phosphate, without a 5-coordinate phosphorus intermediate.
Thus we investigated it further. Our experiments with ribonuclease and various
substrates made it clear that it proceeds via the phosphorane intermediate (56),
and most modern biochemistry books now have this as the mechanism.

The Formose Reaction and Prebiotic Chemistry

In the course of reading I came across the formose reaction, in which
formaldehyde reacts in calcium hydroxide solution to form various sugars,
although not with any defined stereochemistry. There was no proposed reasonable
mechanism for it--formaldehyde can’t lose its proton to make a reactive anion--so
we examined the reaction. It had two phases. In the first the system just sat
there, with nothing happening, and then it started to make sugars with increasing
velocity. Thus it was autocatalytic.
70
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 I proposed a mechanism in which two formaldehydes first condensed to
glycolaldehyde, in a slow reaction that may well have involved free radicals and
maybe even cosmic rays to initiate the reaction. However, once a single molecule
of glycolaldehyde was formed it could enolize and react with formaldehyde to
form racemic glyceraldehyde. This then could isomerize to dihydroxyacetone,
in analogy to a well-known biochemical triose isomerase process, and this could
enolize and add a formaldehyde to form a four-carbon keto sugar mixture. This
could isomerize to the aldehyde sugar, which could undergo a reverse aldol
reaction to form two glycolaldehydes. Then four, then eight, etc. My proposed
mechanism (57), consistent with our experiments, was widely accepted--until
Steve Benner reinvestigated it.
Steve performed the reaction in D2O, and saw the various intermediates
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

we had proposed, and that many others had seen subsequently, but they had no
deuterium on the carbons (58)!! Thus we have recently reinvestigated it, and found
the explanation that generally validates the process we had originally proposed,
with some important differences (59). We had suggested that glyceraldehyde
isomerizes to dihydroxyacetone through the initial enol of glyceraldehyde, and
this should incorporate deuterium in the dihydroxyacetone from D2O. However,
we have now shown that the isomerization occurs not by enolization but by a
hydride shift, by showing that 2-deuteroglyceraldehyde goes to dihydroxyacetone
under formose conditions with migration of the deuterium, not its loss as would
happen with enolization (59). We have recently studied this reaction in detail,
and have evidence that it proceeds with quantum mechnical tunneling from a
thermally excited state (60). Then of course the dihydroxyacetone must enolize,
but it is irreversible since the formaldehyde immediately traps the enol. The
isomerization from ketosugar to aldosugar at the end also involves hydride shifts
and the reverse aldol reaction liberates an enediol that is immediately trapped by
formaldehyde. The details are in our paper.

The Creation of Selective Stereochemistry in Amino Acids and

Sugars on Prebiotic Earth
Of course the formose reaction could have occurred on prebiotic earth, but it
does not select for the stereochemistry we now know is present and necessary in
sugars. We came upon the findings that some meteorites that have landed on Earth
in recent years carry with them several amino acids with methyl groups instead of
alpha protons, so they cannot racemize (normal amino acids are also seen on the
meteorites, but racemic unless they have a second chiral center.) The α-methyl
amino acids were not racemic, but had small excesses of the L isomer (counting
the methyl group as if it were a hydrogen atom) (61). So we showed that they could
induce the formation of normal L-amino acids under sensible prebiotic conditions
(62), and we (63), and others (64, 65), saw how small enantioexcesses could be
amplified to high e.e’s under sensible prebiotic conditions.
We then examined the formation of sugars, in particular of D-glyceraldehyde,
since it is the simplest sugar with chirality. All other sugars in the D series can
form by building on the glyceraldehyde aldehyde group. We saw that under mild
71
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 acidic conditions the L-amino acids catalyzed the formation of small excesses
of D-glyceraldehyde (66, 67), and that this could also be amplified to high
enantioexcesses under sensible prebiotic conditions. In current work we are
investigating the formation of other important sugars including the D-ribose of
RNA and the D-deoxyribose of DNA.
In one earlier study, we examined why the current structure of DNA, 2-deoxy
with a 3-5 phosphate linkage, is better than the other possibility: 3-deoxy with
a 2-5 linkage (68). This was in the spirit of earlier work exploring why thiamine
pyrophosphate was particularly suited for its catalytic role in modern biochemistry,
all part of what we call Nature’s Chemistry Preferences. Our work with the isomer
made it clear that the current structure has significant advantages.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

A Novel Mechanism To Combat Cancer

About 30 years ago I was contacted by a friend who had become the head
of the Sloan-Kettering Insitute fror Cancer Research, Dr. Paul Marks. He
had learned that dimethylsulfoxide (DMSO) was able to induce certain cancer
cells to differentiate into adult erythrocytes (red blood cells), and he suggested
a collaboration to understand the process and perhaps to then create a more
potent compound than DMSO. This was the normal physical organic attempt to
understand the relationship between structures and properties, the properties being
biological. I set out to see whether our abilities in physical organic chemistry
could guide us to better compounds that might, with luck, prove to be useful in
the war against cancer (reviewed in (69)).
At first we established that DMSO could be replaced by simple amides,
and then we decided to use the chemists’ version of how to increase potency:
double the compound up so the chelate effect comes into play. Starting with
acetamide as the DMSO surrogate, we linked two acetamides with various
polymethylene chains and found that the magic number for best improvement
was six, in hexamethylene-bis-acetamide. Actually, that was my first guess. The
compound was safe and effective in animal studies, and even was used in some
cancer patients, but it was not potent enough.
I then used chemical reasoning. If an acetamide group was binding to a
biological receptor it was either binding to a metal ion or making hydrogen
bonds, and in any case an N-hydroxyamide could be better. It is a better metal
ion ligand (my guess of what was involved) and could make more hydrogen
bonds. Thus we made suberoyl bis-hydroxamic acid (SBHA), and it was a better
inducer of differentiation in the cancer cells. However, it was still not quite potent
enough. I then decided that it was probably binding to a metal ion, and perhaps
one end needed to be a hydroxamic acid for that role but the other end probably
contributed something else rather than curl back to the metal ion. We tried
hydrophobic binding with a phenyl group, and it was better than SBHA. We called
the new compound SAHA (Figure 6), suberoylanilide hydroxamic acid, and it
was very effective. By then we knew that it was indeed binding to Zn2+, and that
the target was a histone deacetylase with the Zn2+ binding to the carbonyl oxygen
of the acetyl group. Our drug SAHA bound to this catalytic site and blocked the
72
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 hydrolysis of acetylated histone, and indeed its structure was exactly what one
would design as an active-site directed inhibitor. But it was not derived that way.

Figure 6. Suberoylanilide hydroxamic acid

Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

Since then the compound has had extensive testing, and was approved
for human use against cancer in the U.S. Then approval followed in Canada,
and Japan, and the compound SAHA has been in human use for many years.
Others have created related compounds, but SAHA was the pioneer. It led to
the wide appreciation of HDAC inhibition as a mode of cancer treatment, and
of the mechanism involving modulation of post-translational regulation of gene
transcription. The field has even received a name, epigenetics, referring to the
role that regulation of acetylation levels, and other protein modifications after
translation, can have in gene expression.

Predicting the Future of Physical Organic Chemistry

Some time ago I was asked to write a brief prediction of the future of physical
organic chemistry (70). After describing the wonderful strengths of our field--it
combines the strength of synthetic organic chemistry with its unique ability
to design and create novel molecules, with the serious scientific aim to fully
understand molecular properties in chemical reactions in full detail--I described
"Some Goals for the Future." Here I discussed what abilities our field still needs,
and what important questions remain to be answered. The list is extensive, but
ends with the following statement: "These are some of the challenges, and likely
advances, that I can predict, but there is one more prediction I can make with
certainty. Smart physical organic chemists will come up with advances and
inventions that are anticipated by none of the chapters in this volume, including
this one, and we will all say: why didn’t we think of that?"
Physical Organic Chemistry has a wonderful history, putting understanding
into otherwise baffling organic processes and properties. It also has a great future,
giving us insights into the relationship of structure and properties of species as
diverse as molecular electronic materials and natural enzymes. It will also guide us
in designing new molecules and new chemical processes. It will increasingly use
modern computational methods to help in these adventures. I look forward eagerly
to participate in these efforts, and to see with admiration the progress being made
by the many other students of that most scientific and creative aspect of organic
and physical chemistry--Physical Organic Chemistry.
73
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 References
1. Breslow, R. J. Am. Chem. Soc. 1958, 80, 3719–3726.
2. Breslow, R.; Groves, J. T.; Ryan, G. J. Am. Chem. Soc. 1967, 89, 5048.
3. Breslow, R.; Altman, L. J.; Krebs, A.; Mohacsi, E.; Murata, I.; Peterson, R.
A.; Posner, J. J. Am. Chem. Soc. 1965, 87, 1326–1331.
4. Breslow, R.; Ryan, G. J. Am. Chem. Soc. 1967, 89, 3073.
5. Bandar, J. S.; Lambert, T. H. J. Am. Chem. Soc. 2013, 135, 11799–11802.
6. Breslow, R. Chem. Rec. 2014, xxxxx.
7. Breslow, R.; Brown, J.; Gajewski, J. J. J. Am. Chem. Soc. 1967, 89,
4383–4390.
8. Breslow, R. Angew. Chem. 1968, 7, 565–570.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

9. Breslow, R. Pure Appl. Chem. 1971, 28, 111–130.

10. Breslow, R. Acc. Chem. Res. 1973, 6, 393–398.
11. Breslow, R.; Brown, J.; Gajewski, J. J. J. Am. Chem. Soc. 1967, 89,
4383–4390.
12. Wasielewski, M. R.; Breslow, R. J. Am. Chem. Soc. 1976, 98, 4222–4229.
13. Breslow, R.; Goodin, R. J. Am. Chem. Soc. 1976, 98, 6076–6077.
14. Breslow, R.; Grant, J. J. Am. Chem. Soc. 1977, 99, 7745–7746.
15. Breslow, R.; Grubbs, R.; Murahashi, S.-I. J. Am. Chem. Soc. 1970, 92,
4139–4140.
16. Breslow, R.; Chang, H. W.; Yager, W. A. J. Am. Chem. Soc. 1963, 85,
2033–2034.
17. Breslow, R.; Hill, R.; Wasserman, E. J. Am. Chem. Soc. 1964, 86,
5349–5350.
18. Saunders, M.; Berger, R.; Jaffe, A.; McBride, J. M.; O’Neill, J.; Breslow, R.;
Hoffman, J. M., Jr.; Perchonock, C.; Wasserman, E.; Hutton, R. S.; Kuck, V.
J. J. Am. Chem. Soc. 1973, 95, 3017–3018.
19. Breslow, R.; Jaun, B.; Kluttz, R. Q.; Xia, C-Z. Tetrahedron 1982, 38,
863–867.
20. Breslow, R. Pure Appl. Chem. 1982, 54, 927–938.
21. Breslow, R. Mol. Cryst. Liq. Cryst. 1989, 176, 199–210.
22. Thomaides, J.; Maslak, P.; Breslow, R. J. Am. Chem. Soc. 1988, 110,
3970–3979.
23. Chen, W.; Widawsky, J. R.; Vazquez, H.; Schneebeli, S. T.; Hybertsen, M. S.;
Breslow, R.; Venkataraman, L. J. Am. Chem. Soc. 2011, 133, 17160–17163.
24. Cheng, Z-L.; Skouta, R.; Vazquez, H.; Widawsky, J. R.; Schneebeli, S.;
Chen, W.; Hybertsen, M. S.; Breslow, R.; Venkataraman, L. Nat. Nanotech.
2011, 6, 353–357.
25. Chen, W.; Li, H.; Widawsky, J. R.; Appayee, C.; Venkateraman, L.;
Breslow, R. J. Am. Chem. Soc. 2014, 136, 918–920.
26. Breslow, R. Chem. Ind. 1957, 893–894.
27. Breslow, R. J. Am. Chem. Soc. 1957, 79, 1762.
28. Breslow, R. J. Am. Chem. Soc. 1958, 80, 3719–3726.
29. Lopwz-Calahorra, F.; Castro, E.; Ochoa, A.; Marti, J. Tetrahedron Lett. 1996,
37, 5019–5022.
30. Breslow, R.; Schmuck, C. Tetrahedron Lett. 1996, 37, 8241–8242.

74
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 31. Breslow, R.; Baldwin, S. W. J. Am. Chem. Soc. 1970, 92, 732–734.
32. Breslow, R.; Baldwin, S.; Flechtner, T.; Kalicky, P.; Liu, S.; Washburn, W.
J. Am. Chem. Soc. 1973, 95, 3251–3262.
33. Breslow, R.; Dale, J. A.; Kalicky, P.; Liu, S. Y.; Washburn, W. N. J. Am.
Chem. Soc. 1972, 94, 3276–3278.
34. Breslow, R.; Corcoran, R.; Dale, J. A.; Liu, S.; Kalicky, P. J. Am. Chem. Soc.
1974, 96, 1973–1974.
35. Breslow, R.; Corcoran, R. J.; Snider, B. B. J. Am. Chem. Soc. 1974, 96,
6791–6792.
36. Snider, B. B.; Corcoran, R. J.; Breslow, R. J. Am. Chem. Soc. 1975, 97,
6580–6581.
37. Yang, J.; Breslow, R. Angew. Chem., Int. Ed. 2000, 39, 2692–2694.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

38. Breslow, R.; Yan, J.; Belvedere, S. Tetrahedron Lett. 2002, 43, 363–365.
39. Fang, Z.; Breslow, R. Biorg. Med. Chem. Lett. 2005, 15, 5463–5466.
40. Breslow, R.; Campbell, P. J. Am. Chem. Soc. 1969, 91, 3083–3084.
41. Breslow, R.; Campbell, P. Bioorg. Chem. 1971, 1, 140–156.
42. Breslow, R.; Kohn, H.; Siegel, B. Tetrahedron Lett. 1976, 1976, 1645–1646.
43. Breslow, R.; Hammond, M.; Lauer, M. J. Am. Chem. Soc. 1980, 102,
421–422.
44. Breslow, R.; Doherty, J.; Guillot, G.; Lipsey, C. J. Am. Chem. Soc. 1978,
100, 3227–3229.
45. Anslyn, E.; Breslow, R. Geometric Evidence on the Ribonuclease Model
Mechanism. J. Am. Chem. Soc. 1989, 111, 5972–5973.
46. Breslow, R.; Dong, S. D.; Webb, Y.; Xu, R. J. Am. Chem. Soc. 1996, 118,
6588–6600.
47. Levine, M.; Kenesky, C. S.; Zheng, S.; Quinn, J.; Breslow, R. Tetrahedron
Lett. 2008, 49, 5746–5750.
48. Cheng, L.; Abhilash, K. G.; Breslow, R. Proc. Natl. Acad. Sci. U. S. A.
2012, 109, 12884–12887.
49. Cheng, L.; Mahendran, A.; Gonzalez, R. L., Jr.; Breslow, R. Proc. Natl.
Acad. Sci.USA 2014, 111, 7920–7924.
50. Rideout, D.; Breslow, R. J. Am. Chem. Soc. 1980, 102, 7816–7817.
51. Breslow, R.; Rizzo, C. J. J. Am. Chem. Soc. 1991, 113, 4340–4341.
52. Breslow, R.; Groves, K.; Mayer, M. U. Pure Appl. Chem. 1998, 70,
1933–1938.
53. Ginodman, L. M.; Mal’tsev, N. I.; Orekhovich, V. N. Biochemistry (USSR)
1966, 31, 931–935.
54. Breslow, R.; Wernick, D. J. Am. Chem. Soc. 1976, 98, 259–261.
55. Breslow, R.; Wernick, D. L. Proc. Natl. Acad. Sci. U. S. A. 1977, 74,
1303–1307.
56. Breslow, R.; Dong, S. D.; Webb, Y.; Xu, R. J. Am. Chem. Soc. 1996, 118,
6588–6600.
57. Breslow, R. Tetrahedron Lett. 1959, 21, 22–26.
58. Kim, H. J.; Ricardo, A.; Kim, M. J.; Carrigan, M. A.; Frye, F.; Benner, S. A.
J. Am. Chem. Soc. 2011, 133, 9457–9468.
59. Appayee, C.; Breslow, R. J. Am. Chem. Soc. 2014, 136, 3720–3723.
60. Cheng, L.; Breslow, R. Submitted for publication.

75
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 61. Pizzarello, S.; Weber, A. L. Science 2004, 303, 1151.
62. Levine, M.; Kenesky, C. S.; Mazori, D.; Breslow, R. Org. Lett. 2008, 10,
2433–2436.
63. Breslow, R.; Levine, M. Proc. Natl. Acad. Sci. U. S. A. 2006, 103,
12979–12980.
64. Morowitz, H. J. Theor. Biol. 1969, 25, 491–494.
65. Hein, J. E.; Blackmond, D. G. Acc. Chem. Res. 2012, 45, 2045–2054.
66. Breslow, R.; Cheng, Z.-L. Proc. Natl. Acad. Sci. U. S. A. 2010, 107,
5723–5725.
67. Breslow, R; Ramalingam, V.; Appayee, C. Orig. Life Evol. Biospheres 2013,
43, 323–329.
68. Breslow, R.; Sheppard, T. L. Pure Appl. Chem. 1996, 68, 2037–2041.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch004

69. Marks, P. A.; Breslow, R. Nat. Biotechnol. 2007, 25, 84–90.

70. Breslow, R. Pure Appl. Chem. 1997, 69, 223–225.

76
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Chapter 5

Hydrogen Isotopes in Physical Organic

Chemistry
Andrew Streitwieser*
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

Dept. of Chemistry, University of California, Berkeley, California

*E-mail: [email protected]

I was the fortunate recipient of the Norris Award in 1982 but I

also played a role in the very first Norris symposium in honor of
the first Norris awardee, Prof. Christopher K. Ingold. This first
Norris award symposium was a full day affair on Sept 14, 1965,
at the 150th American Chemical Society meetjng in Atlantic
City, NJ. There were three talks in the morning session, by R.
B. Woodward, me and Ingold in that order. Joe Bunnett was
the chair (he also gave the final talk in the afternoon). While
we were getting set up for the morning session, Joe approached
me rather diffidently and hemmed and hawed a bit but I knew
exactly what his concern was. Woodward was well known for
giving beautiful and elegant talks. In a chalk talk he would
start at the top left corner of the board, fill the board with clean
graceful structures, and finish at the bottom right corner. But
his talks were usually long – sometimes very long. I realized
that no matter how long Woodward took for his talk I could not
infringe on the awardee’s time. I assured Joe that I was aware of
this problem and that I was prepared to give an accordian talk.
No matter how little time Woodward left for me I would finish
on time so that Ingold would have his full allotted time. As it
turned out, our fears were groundless. Woodward finished in his
allotted time, I gave my normal talk and Ingold gave his award
talk as planned.
Seventeen years later I was the Norris awardee “for his
pioneering role in the development and application of new
experimental and theoretical tools for the study of physical
organic chemistry.” In this chapter I will emphasize the
experimental tools, most of which involve isotopes of one

© 2015 American Chemical Society

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 form or another, and particularly deuterium. Deuterium has a
number of uses as a tool in physical organic chemistry, many
of which are not original with me. Its use as a tracer is one.
With hydrogen – deuterium asymmetry there is the possibility
of optical activity. In addition there are both equilibrium and
kinetic isotope effects. These isotope effects are normally
divided into primary and secondary isotope effects; primary
when the bond to the hydrogen is broken, secondary when the
bond is not broken but can be altered.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

Discovery of Deuterium

The existence of isotopes was realized in the early 1900’s during a period of
confusing results from radiochemistry experiments (1) . By 1920, isotopes were
reasonably well understood, although deuterium, the stable isotope of hydrogen,
was not discovered until 1931 by Harold Urey. Ferdinand G. Brickwedde, a
physicist friend of Urey’s at the National Bureau of Standards, allowed four liters
of liquid hydrogen to evaporate slowly near the triple point. Spectroscopic study
of the last milliliter in the Fall of 1931 allowed Urey and his assistant, Dr. G. M.
Murphy, to identify the so-called Balmer lines of deuterium. These lines occurred
in natural hydrogen but were so faint that they could have been “ghost” images on
the photographic plate. In the enriched material, the signal was now much larger
and was clearly real. With that Urey announced the discovery of deuterium (2,
3). Two years later, Urey received the Nobel prize in chemistry for the discovery
of heavy water (4–6). This was achieved by electrolysis of water, a method
conceived and developed by Edward W.Washburn of the Bureau of Standards and
continued by Urey after Washburn’s death early in 1934 (7). It seems quite likely
that Washburn would have shared this Nobel prize except for his untimely death.
Pure heavy water was first accomplished by G. N. Lewis at Berkeley in 1933
(5, 6), but followed closely behind by Bonhoeffer in Frankfurt (8). Lewis also
suggested the “deut” stem that Urey adopted in coining the name “deuterium” (9).
What is amazing is how quickly deuterium went from being an oddity to a
tool in chemistry. By 1935, there were already several exchange reactions that
had been studied with heavy water with various organic and inorganic compounds
and even with living materials like fungi. So much so that even as early as 1935,
Wynne-Jones was able to write a Chemical Reviews article summarizing the results
of the exchange reactions that had been run up to that time (10).
Fifteen years later I was a graduate student at Columbia in the laboratory
of William von Eggers Doering. In his group were several projects making use
of deuterium, usually as a tracer, following the course of hydrogen in various
reactions. An example was that of Alfred Wolf, whose lab bench was close to
mine. Al used deuterium to trace hydrogen migrations during the acid-catalyzed
rearrangements of fenchene to other terpenes (11).
78
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Hydrogen-Deuterium Asymmetry

As shown above, the idea of deuterium as a tracer was not strange to me.
When it came time for me to think about post-doctoral work I applied to the
Atomic Energy Commission for a post-doctoral fellowship. For the original
project required for the application I proposed a base-catalyzed exchange study
of esters using deuterium. This proposal was good enough to get my fellowship,
but by the time I got ready to go to MIT to work with Jack Roberts, I thought
of a better project. I proposed to synthesize an optically-active primary alcohol
of the type RCHDOH in which the chirality comes from hydrogen – deuterium
asymmetry. With it I could study the stereochemistry of various reactions at
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

a primary alcohol center. The genesis of this idea is clear. In my PhD work
with Doering my project was working with an optically-active tertiary alcohol.
The first tertiary alcohol had been resolved by Harold Zeiss (12) before I got to
Columbia, and my project was to follow-up that work and study that chemistry.
Subsequent thinking about the stereochemistry of reactions of a primary alcohol
is an obvious association. I had expected hydrogen-deuterium asymmetry to
provide sufficient optical activity to be experimentally useful in studying its
transformations based on recent results by others.
The search for measureable optical activity in compounds of the type
RCHDR’ had been tried for several years, but the first apparent success was that
of Elliott R. Alexander, who took camphene and on reducing it with deuterium got
the di-deuterio system having a small rotation (Figure 1) (13). That small rotation
was a problem, because in those days, before chromatography, the purification of
small amounts of liquids was difficult. Alexander did the control reaction with
hydrogen to give product that was racemic but there was always the possibility
that in the deuterium case a trace of some impurity which was highly optically
active – we note that camphene itself has a very high optical activity – was present.
It wouldn’t take much to give this tiny rotation. Ernest Eliel, shortly thereafter,
looked at the same problem, but in a more straightforward way. He took optically
active phenylethyl chloride and treated it with lithium aluminum deuteride to
yield a deuterio-ethylbenzene that was optically active (Figure 1) (14). He had
the same problem; the product is a liquid having a small rotation, but he was able
to convert it to a solid derivative and then on numerous recrystallizations show
that the optical activity did not change upon subsequent crystallizations. Thus,
there is no question that his ethylbenzene really was optically active because of
the hydrogen-deuterium asymmetry.
I expected to prepare the optically active alcohol by use of a partially
asymmetric reduction such as that being done in Doering’s laboratory at that
time. A labmate, Dick Young, was studying the Meerwein-Ponndorf-Verley
reduction using an optically active secondary alcohol as a hydride source to
make other optically active alcohols (15). I first made 2-octanol-2-d by reduction
of the ketone with LiAlD4 and resolved it in a standard manner. The alcohol
was converted to its oxymagnesium salt by treatment with a Grignard reagent
and was then allowed to react with butyraldehyde (Figure 2). This use of the
magnesium derivative instead of the aluminum salt that is commonly used in the
79
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Meerwein-Ponndoff-Verley reduction was suggested by Bill Saunders, who was
also a post-doc with Jack Roberts at that time. His suggestion worked out quite
well. Because my reaction involved the exchange of an aldehyde carbonyl group
to a ketone carbonyl, the reaction has a much better driving force than the normal
Meerwein-Ponndoff-Verley reactions using secondary alcohols to ketones. The
presumed transition state is shown in Figure 3, a cyclic transition structure with
the magnesium coordinating to the two oxygens. Of the two diasteriomeric
transition structures the favored one is shown in which the smaller group is
opposite the larger group on both sides of the coordination ring.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

Figure 1. Early optically active deuterium compounds

80
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

Figure 2. First preparation of optically active 1-butanol-1-d.

Figure 3. In the favored transition structure the smaller methyl is opposed by

the larger propyl group and hexyl is opposed to hydrogen. In the alternative
diasteriomer, the large propyl is opposed to the large hexyl.

I still had the same problem as my predecessors – proving that the small
observed rotation was that of the butanol-d and not from some impurity. I made use
of the famous experiment (Figure 4) of Hughes, Juliusburger, Masterman, Topley
and Weiss in the mid 30’s in which they showed that in the radioactive iodine
exchange of an SN2 reaction, the rate of racemization of the optically active iodide
was exactly twice that of the rate of isotope exchange (16). This experiment was
one of the important demonstrations that each SN2 reaction occurs with complete
inversion of configuration. In my case, La Roux and Sugden had already published
the rate of reaction of butyl bromide with bromide ion using radioactive bromine
(17).

81
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

Figure 4. Each SN2 exchange of iodine isotopes occurs with inversion of

configuration.

Treatment of the butanol-d with phosphorus tribromide gave optically active

1-bromobutane-1-d which was allowed to react with lithium bromide in 90%
aqueous acetone under the same conditions as Le Roux and Sugden (17). After
varying lengths of time the butyl bromide was isolated and its optical rotation
was measured and compared with the radiochemical exchange rate as shown in
Figure 5 (18). The good agreement shows that that the optical activity is truly that
of the 1-deuterio species.
My first graduate student who got his PhD with me, Bill Schaeffer, took over
this problem. He made a large amount of the optically active alcohol and used it
in several studies that resulted in a number of papers in the late 1950’s, under the
general title of The Stereochemistry of the Primary Carbon. An example is his
study of the reaction of the alcohol with thionyl chloride, an important procedure
for the preparation of alkyl chlorides. The reaction first forms a chlorosulfite that
decomposes to give the alkyl chloride and sulfur dioxide. When applied to our
optically active butanol-1-d the product 1-chlorobutane-1-d was found to have
about 90% inversion of configuration (19). This result suggested that almost all of
the reaction involves an SN2 displacement of the type shown in Figure 6.
At about the same time and independently of my work, Westheimer,
Venneslan and Loewus showed that deutero acetaldehyde could be reduced in an
enzymatic reduction to optically active ethyl alcohol. In their first experiment
in 1953 the amount of ethyl alcohol that they made was too small to measure
the optical activity directly (20). So they showed by a clever technique that their
ethanol-d was enantiomerically pure and that the SN2 reaction used went with
complete inversion of configuration. Later, enough of the alcohol was made so
that the optical rotation could be measured (21). Enzyme routes have been used
82
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 to prepare a number of optically active primary alcohols (22–26). This approach
has the important advantage that it gives product with high enantiomeric purity
but the method fails for some alcohols (27). We showed (28) that the use of
isoborneol is advantageous compared to 2-octanol but other partially asymmetric
chemical reducing agents have been developed that are improvements over my
original synthesis (29, 30).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

Figure 5. Racemization of 1-bromobutane-1-d with LiBr in 90% aqueous

acetone. The line shown is that calculated friom the rate constants of Le Roux
and Sugden (ref. (17) . The experimental points are shown with estimates of
uncertainty. The dotted line is that calculated for isopropyl bromide showing that
the observed racemization cannot be that of a secondary bromide derived from
some rearrangement. This figure is taken from ref. (18)

83
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 6. SN2 reaction on chlorosulfite gives inverted product

Within the next few years a number of optically active 1-deuterio primary
alcohols had been prepared and their absolute configurations were determined.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

These compounds rapidly became quite valuable in the study of the mechanisms of
various chemical and biochemical reactions (29). Many of the early studies have
been reviewed (31, 32). Two examples will illustrate the range of applications.
H. C. Brown’s partially asymmetric hydro-boration (33) method applied to cis-
butene-1-d gave optically active butanol-1-d of relatively high optical purity but
of opposite configuration from that expected on the basis of the accepted reaction
mechanism. A small but significant change in this mechanism accounted for the
observed configuration of the product; the new mechanism was also consistent
with the other chemistry known for the reaction (34).
Mosher (35) prepared optically active neopentyl alcohol, (CH3)3CCHDOH,
and ran a number of reactions with the tosylate. The neopentyl system is known
to rearrange readily under carbocation conditions and its t-butyl group is so close
to the reaction center that SN2 reactions are relatively slow. Nevertheless, reaction
with a number of nucleophiles was shown to proceed with complete inversion of
configuration, just as normal SN2 reactions on less hindered systems.

Secondary α-Deuterium Isotope Effects

Even early in the development of deuterium chemistry, in the middle 30’s,
when many exchange reactions were being studied, it had been noted that
reactions with deuterium were frequently slower than the corresponding reactions
with hydrogen. Even then it was realized that this was probably an effect of
zero-point energies. But it wasn’t until after the war that Jake Bigeleisen was
able to publish his work with Maria Goeppert-Mayer on the theory of equilibrium
isotope effects (36). Shortly thereafter, Bigeleisen also applied the theory to
reaction rates as well. One expression of the theory for hydrogen-deuterium
isotope effects is given as 1:

84
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 This equation embodies a number of approximations such as the
Born-Oppenhimer approximation and Absolute Rate Theory (37–39). In essence,
the isotope effect arises from differences in vibration frequencies between the
reactant state and the transition state. Wolfsberg prefers to think in terms of the
corresponding differences in force constants (40, 41), but these are essentially
equivalent.
One of the first projects I worked on when I got to Berkeley was an isotope
effect of deuterium in the solvolysis reaction. The secondary deuterium effect in
the β position had already been done and turned out to be useful as a measure
of hyperconjugation (42, 43). Instead, I suggested to an undergraduate student,
Bob Fahey (now a chemistry professor at UCSD), that we measure the effect of an
α-deuterium. Bob compared the rate of solvolysis of cyclopentyl-1-d tosylate in
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

acetic acid with that of the undeuterated compound (44, 45). Considering only the
C-H stretching frequency, which is greater for the Csp2 –H bond of a carbocation
than the Csp3 –H bond in the reacant I had anticipated an inverse isotope effect.
I was quite surprised when Bob found that the reaction rate was actually slower
for the deuterio system. I then quickly realized the importance of considering the
bending vibrations. One in particular stood out, the out-of-plane bending of the
α-CH bond in the carbocation. As shown in Figure 7 in which an aldehyde C-H
is used as a model for a carbocation, the inplane bending vibration is similar to
that in a tetrahedral C-H but the out-of-plane bending vibration has much lower
frequency.

Figure 7. Some bending vibrations

With some additional approximations, including putting in the reduced masses

of C-H and C-D and considering the isotope effect to come primarily from zero
point energy differences (46), I simplified 1 to 2, in which the isotope effect is

85
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 related directly to frequency differences between reactant state and transition state
(45).

The largest frequency difference is that from a tetrahedral C-H bend to the
out-of-plane Csp2 –H of the carbocation, about 550 cm-1 but this leads to an
isotope effect of about kH/kD = 1.4, much higher than the observed kH/kD = 1.15.
It was clear that the free carbocation is not a suitable model for the transition
state – the leaving tosylate group is still close and would impede the vibration
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

of the hydrogen. This interpretation was consistent with the result that Shiner had
reported (47) of kH/kD = 1.0 for the SN2 reaction of isopropyl bromide with sodium
ethoxide. In this case, the entering and leaving groups would doubly impede the
H vibration leading to a higher frequency and a smaller isotope effect. This effect
is summarized in Figure 8. In this figure A is a model for an isolated carbocation,
B is a model for an SN1 transition state, C is that of an SN2 reaction and D shows
the bending vibration of a tetrahedral hydrogen. The effect of the X and Y groups
in increasing the zero point energy is apparent from the simplest particle-in-a-box
model.

Figure 8. Out-of-plane bending motions in various structures

Alternatively, we can now make quite good calculations of vibration

frequencies for different models although such computations usually apply to the
gas phase. Moreover, in real molecules the various atomic motions frequently
couple to give complex vibrations in which it can be difficult to perceive simple
components such as the motions of only one hydrogen as in Figure 8. For
86
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 demonstration purposes the computed frequencies in Figure 9 were made on
chlorine compounds in which the heavy chlorines effectively do not couple with
the motions of the light hydrogen and one can clearly see the effect of additional
groups near the vibrating hydrogen. Structure A is that of an isolated carbocation
with no group at either end of the bending motion of the hydrogen. B is a model
of an ion pair with a group at one ond of the vibrating proton. C is a model of
an SN2 reaction with groups at both ends hemming in the hydrogen. The lines
attached to the hydrogen are the displacement vectors for the vibrations. The
absence of such vectors on the other atoms indicates that they all are essentially
motionless in these vibrations.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

Figure 9. Effect on the vibration frequency of groups near the bending hydrogen

The net result of this analysis is that the α-deuterium provides a sensitive
measure of the region near the α-hydrogen in the transition state. This
interpretation was quickly accepted (48) by others and has become an important
tool for the physical organic chemist. Our experimental results were soon
confirmed by others (48, 49) and were extended to related solvolyses. Shiner
(50) has given a thorough review on applications of the α-isotope on solvolytic
displacement reactions. Lee and Wong (51) provided an important application
in their studies of the neighboring group effect in solvolyses. As summarized
in Figure 10, the solvolysis of endo-norbornyl brosylate gives an essentially
normal α-d effect as expected for a solvolyis of the limiting type. The exo-isomer,
however, shows a much lower isotope effect, indicative of neighboring group
participation to give the bridged norbornyl cation intermediate. This and other
examples have been reviewed by Sunko and Borcic (52) and by Scheppele (53).

87
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

Figure 10. α-d Isotope effects in solvolysis of norbornyl brosylates.

Another useful application has been to cycloaddition reactions. The question

indicated in Figure 11 is whether one bond forms first to give an intermediate
diradical or whether the two bonds form concurrently and, if so, whether they are
of equal strength at the transition state or different.

Figure 11. Synchronicity question of cycloaddition reactions.

α-Deuterium isotope effects have been used for this problem as shown in
the recent example in Figure 12 (54). For a two-step reaction there would be no
isotope effect at one end. The isotope effects found are on the low side and slightly
different indicating that the reaction is concerted but not equal and with an early
transition state. Note that the isotope effects are inverse because the CH bonds
involved are changing from sp2 to sp3.
88
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 12. Isotope effects in cycloaddition of nitrostyrene to cyclopentadiene
(54).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

Tritium
Tritium is highly radioactive with a half-life of 12.3 years but as a weak
beta-emitter it is safe in external use and needs only reasonable precautions in
use. It is useful as a tracer just as deuterium but with some distinctive differences.
In particular, it can be used to measure small amounts of reaction. For example,
in early experiments in my laboratory we used tritium exchange to measure
the kinetic acidities of hydrocarbons using a strongly basic system, cesium
cyclohexylamide (CsCHA) in cyclohexylamine (CHA).
Some measurements were relative to benzene; others referred to cyclohexane.
To put all on the same scale we required the relative rates of benzene and
cyclohexane. Under typical conditions benzene has a reaction half-life of about
half a minute, whereas that of cyclohexane is several years. Nevertheless, we
were able to run a mixture of both hydrocarbons in CHA containing tritium,
taking several aliquots within a few minutes and spacing others over days. This
type of procedure requires careful technique and pure materials. The results of
three runs gave k(cyclohexane)/k(benzene) = 1.11 x 10-8, 1.05 x 10-8, 0.97 x 10-8
(55). The good reproducibility suggests that the results are reliable.
We were also able to measure the primary isotope effect for the exchange with
cyclohexane. Deuterium exchange with cyclohexane is too slow to be practical
but could be done using tritium exchange. We used a mxture of cyclopentane
and cycloheptane in CHA containing tritium with cyclohexane-d12 in one run and
cyclohexane in another. The relative rate cyclopentane/cycloheptane should be the
same in both runs and serves as a check. One can show that for small extents of
reaction, the rate of incorporation of tritium into cyclohexane-d12 is actually kD.
The experimental results gave k(cyclopentane)/k(cycloheptane) = 7.5 and 7.8 for
the two runs (56). These results were part of a larger study on the kinetic acidities
of cycloalkanes (57, 58). The isotope effect found in this study was kH/kD = 6.5
after correction for the small amount of H in the C6D12 (55).
These results with tritium point out that as a highly radioactive element, it can
be uniquely useful for measuring slow hydrogen exchange rates. One limitation,
however, needs to be added. The experimental kinetic acidity measured by the
89
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 rate of incorporation of tritium into a molecule does not tell us which hydrogen
positions are involved. This is not a problem for symmetrical molecules like
benzene in which all of the hydrogens are equivalent. It is a limitation for
cyclohexane; the reactivity discussed above does not distinguish between the
exo and endo-hydrogens. The measured kinetic acidity is the average of the two
positions.

Acknowledgments
At the time of my Norris award in 1982 I was just finishing my third decade
at Berkeley. During that time some 168 undergraduate, graduate and postdoctoral
coworkers had worked in my laboratory and I thank them for enjoying chemistry
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

with me. I also thank the several funding agencies (PRF, AFOSR, NIH, NSF) for
making it all possible.

References
1. For an excellent account of this period, see Soddy, F. Nobel Lectures,
Chemistry, 1901-1921; World Scientific: River Edge, NJ, 1999; pp 371−399.
2. Urey, H. C.; Brickwedde, F. G.; Murphy, G. M. Phys. Rev. 1932, 40 (1),
1–15.
3. Urey, H. C.; Brickwedde, F. G.; Murphy, G. M. Phys. Rev. 1932, 39 (1),
164–165.
4. Palmer, W. Presentation Speech. In Nobel Lectures in Chemistry 1922-1941;
World Scientific: Singapore, 1934; pp 333−338.
5. Lewis, G. N. J. Am. Chem. Soc. 1933, 55, 1297–1298.
6. Lewis, G. N.; Macdonald, R. T. J. Chem. Phys. 1933, 341–344.
7. Washburn, E. W.; Urey, H. C. Proc. Natl. Acad. Sci. 1932, 18, 496–498.
8. Bonhoeffer, K. F. Angew. Chem. 1933, 46, 776–779.
9. O’Leary, D. Nat. Chem. 2012, 4, 236.
10. Wynne-Jones, W. F. K. Chem. Rev. 1935, 35, 115–123.
11. Doering, W. v. E.; Wolf, A. P. Perfumery and Essential Oil Record 1951, 42,
414–415.
12. Doering, W. v. E.; Zeiss, H. H. J. Am. Chem. Soc. 1953, 75, 4733–4738.
13. Alexander, E. R.; Pinkus, A. G. J. Am. Chem. Soc. 1949, 71, 1786–1789.
14. Eliel, E. L. J. Am. Chem. Soc. 1949, 71, 3970–3972.
15. Doering, W. v. E.; Young, R. W. J. Am. Chem. Soc. 1950, 72, 631.
16. Hughes, E. D.; Juliusburger, F.; Masterman, S.; Topley, B.; Weiss, J. J. Chem.
Soc. 1935, 1525–1929.
17. Le Roux, L. J.; Sugden, S. J. Chem. Soc. 1939, 1279–1283.
18. Streitwieser, A. J. Am. Chem. Soc. 1953, 75, 5014–5018.
19. Streitwieser, A.; Schaeffer, W. D J. Am. Chem. Soc. 1957, 79, 379–381.
20. Loewus, F. A.; Westheimer, F. H.; Vennesland, B. J. Am. Chem. Soc. 1953,
75, 5018–5023.
21. Levy, H. R.; Loewus, F. A.; Vennesland, B. J. Am. Chem. Soc. 1957, 79,
2949–2953.
90
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 22. Althouse, V. E.; Feigl, D. M.; Sanderson, W. A.; Mosher, H. S. J. Am. Chem.
Soc. 1966, 88, 3595–3599.
23. Althouse, V. E.; Kaufmann, E.; Loeffler, P.; Ueda, K.; Mosher, H. S. J. Am.
Chem. Soc. 1961, 83, 3138–3142.
24. Althouse, V. E.; Ueda, K.; Mosher, H. S. J. Am. Chem. Soc. 1960, 82,
5938–5941.
25. Günther, H.; Alizade, M. A.; Kellner, M.; Biller, F.; Simon, H. Z. Naturfor.
1973, 28, 241–246.
26. Günther, H.; Biller, F.; Kellner, M.; Simon, H. Angew. Chem., Int. Ed. 1973,
12, 146–147.
27. Mosher, C. W.; Mosher, H. S. Indian J. Chem. 1992, 31B, 900–902.
28. Streitwieser, A; Granger, M. R. J. Org. Chem. 1967, 32, 1528–1529.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

29. Reich, C. J.; Sullivan, G. R.; Mosher, H. S. Tetrahedron Lett. 1973, 14 (17),
1505–1508.
30. Noyori, R. Pure Appl. Chem. 1981, 53, 2315–2322.
31. Arigoni, D.; Eliel, E. L. In Topics in Stereochemistry; Eliel, E. L., Allinger,
N. L., Eds.; Wiley-Interscience: New York, NY, 1969; Vol. 4, pp 127−244.
32. Verbit, L. Prog. Phys. Org. Chem 1970, 7, 60–80.
33. Brown, H. C., Hydroboration; W. A. Benjamin, Inc.: New York, NY, 1962.
34. Streitwieser, A.; Verbit, L.; Bittman, R. J. Org. Chem. 1967, 32, 1530–1532.
35. Mosher, H. S. Tetrahedron 1974, 30, 1733–1745.
36. Bigeleisen, J.; Mayer, M. G. J. Chem. Phys. 1947, 15, 261–267.
37. Melander, L. Isotope Effects on Reaction Rates; Ronald Press Co.: New York,
NY, 1960; p 189.
38. Melander, L.; Saunders, W. H., Jr. Reaction Rates of Isotopic Molecules;
Wiley: New York, 1980.
39. Van Hook, W. A. Kinetic Isotope Effects: Introduction and Discussion of the
Theory. In Isotope Effects in Chemical Reactions; Collins, C. J., Bowman,
N. S., Eds.; Van Nostrand Reinhold: New York, 1970.
40. Wolfsberg, M. Ann. Rev. Phys. Chem. 1969, 20, 449–478.
41. Wolfsberg, M. Acc.Chem. Res. 1972, 5, 225–233.
42. Shiner, V. J., Jr. J. Am. Chem. Soc. 1953, 75, 2925–2929.
43. Shiner, V. J., Jr. J. Am. Chem. Soc. 1954, 76, 1603–1606.
44. Streitwieser, A., Jr.; Fahey, R. C. Chem. Ind. 1957, 1417–1418.
45. Streitwieser, A., Jr.; Jagow, R. H.; Fahey, R. C.; Suzuki, S. J. Am. Chem.
Soc. 1958, 80, 2326–2332.
46. Melander, L. Arkiv. Kemi 1950, 2, 213–292.
47. Shiner, V. J., Jr. J. Am. Chem. Soc. 1952, 74, 5285–5288.
48. Johnson, R. R.; Lewis, E. S. Proc. Chem. Soc. 1958, 52.
49. Mislow, K.; Borcic, S.; Prelog, V. Helv. Chim. Acta 1957, 40, 2477–2480.
50. Shiner, V. J. Deuterium Isotope Effects in Solvolytic Substitution at Saturated
Carbon. In Isotope Effects in Chemical Reactions; Collins, C. J., Bowman,
N. S., Eds.; Van Nostrand Reinhold Co.: New York, NY, 1970; pp 104−135.
51. Lee, C. C.; Wong, E. W. C. J. Am. Chem. Soc. 1964, 86, 2752–2753.
52. Sunko, D. E.; Borcic, S. Secondary Isotope Effects and Neighboring Group
Prticipation; Van Nostrand Reinhold: New York, NY, 1970.
53. Scheppele, S. E Chem. Rev. 1972, 72, 511–532.

91
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 54. Kwiatkowska, M.; Jasinski, R.; Mikulska, M.; Baranski, A. Monatsh. Chem.
2010, 141, 545–548.
55. Young, W. R. Kinetic Acidities of Cycloalkanes and Other Hydrocarbons.
Dissertation, University of California, Berkeley, 1967.
56. Streitwieser, A., Jr.; Young, W. R.; Caldwell, R. A. J. Am. Chem. Soc. 1969,
91, 527–528.
57. Streitwieser, A., Jr.; Caldwell, R. A.; Young, W. R. J. Am. Chem. Soc. 1969,
91, 529.
58. Streitwieser, A., Jr.; Young, W. R J. Am. Chem. Soc 1969, 91, 529–530.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch005

92
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Chapter 6

Exploring Free Radicals: The Life and

Chemistry of Glen A. Russell
Kathleen M. Trahanovsky1,* and E. Thomas Strom2,*
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

1Department of Chemistry, Iowa State University (retired), 1605 Gilman

Hall, Ames, Iowa 50011
2Department of Chemistry & Biochemistry, University of Texas at Arlington,

Box 19065, 700 Planetarium Place, Arlington, Texas 76019

*E-mail: [email protected]; [email protected]

The career of Glen A. Russell essentially covered the second

half of the 20th century. Russell carried out free radical
research using traditional methods and the then new technique
of electron spin resonance spectroscopy. The results were
ground-breaking research on free radical conformations,
halogenations, oxidations, directive effects, substituent effects,
and electron transfer. This chapter deals with the early research
themes of the Russell group and describes Russell’s career,
accomplishments, and personality.

Introduction
Even today, some 115 years after the first convincing proof of their existence,
organic free radicals tend to be the Rodney Dangerfields of chemistry, i.e., “They
don’t get no respect!” Consider the normal organic chemistry textbook and
measure the page counts devoted to heterolysis versus homolysis or heterolytic
reactions versus homolytic reactions. The page count for the former is normally
an order of magnitude greater than that for the latter. This ratio faithfully reflects
the emphasis given ionic reactions versus free radical reactions during the first
half of the 20th century. The second half of the 20th century featured large
advances in the knowledge of free radicals and their reactive chemistry, and Glen
Alan Russell (Figure 1) was one of the prime movers in this effort.

© 2015 American Chemical Society

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 1. Glen Russell as an Iowa State faculty member in the ‘60’s. Photo
courtesy of Iowa State.

Most of the chemical career of Glen A. Russell was spent searching for free
radicals and exploring their chemistry. This chapter proposes to tell something of
the life of Glen Russell and to give a broad overview of his chemistry research.
This is an apt time to do so, as many of the people who knew Russell and helped
in his work are passing from the scene. We two were part of the early group of
Russell graduate students, and we can testify first hand about the brilliance of the
man. Russell’s many avenues of free radical research helped move this area into
the mainstream of physical organic chemistry, and consequently he was honored
with the 1983 Norris Award.
For another view on research in the Russell group, we also recommend a
memoir by our fellow graduate student Edward G. Janzen (1) that covers Janzen’s
early days in the group. The Janzen book covers some of Russell’s career during
the 1960-64 time period, gives a sense of the excitement of being in a burgeoning
research group about to “take off” in creativity, and shows the transitions a graduate
student goes through in the path to becoming an independent researcher.

94
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The proof that such an entity as a free radical existed is quite reasonably
attributed to the synthesis of triphenyl methyl radical (Figure 2) by Moses
Gomberg in 1900 (2). However, the radical was not stable enough to be
characterized by the traditional methods, such as taking a melting point. Syntheses
of free radicals were viewed with some suspicion by other chemists.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 2. One of the resonance structures of triphenyl methyl radical.

Indeed, as you can read elsewhere in this volume in Arthur Greenberg’s

chapter on James Flack Norris, Norris himself was dubious of Gomberg’s
claim. Examples of stable radicals were found in the realm of organic
nitrogen compounds, with Goldschmidt’s 2,2-diphenyl-1-picrylhydrazyl (DPPH)
(Figure 3) the prime example (3). Nevertheless, the agnostic view of many
chemists toward free radicals is shown by the treatment of Koelsch’s radical,
1,3-bisbiphenylene-2-phenylallyl, by some unknown referee (Figure 4).

Figure 3. One of the resonance structures of DPPH.

95
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 4. One of the resonance structures of Koelsch’s radical.

In a series of elegant synthetic papers, C. Frederick Koelsch painstakingly

worked his way toward his landmark synthesis of the radical. The last paragraph
of a paper published in 1932 claimed the synthesis of the radical and promised a
further paper on the radical (4). That paper on the radical’s synthesis was submitted
on June 9, 1932, but the paper on Koelsch’s radical had to wait 25 years for
publication (5). A referee held that “the properties of the chief compound could
not be those of a radical.” Koelsch’s radical was incredibly stable to oxygen. The
convincing factor behind the final, long-delayed acceptance of Koelsch’s paper on
the radical was that of electron spin resonance (ESR). An ESR experiment proved
that this stable material was in fact a free radical.
The paramagnetism associated with unpaired electrons obviously provided a
clear route to proving the existence of a free radical, provided the radical stayed
around long enough for a measurement. Magnetic susceptibility measurements
were one way of accomplishing this. However, it was the invention of electron
spin resonance spectroscopy in 1944 by the Russian scientist E. K. Zavoisky at
the University of Kazan that opened a new era in the exploration of free radical
chemistry (6). (It should be noted that the current fashionable term for this type
of spectroscopy is electron paramagnetic resonance (EPR). The reasons for this
switch are neither clear nor convincing to us.)
If not for the dislocations caused by World War II, Zavoisky might have been
known as the inventor of nuclear magnetic resonance spectroscopy (NMR) and a
Nobel Laureate to boot. In 1940 Zavoisky and two colleagues set out to search
96
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 for proton magnetic resonance (7). They decided to use pure water containing
copper and manganese salts. Zavoisky detected the proton magnetic resonance of
water in May/June, 1941. Hitler invaded the USSR in June, 1941. Many important
Russian scientists were sent east to Kazan, causing disruption of research by home
scientists. Zavoisky’s laboratory was closed down. When he started up work
again, he moved his efforts into paramagnetism, with ESR spectroscopy the result.
Glen Russell’s entry into independent free radical chemistry took place as
ESR spectroscopy began to flower. When Varian Corp. began to manufacture
commercial ESR spectrometers, the field of free radical chemistry was forever
changed. Instead of having to prove the existence of free radicals by indirect
methods, one could actually see them form, react, and decay. For coverage of the
foundations and history of this spectroscopy, we recommend the volume edited by
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Eaton, Eaton, and Salikhov (8). ESR studies became a major element in Russell’s
research, although he accomplished much also with traditional methods. After this
introduction, we continue on to cover some of the details of Russell’s life, followed
by an examination of much of his important chemistry.

Russell’s Life and Career

Unless otherwise referenced, the material in this section was obtained from
Russell’s daughters June and Susan, his widow the late Dr. Martha Russell, the
Russell archive at Iowa State University, and our own personal knowledge. Since
there are two of us writing this material, the authors will normally use the term
“we” when referring to material known to the two of us. Occasionally stories
and incidents will come up involving only one of us. We will introduce that
material by giving the name of the person involved, either Kathy or Tom, in the
first sentence, and then using the first person pronoun for most of the material.
The early biographical material cited immediately below was obtained from Susan
Russell.
Glen Allan Russell was born on Aug. 23, 1925, in Pittstown in Rensselaer
County, NY, near the border with Vermont. He was the only child of John Allan
and Marion Russell. Russell had Indian ancestry on his father’s side and was 1/16
Cherokee. He grew up in the country on a farm that had dairy cows (Figure 5).
With the onset of the great depression, times were hard, and Russell’s father
supplemented the farm income by doing work with explosives. This resulted in his
losing sight in one eye. The house was an old foursquare farm house with four large
upstairs bedrooms. Russell’s parents had one bedroom, each set of grandparents
had another bedroom, and Russell had the final room.
Russell was interested in science and math early on. His mother bought him
a chemistry set when he was young. He was fascinated with the set and regularly
conducted experiments in his room. At least one of these experiments had a bad
outcome, resulting in an explosion that left burn marks on the ceiling. When
he wasn’t doing science experiments in his room, young Glen was busy hunting
squirrels which he shot and the family ate.
97
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 5. Glen Russell around age 6. Photo by Marion Russell.

He attended public schools in Hoosick Falls, NY. His elementary school

was very small—a one-room rural school house. He finished his studies there
quickly, because he was soon convinced that he knew more than the teacher. He
did study chemistry in high school. At the age of 16 he entered nearby Rensselaer
Polytechnic Institute (RPI). He had a full scholarship, and that way he could still
help out on the farm. He majored in chemical engineering, because his mother
thought this was a more practical and profitable choice with respect to jobs.
(Sadly, that’s still true!) He didn’t care for the engineering classes, though, and
told his daughter the story of the only poor grade he got. This was in a drafting
class. The student behind him kept pestering him during the exam, asking how to
complete the question. Russell finally stood up, went to the student’s desk, and
completed the question for him. As a result, Russell got an F for the exam. At
RPI he was awarded the Ricketts Prize in Chemical Engineering, given to the top
student, and the Research Prize awarded for the best senior thesis. That topic was
“The Hydrolysis of Diphenylketimine Hydrochloride.” He was also elected to Phi
Lambda Upsilon, Tau Beta Phi, and Sigma Xi.
After receiving his BChE degree from RPI in 1947, he stayed on to receive
an MS in chemistry in 1948. He was accepted to graduate school at both Purdue
and Harvard, but he chose Purdue, because, unlike Harvard, he could enter Purdue
mid-year. He didn’t start graduate school with the idea of working for H. C. Brown,
but Russell realized that his first choice as a mentor was not scientifically sound.
He switched to Brown rather soon, and his Ph.D. was completed in 1951 (Figure
6). Brown felt that free radicals were not a “hot” area for research. Russell felt
otherwise, and he went ahead single mindedly on free radical projects to prove
Brown wrong. His dissertation topic was “The Effect of Structure on the Relative
Reactivities of Hydrogen Atoms as Determined in Photochemical Halogenations.”
98
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 6. Russell in academic gowns at the awarding of his Ph.D. Photo by

Marion Russell.

Russell’s thesis is an intimidating 150 pages, but there was more. The table
of contents also promised four abstracts. In fact, those four abstracts were instead
four full papers that totaled 60 additional pages; papers that were clearly destined
for J. Am. Chem. Soc. They were written up as authored by Russell and Brown,
with Brown’s name first. These four papers were eventually published over the
next four years with some of the titles slightly changed (9–12).
There was a perception among his graduate students that Russell had little
or no help from Brown, and that Brown did no free radical work before or after
Russell’s time. As noted above, Russell worked essentially independently, but any
belief that Brown was a novice at free radical studies is incorrect. After obtaining
his Ph.D. at Chicago, Brown did a post doc with Kharasch on free radicals. One
important paper dealt with peroxide catalyzed chlorination of hydrocarbons (13).
Brown continued to publish occasional free radical papers in the ‘40s. A 1948
paper written with Ash covered directive effects in aliphatic chlorination (14). It
is true, however, that after Russell’s time Brown focused more on the research
which eventually resulted in a Nobel Prize. Certainly there was mutual respect
between the two. An incident related by Russell graduate student Kirk Schmitt (15)
illustrates the bond. In a group meeting a debate was going on between a graduate
student and a post doc about the nature of the bonding in a cyclopropyl-substituted
radical. This was still at the time of the dispute between Winstein and Brown about
99
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 whether norbornyl cations were non-classical or two rapidly equilibrating classical
cations, a topic covered elsewhere in this volume by Paul Schleyer. In the course
of arguing about the sign of the coupling constants, all sorts of structures were
drawn, one of which the graduate student referred to as “classical.” The post-doc
said, “Aha, I think I smell a Brown man.” From the back of the room in a clear,
low voice Russell said, “We’re all Brown men here.” End of discussion!
Russell was definitely interested in an academic position, and he had
won a Fulbright scholarship for post-doctoral work in England. However, the
Korean War was still going on at that time. Russell was under 26, and his draft
board informed him that he would be drafted from any post-doctoral position.
Consequently, he took a position with GE in Schenectady, NY, close enough to
the family farm that he could work there weekends. Russell was promised he
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

could have substantial amounts of time to do basic research at the GE facility.

Both Kirk Schmidt (15) and Keith Ingold (16) report that what that meant in real
life was that he could spend all the time at night he wanted to do basic research,
but during the day he had better work on silicones.
Nevertheless, noted free radical chemist Frank Mayo was also at the GE
lab, and the situation regarding pure research was much better than at the usual
industrial laboratory.
At Purdue Russell met fellow Brown graduate student Martha Havill. Martha
had graduated with a mathematics degree from the University of Rochester in 1945
at the age of 19, and then she obtained an M.S. in chemistry from the University
of Buffalo in 1947. She went to work at National Aniline. She described her job
as “Diazotize in the morning and couple in the afternoon, day after day.” To find
a more fulfilling career, she entered graduate school at Purdue, where she joined
Brown’s group (Figure 7). When Martha took Brown’s course in organometallic
lab techniques, Glen Russell was the “obnoxious” TA. On a long train trip home
to upstate New York for Christmas, they became better acquainted and “formed a
more comfortable relationship” from then on. They became engaged after Russell
graduated and took the position at GE. It was another two years before Martha
graduated. She frequently said that Brown was fond of telling people that women
in his group got TWO Ph.Ds, their husband’s and their own. Brown no doubt
thought that was clever, but nowadays we would view his comment as sexist. And
it was.
Russell and Martha were married on June 6, 1953. Figure 8 shows the happy
couple in the MG that Glen had purchased in 1953 and which became one of his
great passions in life. Russell drove the car to the university nearly every day
for 40 years, including through brutal Iowa winters. After his death his daughter
Susan had it shipped to Portland, where she still drives it regularly. To us graduate
students, it seemed that Glen and Martha complemented each other very well.
Where Russell was very reserved and reflective, Martha was warm, outgoing, and
bubbly. She kept up contact with all of his graduate students, and she had a long,
successful teaching career at Iowa State. When she retired in 2005 at the age of
79, the Iowa State Chemistry Department changed the name of the chemistry help
room to the Martha E. Russell Chemistry Help Center in recognition of her career
as an outstanding educator. She died on July 6, 2014 and is buried with her husband

100
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 at the Iowa State University Cemetery. To hear Martha Russell in her own words,
we recommend reading her letter in Janzen’s memoir (17).
After she received her Ph.D., Martha Russell eventually went to work at GE.
Their daughters testify that neither Glen nor Martha were motivated by financial
gain. Since most people spend the majority of their lives working, their view was
that it was absolutely critical to work at something you love, which for them would
have been academic jobs. At GE they lived on one salary, with the other salary
being put aside for the purchase of a house one day for cash, which they did after
their move to Iowa State. When GE suffered some financial reverses, it was Martha
who was laid off. Martha said she was told this was because she was a married
woman and had a husband to support her.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 7. Graduate student Martha Havill at Purdue. Photo by Glen Russell.

101
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 8. Glen and Martha Russell in the MG on their wedding day. Photo by
Marion Russell.

Whatever limitations on basic research that GE put in Russell’s way proved

ineffective, because he published 13 papers from there, all but two with himself as
sole author and all but two appearing in the prestigious Journal of the American
Chemical Society (JACS). In 1956 he published four significant papers back
to back in JACS (18–21) as sole author, and in 1957 he published a JACS
communication on his ground-breaking work on solvent effects on the reactions
of chlorine atoms (22). The story in our group was that Cheves Walling from
Columbia had sent Russell a congratulatory telegram on the publication of this
work. With his credentials as a basic researcher demonstrated, Russell was ready
to seek an academic job, and one was about to open up.
Iowa State’s George Hammond of “Hammond postulate” fame had just
started working in the area of photochemistry. Iowa State not only had chemistry
courses for chemistry and chemical engineering students, but agriculture and home
economics students also had to take elementary chemistry courses. Hammond
did not like teaching agriculture students and expressed his distaste for this task
quite colorfully—so colorfully that we won’t repeat it. In any event, Cal Tech
lured Hammond away, so there was an opening for a physical organic chemist on
the Iowa State faculty. Glen Russell submitted his application to the head of the
search committee, the renowned organometallic chemist, Henry Gilman. H. C.
Brown (23) wrote an extremely strong supporting letter, mentioning that Russell
102
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 was his leading student and that he had published a huge number of significant
publications. Brown finished his letter by saying

“In my opinion Dr. Glen Russell is the most capable of our younger
men in the physical organic field. Considering what he has been able
to accomplish primarily through his own efforts, there is little doubt that
with only a few students to assist his research progress he would soon
develop into one of the top leaders in the field.”

Consequently, in May, 1958, Charles Goetz, Chair of the department, offered

Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Russell a position of Associate Professor for the sum of $10,000 a year (24). At
that time GE was paying Russell $12,000 a year. His teaching load was to be six
lecture hours a week.
Readers who are non-academics would not find anything remarkable in this
offer. Academic readers, however, would be surprised. All chemists know that
“Like dissolves like.” Academicians also know that “Like hires like.” For a chemist
to come in from industry at the tenured level of associate professor is almost
unheard of. The Iowa State offer testifies to the high quality of Russell’s previous
publications and the respect that everyone had for H. C. Brown’s judgment.
Despite the loss of Hammond, the organic faculty at Iowa State was still of
high quality. Henry Gilman had an active research group, although Gilman’s
failing eyesight made it difficult for him to keep up with the literature. Ernest
Wenkert was a fine natural products chemist, and Charles DePuy was well
underway in his career, which resulted in his being chosen as a Norris Award
winner in 2002. In 1957 Iowa State had hired as instructor a bright young chemist
named Orville Chapman. We graduate students were intimidated by Chapman,
who was just a few years older than we were and incredibly sharp. Chapman
worked in organic photochemistry and won the ACS Pure Chemistry Award in
1968. He was elected to the National Academy of Sciences in 1974, joining Iowa
State chemists Henry Gilman and Frank Spedding in that august body.
Then UCLA snatched him away. Russell and Chapman were close friends
and had coffee together every day, joined later by Walter Trahanovsky when he
joined the faculty.
Russell was offered $500-600 in “start up” funds, which seems puny now,
but this research support was adequate for a new faculty member. Large numbers
of teaching assistants were needed, since Iowa State required so many of their
students to take chemistry. Physical, inorganic, and analytical chemists were
well supported as research assistants at the Iowa State Ames Lab of the AEC.
Consequently, most of the department’s general fund was available for the organic
chemists to use. Russell took up the appointment on Aug. 15, 1958. Glen
and Martha bought the ranch-style house on Murray Drive, which they lived in
throughout their time in Ames. It was to be the site of many memorable parties
for the grad students and post-docs.
Within a couple of months Russell developed three research proposals:
Directive Effects in Aliphatic Substitution (sent to the Office of Naval Research);

103
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Electrophilic Substitution on Saturated Carbon Atoms (directed to the NSF); and
Solvent Effects in the Reactions of Free Radicals and Ions (sent to the PRF). Also,
Goetz recommended Russell for an Alfred P. Sloan Fellowship, and Russell was
a Sloan Fellow from 1959-1962. Russell’s salary for 59-60 had a significant jump
to $11,500, and in 60-61 it was $12,250. Goetz then negotiated a $500 mid-year
raise when he heard that Penn was interested in hiring Russell. In 1961 Russell
was promoted to full professor. We are not sure whether all of Russell’s original
proposals were funded, but our experience was that there was always plenty
of money. After one year as a teaching assistant, a Russell student normally
became a research assistant; and highly qualified graduate students, Ed Janzen for
example, could start out as a research assistant.
Much of the basic equipment from Hammond’s group was turned over to
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Russell. One large lab had a wall of distillation columns and a smaller lab had
a large vacuum rack in the center. Russell was an expert glassblower, and he also
continued to use the carpentry and mechanical skills that he had learned on the
farm, building three gas chromatographs that were used by the group through the
early ‘60’s.
Russell started out with two graduate students, Anthony Moye and Dale
Hendry. In 1959 the group expanded with new graduate students Robert Bridger,
Roger Williamson, Richard Kriens, and Akihiko Ito plus post doc Krishna Nagpal.
Russell’s graduate students seemed to be older than the norm and interested in
eventually working in industry. Several were veterans, and several were married.
The group generally chose physical chemistry as their inside minor (after all, we
were going to be physical organic chemists) and many emulated their mentor by
choosing chemical engineering as their outside minor. Along with the expansion
of Russell’s group came expansion of his family. Martha had lost a still-born girl
just before the move to Ames, but Martha gave birth to daughters Susan in 1960
and June in 1961.
We two entered graduate school in Sept., 1960. Our fellow student Ed
Janzen had come at the beginning of the summer. Kathleen Desmond (later Kathy
Trahanovsky) came from Emmanuel College in Boston with the intention of
getting a master’s degree and returning home to New England. When I did well
on the physical chemistry qualifying exam, Russell suggested that I major in that
area; but I liked organic more and became the first female member of the group.
There was only one other woman organic graduate student in the department,
and she was finishing up an MS with DePuy. Russell suggested I study the
reactivity of the succinimidyl radical, as it would make a tidy master’s project.
That project turned out to be very significant, and we will return to it in part 3 of
this chapter. After completing my MS degree, I returned to New England to a
temporary teaching position at Emmanuel for three years. I enjoyed teaching and
realized that I would need a Ph.D. to make a career of it. I considered Princeton,
but, when I requested an application, I was told that they did not accept women
graduate students. I had left Iowa State in good standing, so I returned to the
Russell group, doing a thesis on ESR of some semiquinones and graduating in
1969. Walter Trahanovsky joined the Iowa State organic faculty in 1964 while I
was gone. We were married in 1967. I eventually became a teaching member of
the Iowa State chemistry faculty. Martha Russell and I progressed together from

104
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 part-time Temporary Instructors to Continuous Adjunct Associate Professors and
Associate Coordinators of General Chemistry over the next thirty years.
In the early ‘60’s, students routinely worked in the lab nights and weekends.
Russell was in his office every Saturday morning for 40 years, if he was in Ames.
When he was hired, the college work week was 44 hours, and there were Saturday
morning classes. There was an easy camaraderie among students and lots of
interactions among groups. In a time before so many OSHA/EHS rules, lab doors
were usually left open and cross fertilization of research ideas and equipment
sharing flourished, along with coffee breaks and football pools. I was accepted
as a peer, even garnering an invitation for beer at Dave’s Lounge in downtown
Ames shortly after joining the group. However, I was presented with a large bell
to ring when approaching the large lab, because there Bob Bridger and Roger
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Williamson often regaled their lab mates with war stories and bawdy songs.
E. Thomas (Tom) Strom entered graduate school with an MS in nuclear
chemistry from UC-Berkeley. My mother had developed leukemia, so I wanted
to finish my graduate work near to our home in Des Moines, IA. Ames was
just 30 miles away. Although my BS Chem degree was from the University of
Iowa, I was well aware of Iowa State’s strength in engineering and the sciences.
My experience in nuclear chemistry convinced me I would do better in organic
chemistry, but my time at UC-Berkeley did give me a strong P-Chem background.
Russell first thought I might do some P-Chem on the solvent-chlorine
complexes he had proposed earlier (22), but then he suggested that Ed Janzen and
I look into using the electron spin resonance (ESR) spectrometer he had recently
acquired. Dr. Roy King from England was a former DePuy post-doc, who was
very skilled in electronics. His position had morphed into that of instrument
specialist. Russell knew nothing about the instrument, so Roy taught us both.
Roy believes that the ESR and NMR spectrometers recently purchased came
from a $50,000 Army grant that Russell had obtained, but we cannot confirm
that (25). There was one huge problem with the instrument set up. There was
only one 12-inch magnet for the two spectrometers! Consequently, every time we
wanted to do ESR work, the NMR spectrometer had to be shut down. Then as
now, NMR was the work horse instrument for proving the structure of an organic
compound. When we did ESR work, Ed and I got dirty looks; and we felt that
we had to be on the instrument 24 hours a day to justify shutting down the NMR.
Roy worked out a procedure wherein the ESR would be operating one week out
of four. As our ESR work progressed, we needed more and more time. Russell
finally got the funds to buy a new six-inch magnet solely for the ESR, which
ended the problem. However, he did not buy the Varian water-cooling system
that went with it. Instead he bought a huge milk can, a pump, and lots of copper
tubing and had the shop put it all together to form a water cooling system for the
magnet. Perhaps this was a throwback to his old dairy farming days. Roy felt this
makeshift setup looked unprofessional, but it did the job for a long time.
More top notch students such as Ed Geels, Jerry Hunt, Al Bemis, Joe Schoeb,
Leo Chang, Steve Weiner, Gary Mikol, Maria Young, and Wayne Danen joined
the group, as did post-docs Hans Dieter-Becker (a wizard in the lab!), Frank
Smentowski, Ryusei Konaka, and John Gerlock; and the research really started
to roll. A tangible demonstration of this was Russell’s winning the 1965 ACS

105
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Award in Petroleum Chemistry, now known as the George A. Olah Award in
Hydrocarbon or Petroleum Chemistry. Kathy and Tom were in Detroit for the
ACS meeting at which Russell received this award. Russell had the custom of
inviting all of his students, both present and past, to dinner at his expense at
national ACS meetings. We both remember having the Detroit dinner at the
Playboy Club, a first visit for both of us. It was tamer than Kathy expected, but
the matchbook was a super souvenir for her, then a teacher at a Catholic women’s
college.
There seemed to be some mystique about this award, but its origins and
character are not clear to us. Russell received congratulatory messages from
previous winners Vladimir Haensel (26), Robert Taft (27), George Hammond
(28), Harold Hart (29), and George Olah (30). A letter in the Russell archives from
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

George Pimentel to George Olah (31) refers to a “club” or society of petroleum

chemistry award winners, a term also used by Harold Hart (29) in his letter. At
this late date neither 1964 winner George Olah (32) nor 1967 winner Andrew
Streitwieser (33) remembers what this club or society of recipients was. 1968
winner Keith Ingold believes this might have been something set up by the award
sponsor (34). He thinks that the sponsor, then Precision Scientific, displayed full
face, color photos of the winners at their head office, since he had to supply such
a photo. Perhaps the “club” may only have referred to a dinner given by the
sponsor for the current winner to which previous winners were invited.
As the years passed, the honors and recognition for Russell increased. He
was chosen as a Fulbright Hayes Lecturer at the University of Wurzburg. In 1966
he was the Reilly Lecturer at Notre Dame. In 1967 he was Chair of the Gordon
Conference on Radical Ions. In 1972 he was a Guggenheim Fellow at the Centre
d’Etude Nucleaires de Grenoble. That year he also won the Iowa Medal given by
the Iowa City ACS Section, while being promoted to the rank of Distinguished
Professor at Iowa State. He was the recipient of the ACS Midwest Award in 1974.
In 1975 he participated in the USA-USSR NAS Exchange Program. His entire
family was with him for summer conferences in Europe or Japan, for his sabbatical
in Grenoble, and for an extended stay in the then Soviet Union. He also was fond
of finding an excuse to go to Germany for Oktoberfest. In 1983 he won the award
that has justified this chapter, the Norris Award in Physical Organic Chemistry. He
won the Iowa Governor’s Medal for Science Achievement in 1988.
Russell thought every family should have a sport they could do together, and
he chose skiing (more about that later). Figure 9 shows the Russell family at Sun
Valley during the week of Christmas of 1970. On his trips to Germany, Glen
picked up various beer drinking dances, which he was fond of demonstrating
during parties for his group (Figure 10). Through the years his research output
slowed down a little, but it was still significant. He participated in the various
meetings and conferences concerning free radicals. Figure 11 shows Russell
with his former graduate student James Tanko, now of Virginia Tech, at the 1993
Gordon Conference on Free Radicals.

106
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 9. From left to right, Glen, Susan, Martha, and June Russell at Sun Valley.
Photo by Glen Russell using a photo timer.

Figure 10. Russell showing off his dance skills. Photo by Susan Russell.

107
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 11. James Tanko and Glen Russell at the 1993 Gordon Conference. Photo
by Naushadalli Suleman.

In either late 1993 or early 1994, Tom realized that Russell was nearing his
70th birthday. I remembered that the Journal of Organic Chemistry (JOC) had
published a special issue for the 70th birthday of Henry Gilman, so I thought that
surely they would do the same for Glen Russell. I contacted JOC Editor-in-Chief
Clayton Heathcock about my suggestion. Heathcock gently but firmly told me that
Henry Gilman was a special case in view of his effort towards establishing that
journal and also, if he published a special issue every time a significant organic
chemist reached his or her (back then it would have been his) 70th birthday, he’d
have nothing but special issues. However, there would be nothing wrong with
having a number of Russell ex-students, friends, and colleagues submit papers
and asking them to be published in the same issue, each paper with a dedication to
Russell. Therefore, that was the route we took. Tom asked all the Russell students,
friends, and colleagues he knew to submit papers, and at Iowa State Walter and
Kathy Trahanovsky did the same. For a long time we kept Russell in the dark
about the effort, but eventually we had to let him know what was happening.
For one thing, what would a special Russell issue of JOC be without a Russell
paper. Issue No. 17, Volume 60 of JOC appeared on August 25, 1995, just two
days removed from Russell’s 70 birthday. The articles in the issue led off with a
paper by Russell, Wang, and Yao on “tert-Butylation of Quinolinium Cations and
Quinoline N-Oxides by tert-Butyl Mercury Halides” that was paper number 60 in
Russell’s long-running series on Electron Transfer Processes (35). There followed
19 consecutive articles, all dedicated to Russell for his 70th birthday. The first three
notes in the issue were dedicated to Russell as well. The writers were ex-students
or post-docs, friends, or academic colleagues, and the countries of origin besides
the US were Japan, Korea, Germany, Italy, Australia, Poland, Spain, and Canada,
testifying to Russell’s wide network of influence.

108
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 We also planned a Russell 70th birthday symposium at Iowa State, with the
Trahanovskys handling the details. The celebration at Iowa State, which took
place on Aug. 26, just three days after Russell’s birthday, featured talks by former
student Ed Janzen, former post-doc John Gerlock, friends Keith Ingold, Henry
Shine, and Cheves Walling plus Russell himself (Figure 12). At least 50 students
and friends with their spouses were at the function. The symposium was supported
by Ford, Chevron, Rohm and Haas, Amoco, and Bridgestone-Firestone plus Dr.
Christoph Osuch, Dr. David Lawson, Dr. and Mrs. George Holland, and Dr. and
Mrs. Kenneth Mattes. In retrospect, it was providential that Russell’s 70th birthday
was chosen for the celebration, for he would not see his 75th birthday.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 12. Glen Russell, Edward Janzen, Henry Shine, Keith Ingold, Cheves
Walling, and John Gerlock at Russell Symposium, Aug. 26, 1995. Photo by
Iowa State.

The year 1997 started out like any other. Russell still drove to work six days
a week in the old MG (Figure 13). Russell rebuilt the MG engine himself and
had the MG repainted in the exact shade of British racing green originally used
on the car. This was thanks to daughter Susan who had spent four years at the
London School of Economics and who brought MG parts and the green paint in
her carry-on luggage when she came home for visits. In June, Russell traveled to
Portland, OR, to help Susan put a deck on her house. That summer also marked
the birth of the first Russell grandchild on June 30, granddaughter Kylie (Figure
14), born to Russell’s chemist daughter June Russell and her chemist husband Jeff
Bjorklund. (Grandson Tristan Russell was born on July 17, 2000.)

109
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 13. Russell in the Chemistry Department Parking Lot with the 1952 MG.
Photo by Martha Russell

Figure 14. Russell with new granddaughter Kylie. Picture by June Russell.

Russell returned from a summer Gordon conference not feeling well. There
was fluid build up in his chest. A large amount of fluid was drained from his
pleural cavity, and it was found to contain cancer cells. We can testify that Russell
for most of his life was a chain smoker. He gave it up for a time, but he took up the
110
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 habit again when he was in Russia in 1975 as part of the USA-USSR Exchange
Program. Whatever the cause, he was diagnosed with stage 4 lung cancer. There
was little hope for successful treatment. He was cared for at home by his family
with the help of hospice, and he died on Jan. 1, 1998, age 72. The memorial service
took place on Jan. 16 at the Memorial Union at Iowa State with burial in the Iowa
StateUniversity Cemetery. Interestingly, his mother Marion Russell survived him
and lived to be 100.
Nobel Laureates H. C. Brown and Jean-Marie Lehn sent memorial tributes
to Russell’s friend Walter Trahanovsky, as did great chemists such as Cheves
Walling, George Hammond, Frederick Greene, Emanuel Vogel, Keith Ingold,
Joseph Bunnett, and Robert Norris. However, the tribute with the most resonance
for us is from his former colleague Orville Chapman (36).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

“When I think of Glen, the word that most frequently comes to mind is
silence. That’s odd, isn’t it? Some of you know that for sixteen years
Glen and I went out for coffee twice a day, six days a week. Sometimes
others joined us, but more frequently it was just the two of us. On frequent
occasions, we never said a word. We were content with each other’s
company, and we could share silence. How rare and wonderful is a
friendship so complete that it can share silence—share silence and yet
value companionship dearly. Of all the things I missed after leaving
Iowa State University, having coffee with Glen was the thing I missed
most. I enjoyed sharing the latest research results and even more the as
yet unexplained observation. Every conversation lightened with Glen’s
wry observations, his piercing wit, and his unforgettable epigrams of
pompous people. But as the ancient proverb says: “Words fitly spoken are
like apples of gold in pictures of silver, but silence is golden.” And so it is;
silence is golden. In the end, I value the silence—and the companionship.
Death has ended the conversations, but it can never end the silence or the
companionship.”

In view of his 50 years of research on free radicals, we thought appropriate

to have a Russell memorial symposium at a national ACS meeting. The officers
of the organic division were agreeable, and we also got a PRF grant to support
travel. The symposium “Free Radical and Electron Transfer Chemistry: A Legacy
of Glen A. Russell” was held on Mar. 28, 2000 at the ACS National Meeting
in San Francisco. The invited speakers were Kathleen and Walter Trahanovsky,
E. Thomas Strom, Hans Dieter-Becker, Martin Newcomb (Norris Award 2000),
Keith Ingold (Norris Award 1993), Charles Perrin (Norris Award 2015), Henry
Shine, Dennis Tanner, and Bernd Giese (Norris Award 2009). Russell’s family
were present at the symposium.
Thus far we have given you an overview of Glen Russell’s career, and it is
quite clear Russell was a brilliant physical organic chemist. But, how brilliant or
why brilliant? Edward G. Janzen is one of Russell’s best known students. Janzen
had a long academic career, first at Georgia and then at Guelph, and was a pioneer
of the spin trapping method of detecting free radicals. Here is Janzen’s view on
the matter (37).

111
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 “He was creative, yes, but there are a lot of creative people and some
are scientists. So that’s not it. He wasn’t particularly good at designing
experiments. You as a graduate student had to think these up yourself.
But—here is his brilliance. He had the uncanny skill at figuring out
what the results meant and where they would go if continued. And, most
importantly, he had the courage to stick his neck out and predict what
future results would be. I often thought more experiments were needed
to make the statements he made. But Glen said NO! We don’t need more.
He could sense what was really happening even though not all the results
were in. Some people have the “hunch.” Russell did.”
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

What was Russell really like? Orville Chapman’s tribute clearly showed
that he was a quiet man. What adjectives would you use to describe that—shy,
reserved, distant, taciturn, cold? We both thought he was a kindly man, so we
wouldn’t use the adjective “cold,” but we might come down on the side of “shy”
or “reserved.” We’ll give you input from his friends, former students, and his
daughters in the pages to follow, and then you can make your own assessment.
One thing we students had to get used to was that he would seemingly ignore
us when we passed him in the hall. Al Bemis remembers that he would say “Hi,
Dr. Russell” and get no response, although he was always friendly and interactive
to Al when Al saw Russell in class or in his office (38). Russell had the habit of
always walking down the right side of the hall, while dragging the knuckles of his
right hand against the wall. Some of the older graduate students, seeking to play
with his mind, would deliberately get between him and the wall. He would look
vaguely troubled, but he would say nothing about their interference. Ultimately we
concluded that he was deep in thought on those occasions. Shy people are rarely
dynamic lecturers, but Russell was a skilled and interesting teacher. Ed Geels
testifies that his lecture style was casual, but he kept the students up with him at all
times (39). His presentations at meetings were also clever and forceful. Jim Tanko
remembers that he would seem almost dozing during departmental seminars, but,
when the time for questions arose, he roared awake like a lion (40).
Russell was a chain smoker most of his life. Tanko states that when he was
particularly agitated and engaged during a conversation on chemistry, he would
sometime light up a second or even a third cigarette. He thought that perhaps the
number of cigarettes lit was a measure of his interest.
Tanko relates the following incident.

“And of course, there was the time he tossed the lit cigarette into the
trash behind him. It started smoking, and the grad student to whom he
was talking had so much admiration and respect for Glen that it took him
several minutes (and a rapidly decaying situation) to interrupt to say,
“Professor, I think we have a problem.” Glen put another trash can on
top to snuff out the fire and returned to the conversation as if nothing had
happened.”

112
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Geels (39) describes a lecture incident where Russell after finishing his
cigarette took the next one from the pack and placed the non-filtered end in his
mouth.

“Immediately the entire class noticed this as he sort of lost us as we waited

for him to light up the filter end. He kept us in suspense quite a while as
he kept on lecturing while even removing the cigarette from his mouth
while making a point about his subject. Finally he placed it back in his
mouth, removed his lighter from his pocket and proceeded to light the
filter end. The filter got very hot and mostly just melted. This stopped
his lecture, and he also noticed both his mistake and the amusement of
the entire class at this point. This produced a small smile on his lips. He
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

broke the filter from the end and relighted the remainder of the cigarette,
continuing his lecture.”

These two incidents aren’t so amusing when we remember that Russell died of
stage 4 lung cancer.
Russell’s students tended to get jobs in industry rather than in academia. This
may have been because he had developed many contacts through his years at GE
and from his consultancies with a variety of industrial labs, both oil and traditional
chemical companies. His previous industry experience resulted in a practice that
his students appreciated. He felt that, if you could buy a chemical needed for your
research, you should buy it. When you factored in the time it took to make it,
it was cheaper to buy it, no matter how expensive it might seem. Other faculty
members thought graduate student time was cheap. It wasn’t to the graduate
student. Also, Russell tried to get his students finished when he felt they had
completed a significant chunk of research. It was certainly possible to get a Ph.D.
in three years, and the usual time period was between three and four years.
Simply said, Russell respected the time and effort of the graduate students.
Jim Tanko (40) had come to Iowa State with the idea of working for another organic
faculty member. He had corresponded with that individual several times, so when
the time came to choose a research director, he went in to see the faculty member,
expecting to sign up. During the interview the faculty member took a personal
call and stayed on the line for some 30 minutes. This made a very bad impression.
Tanko had been taking a course from Russell and considered him an incredible
lecturer, so he decided to check out what Russell would have to offer. Russell
described about a dozen projects, when the phone rang. It was personal. Russell
said immediately that he was in conference and would return the call later, hung
up, and returned to discussing research projects. The point of Tanko’s story is that
Russell was respectful of people and particularly his students. As an example,
Russell’s office was covered with color photos that he had taken of his graduate
students and post-docs.
Fellow faculty member Walter Trahanovsky noted that Russell enjoyed
interacting with visiting chemists (41). He was very good at summarizing
seminars, and Trahanovsky recalls that he felt he got more out of Russell’s
summaries than the presentation itself. Russell played a major role in establishing
the Gilman Lectures, an endowed lectureship in honor of Henry Gilman. He

113
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 worked very hard to insure that early lectures were presented by the very cream
of organic chemists. Trahanovsky remembers the time an invited chemist turned
down the invitation to be a Gilman Lecturer. Russell wrote this chemist and asked
him to accept the invitation for next year. He pointed out that the only other
person who had turned down the invitation had passed away within a year. The
invitee replied, accepting the invitation for the next year.
Ron Blankespoor recalls with pleasure that Russell was not a micromanager
(42). He was always available to talk about research results and future
experiments, but he wasn’t looking over your shoulder in the lab. Tom recalls that
he was open to your research going in different directions from what had been
envisioned. Above, Janzen had said that you had to design your own experiments
(37). Tom would take the view that Russell was always available to brainstorm
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

with and would help you get started, but then you had to find your own way. Isn’t
that what is supposed to happen, as a graduate student makes the transition to
becoming an independent researcher?
If Russell felt that you were deserving of a Ph.D., he would protect you on
your way through the process. Some background is necessary here. Nowadays
the two main ways of determining whether a student should progress toward a
Ph.D. are (1) defending a research proposal, either your own or an independent
proposal, or (2) passing a set number of cumulative exams over a certain time
period. In those days at Iowa State, the process resembled an obstacle course.
First of all, on entrance to graduate school, you took qualifying exams in the four
areas of chemistry. Back then, biochemistry wasn’t one of the group. If you failed
a qualifying exam, you had the option of taking a refresher course, and then taking
the exam again. Obviously there was course work, and at that time a B was still
considered a pretty good grade. Course work was not only in your major, but in
your inside minor and in your outside minor. Then came written prelims in your
major, your inside minor, and your outside minor. These were followed by an oral
prelim in which the oral committee consisted of your major professor, a second
professor from your major, a professor from your inside minor, a professor from
your outside minor, and one other faculty member.
In contrast to the oral defense of the Ph.D. thesis, which was a love fest, it
was entirely possible to blow your degree in the oral prelim. To show the stakes,
consider two graduate students from our time—student P and student D. Student P
went into the prelims with fear, because he knew that his major professor felt he had
not done enough research to advance. However, student D was taking the prelims
at the same time. Student P reasoned that if he outscored student D on the written
organic prelim, he would be OK, for student D was a Gilman student. Gilman
students who advanced to the prelims always passed, probably in compensation
for the fact that they would spend two more years for their Ph.D. than the other
organic students. However, the faculty were too smart for student P. Both student P
and student D received grades of “condition” on their written exams. This meant
that their advancement to the Ph.D. was conditional on their doing well on the
oral prelim. Of course, student P failed his oral, and student D passed his oral.
However, there was a happy ending. Student P was allowed to take prelims again
the next year, after he had done enough research to satisfy his professor. This time
he passed and progressed to the Ph.D.

114
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 We graduate students were concerned about how our fellow student J would
do on the oral prelim. By chance, the oral was scheduled on the very day the
winners of the 1963 Nobel Prizes in chemistry were announced. They were Karl
Ziegler, honored for work on low pressure polymerization of ethylene, and Giulio
Natta, honored for synthesizing stereoregular polypropylene. (In the interests of
historical accuracy, we should interject that polypropylene had been made several
years earlier than Natta’s synthesis by J. Paul Hogan and Robert Banks of Phillips
(43), a fact normally ignored in the usual polymer chemistry textbook.) Russell
showed up at student J’s desk early in the morning, telling student J that Ziegler
and Natta had won the Nobel Prize for chemistry, and it might be a good idea
to know something about their research. Russell then gave the student several
reprints of Ziegler and Natta’s work. Student J’s oral prelim was in the afternoon.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Russell opened the proceedings by noting they must be in mourning at Harvard

today. The other organic committee member, Orville Chapman, inquired why.
Russell then responded that R. B. Woodward had once more been overlooked for
the Nobel Prize. (Woodward would win in 1965). Chapman then asked who had
won the prize. Russell responded that it was Ziegler and Natta. Chapman took the
bait and opened the oral by asking, “Student J, what do you know of the work of
Ziegler and Natta?” And, of course, he knew a lot, so the rest of the oral prelim
went smoothly.
By all accounts, Russell was a formidable bridge player. If a visiting speaker
was a bridge player, certain skilled graduate students would be invited to the
Russell home to make a fourth in bridge. Neither of us had the skills at bridge
to qualify for such an invitation.
Russell had the reputation of being able to drink anyone under the table. This
was not a contest that either of us wished to enter, but we always wondered whether
this perception was true. Information from Russell student Kirk Schmitt gives
us some hard data on the matter (44). Kirk tells of a memorable Russell group
Christmas party.

“I was used to beer from a small boy, but no one warned me that Martha’s
rum punch was, well, rum and that the “punch” ingredients were mainly
an after-thought. Nevertheless, on my first such party I was doing well
and made it to midnight. I was actually proud that I had done better than
a Polish post-doc who had tried to go head-to-head with Glen, drinking
“wodka” from water glasses. It was reported that Glen winked when the
post-doc said “I am the Polish embassy.”

However, at midnight, Glen got out what he considered an appropriate

nightcap, the Greek liqueur ouzo. He poured us all a shot glass, straight,
and I took mine while leaning on the bar. I had never drunk the stuff
before but had some experience starting siphons for gas and diesel by
mouth and was pretty sure I recognized the vile flavor. As I sat dipping
my tongue into the ouzo, trying to figure out how to get rid of it, Glen
leaned across the bar, looked me in the eye, and said, “You know, no real
man sips his ouzo.” Well, bottoms up. I wish I could report more about
the party, but I only remember waking up the next morning.”

115
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 On this basis of this information, we conclude that the perception was probably
true.
As noted above, the Russell family sport was skiing. His daughters date those
trips back to the ‘60’s. Russell regularly took graduate students on weekend winter
ski trips to Minnesota. Probably an early exposure to heavy duty skiing came
courtesy of Keith Ingold (45). Keith was surprised when Glen expressed a desire
to ski the Expo downhill run at Mont Tremblent in Quebec. The run had opened
recently and had an awesome reputation. Glen came to the National Research
Council in Ottawa to give a talk, and Martha accompanied him. The couple stayed
at Keith’s house, and Keith, his wife Cairine, Glen and Martha went to Tremblent
for a three day stay after Glen gave the talk. Here is Keith’s description of the
skiing trip.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

“I led Glen to the top of Expo, a 1600 foot hair-raising slope with no easy
sections—and no escape routes to easier slopes. Expo is just steep and
steeper! Glen looked a bit aghast, gulped, then grabbed his resolution
(and his poles) in his hands and pushed off. Seconds later a cloud of
snow erupted 80-100 feet down the slope as Glen had a spectacular “egg-
beater” fall with arms, legs, poles, and skis flying everywhere. Fearing
multiple fractures, and even worse, for Glen was lying unmoving, I skied
down to him. There he was lying in the snow with a silly grin on his
face. To my anxious questions he said only: “That was fun!” He got up,
brushing off my concerns (and a whale of snow) while I gathered up his
poles and skis. Then he was off again for another egg-beater fall after
another 100 feet or so. His “take off and fall” skiing technique did get
him down Expo in one piece to my relief. —When we got to the bottom,
I thought he’d be through skiing for the day. Not a bit of it, we had to
return immediately to the top of Expo for another descent, then again,
and again, and again, until the lift closed. —However, his last two runs
on that first day were made without a single fall.

The favorite companions of the Russell clan on skiing trips were his former
post-doc John Gerlock and John’s wife Sandra. With Russell’s encouragement
John and Sandra became pretty good skiers. John tells the tale of the group skiing
at Jackson Hole (46). After the first day, two to three feet of powder snow fell on
the place. Russell advised John to rent “noodles” for skiing that day, and John did
so. “Noodles” are long, flexible, difficult-to-control skis. Russell and John went to
the top, and John started down on his noodles. With his first time on noodles, John
would ski only a few feet, fall, and then reproducibly repeat the process. Russell
told John to ski on down as best he could, and he would follow.

“After I floundered about halfway down the slope, I looked up slope and
saw Russell shooting down the hill at great speed. —It was really an
awesome thing to see. —Suffice it to say that when he blew by me spraying
powder, he looked better than Jean-Claude Killy ever did right up to the
moment that he became airborne, summersaulted and disappeared in a
huge cloud of powder. When I reached him, he was bleeding profusely

116
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 from the mouth. —It was obvious to me that we needed a helicopter to
medevac us to safety, but that was not what happened. Spitting blood
while speaking, just like in the movies, Russell told me in no uncertain
terms that he was going to ski down and I should follow, which I did.”

John states that Russell got a bunch of stitches in his lip. When they had margaritas
afterward, with the usual heavy salt layer on the rim of the glass, Russell drank his
ice-cold, salt-laden margarita through the gash on his lip without flinching. He
was just that tough.
Tough also characterizes the way he carved out his area in research. He
defended his turf strongly. He was very thorough in his reviewing of articles
and research proposals. Occasionally we would have the opportunity to see his
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

reviews of papers. He was a tough reviewer, and sometimes he crossed the line
from “tough” to “nasty.” Yet overall, he was a very generous man, as Tom can
testify.
During my last year with Russell, I had an offer in hand from Mobil Research
and Development Corp. in Dallas. I also had a two year ROTC commitment with
the US Army. Mobil had agreed to keep the position for me while I was in the
army. My Army duty was due to start in March, 1964. I knew that my Army
salary would be a lot less than what Mobil would pay me, so I tried to finish my
thesis early to get as much Mobil money as possible. I had it worked out that I
could start with Mobil Dec. 1, 1963. In the early fall Russell came to me and said
that it would be a pity if I couldn’t spend the Christmas holidays with my Iowa
family. He then offered to pay me a post-doc’s salary for the month of December.
I took him up on his generous offer. I’d like to say that I got a lot of research done
during that extra month, but I didn’t. However, my wife and I did get to spend
Christmas and New Years with our families. Ed Geels (39) tells of Russell putting
him on a post-doctoral appointment for six months after the Ph.D., while Ed was
waiting to start his appointment at Dordt College. When Ron Blankespoor (42)
received his first academic position, he had no research funds to carry him over
the summer. Russell gave him a post-doctoral appointment for that first summer.
The picture that most of us have about a successful academic researcher is
that he/she is a driven individual, who works the students hard and himself/herself
even harder. His daughters testify that Russell was a devoted family man. He
would come home soon after 5 p.m. on weekdays and noon on Saturday. His
daughters remember hearing the sound of the MG’s motor before he ever turned
into Murray Drive. He seemed to almost effortlessly draw a balance between
work and non-work. He left his work behind in Gilman Hall and focused on his
wife and daughters and the recreational activities that all would enjoy. He bought
a trampoline for his daughters before they could barely walk. When they were
young he tried to teach them French from an old Berlitz language book. Nearly
every summer the Russell family went abroad for a month, one week at a chemical
conference and three weeks as a family, traveling and exploring new places. He
was an avid photographer, and he took many photographs of their travels. Overall,
he had incredibly diverse interests.
Russell never lost sight of his farm boy upbringing, and he always relished
working physically with his hands. As soon as he got home in the evening, he

117
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 would change into his work pants and garden, work in the yard or greenhouse,
or complete various house projects, often incorporating his daughters into those
tasks.
The greenhouse was the site of Russell’s orchid collection. When the family
was away on travels, Kirk Schmitt (15) states that certain trusted graduate students
or post-docs were allowed to house sit. Post-docs Gerlock and Balleneger were
trusted to house sit during a bad winter when the heater in the greenhouse failed,
killing Russell’s beloved orchids. Kirk remembers helping restore some of the
plumbing and musing whether the research budget could be diverted into getting
replacement orchids before Russell returned.
The summing up of Glen and Martha Russell’s characters and relationships is
best done by their daughters (47).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

“Our parents were a good match. They complemented one another and
helped each other shine. They were both very social, each in their own
ways, and generous with their time. They loved opening their home to
students and visiting faculty through the rousing parties they hosted.

Our parents were also politically engaged and very mindful of the
opportunities they had and their good fortunes in life. They focused
on doing what they could to ensure that others could achieve similar
outcomes, not only their students, but the community in general.

We were very lucky to have parents that were so positively and actively
engaged in our daily lives yet at the same time always served as strong
role models on the bigger issues, inspiring us to pursue our goals and
visions fairly and to keep a healthy and balanced outlook on life. Our
parents were always there for us, in the little and big ways, and a day
doesn’t go by when they are not in our thoughts.”

Some later chemical historian can measure Russell’s impact on science by

simply reading through and evaluating his many publications. We will do some of
that ourselves in the next section. But, what was Glen Russell really like? Thirty
years from now it is doubtful that there will be anyone living who really knew Glen
Russell. These preceding pages are meant to show something of what Russell was
really like. He was our professor, our mentor, and our lifelong friend. He was
brilliant yet considerate, listened to our research ideas, and helped us think for
ourselves. His was a life well-lived, and it had a powerful influence on our own
lives. We believe that it also had a powerful influence for the better on the course
of free radical chemistry.

Russell’s Research
Glen Russell’s first article was published in 1952 (9), and the last publication
we can find has a date of 2001 (48). His free radical research begins in 1949,
when he began working with Brown, so his research career was around 50 years.
118
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Those 50 years pretty much coincide with the last half of the 20th century. Russell
had well over 200 publications. We will be selective instead of comprehensive in
our coverage of his research. We will focus on the main themes. Those themes
originated early on and continued throughout his career. Furthermore, one can
readily discern those themes, because the titles of his articles would often include
a classification with a number. For example, a 1998 posthumous publication
was titled “Electron Transfer Processes. Part 63. Reactions of p-Nitrobenzyl
Halides with Dialkyl Phosphite Anions in Dimethyl Sulfoxide.” (49). His series
on semidiones was numbered into the 40’s.
We will concentrate mainly on the early research. We will give a little more
detail about the projects on which we worked. This is not because we think our
research was more significant than that of any other Russell student, but the usual
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

journal article is linear, wherein the workers start a project and carry it through
successfully via obvious steps. End of story! Researchers know this is far from
the usual case. There are all sorts of starts, stops, and side steps on a project, and
the results often come from luck rather than design. Like the late Paul Harvey, we
will give you “The rest of the story!”

Early Research at Purdue and GE on Solvent and Polar Effects

on Radical Reactions
Glen Russell’s first paper (9) made it into a significant textbook quite soon.
In the first edition of the classic text by Morrison and Boyd, the work of Brown
and Russell on the chlorination of a deuterium-labeled isobutane was cited via an
illustrative problem as demonstrating that free radicals had very little tendency to
rearrange (50). In the second edition of the text, this result was expanded into
a three page section about the lack of rearrangement of free radicals and on the
utility of isotope labeling to study mechanisms (51). This discussion of the work
of Brown and Russell continued to appear in later editions of the textbook.
That first paper from Russell’s thesis was designated as Part I in the series
“Directive Effects in Aliphatic Substitution” and told of his intent to do further
investigations in “this little explored field.” Three additional papers from his thesis,
now with Brown as the second author, were numbers III, IV, and V in that series
(10–12). (We have not been able to find a part II.) They described further studies of
chlorination and bromination substitution reactions. They pointed out the need to
consider polar effects on the transition states and intermediates in the free radical
reactions and not just the strength of the C-H bond being broken. The bromination
papers (see below) with DeBoer and Desmond in 1963 were numbers XVIII and
XIX in the series.
At the General Electric laboratories, working with Frank Mayo, Russell
became involved in the study of liquid phase autoxidation of hydrocarbons.
These reactions are of great importance because many industrial processes are
based on them. The mechanisms are especially difficult to study. This work
shows Russell’s ability to choose elegant experiments which simply but critically
distinguish among mechanisms. He continued to publish under the Directive
Effects heading. Paper VI, submitted from GE in early 1955, was a study of the
119
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 competitive oxidation of cumene and tetralin, with Russell as the sole author
(52). Here he observed that tetralin can serve as a retarder in the oxidation
of cumene. Think about that for a minute. For the same amounts of initiator,
tetralin is much more oxidizable than cumene, yet Russell found that small
amounts of tetralin could actually act as a retarder. Russell showed that there
was widespread generality to such an effect. Furthermore, he was able to unravel
the complicated kinetics associated with four competing propagation reactions
and three competing termination steps, an analysis that Keith Ingold regarded
as an intellectual tour de force. He also concluded from rate constant data that
the secondary peroxy radicals from tetralin terminated 500 times faster than the
tertiary cumene peroxy radical. Prior to Russell’s work it had been assumed
that all peroxy radicals interact with one another at the same rate to terminate
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

oxidation chains.
A series of papers in 1956 under the heading “Oxidation of Unsaturated
Compounds” began with two papers by A. A. Miller and Mayo on the oxidation
of styrene (53, 54). Russell was the sole author on papers III and IV in that
series, which examined the oxidation of indene with oxygen (18, 19). These
papers were followed by two additional papers on oxidation kinetics by Russell
alone (20, 21). The latter of the two looked at the relative reactivities of a series
of aralkylhydrocarbons toward peroxy radicals. Data were tabulated for 32
hydrocarbons having a 1000 fold difference in reactivity. The differences were
attributed to differing radical stability but also to polar effects on the transition
states of the propagation steps. A communication in Chemistry and Industry
in 1956 (55) and Paper IX in the Directive Effects series, published in 1957
(56), described results of a study initially presented at an ACS meeting in the
fall of 1956. The conclusions from that study were a major contribution to
understanding autoxidation chemistry. Russell suggested that the termination
reaction of secondary peroxy radicals involved two radicals coming together, most
likely in a cyclic transition state or an actual intermediate, which then collapsed
to the observed products: ketone, alcohol, and oxygen. This elegant study using
deuterium-labeled cumene and ethylbenzene explained why secondary peroxy
radicals terminated much more rapidly than tertiary peroxy radicals, where there
was no possibility of transfer of an alpha H. The postulated mechanism is now
known as the Russell mechanism (Figure 15).

Figure 15. The Russell Mechanism.

120
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Russell also continued to illuminate aspects of free radical halogenation
while at GE. He was able to show that confusing data about relative rates of
hydrogen abstraction by various radicals could be understood in terms of a theory
invoking polar effects in the transition state for hydrogen abstraction. He made
the dramatic discovery that different solvents can drastically alter the position of
attack of chlorine atoms on a branched chain hydrocarbon. This was first reported
in a communication (22) and then followed up in a series of papers (57–62).
Russell observed that electron rich aromatic solvents had the ability to change the
orientation of substitution and suggested that this was due to formation of a pi
complex with the Cl atom. The complexed atom is then able to be more selective
in abstracting hydrogen atoms, favoring tertiary over primary halogenation to
a greater extent than was observed for an uncomplexed Cl atom. Solvents like
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

carbon disulfide could form sigma complexes with the Cl atom, which also
modified the selectivity of substitution.
Russell’s lecture entitled “Solvent Effects in Free Radical Reactions” at the
Reaction Mechanisms Conference in 1958 created a great stir. Before that time,
many texts and references stated that the absence of solvent effects was strong
evidence for a free radical mechanism. Russell was one of a very few industrial
chemists to have been invited to speak at the Reaction Mechanisms Conference,
and he was invited to lecture again at the 1966 Conference. He was only the
third person at that time to have been invited twice. The other two were George
Hammond (Norris Award 1968) and Paul Bartlett (Norris Award 1969).

Early Work at Iowa State on Solvent and Polar Effects on

Radical Reactions
Russell would continue to explore solvent and polar effects on radical
reactions during his early years at Iowa State. In 1963 Russell and Bob Bridger
published a monumental study of the reactivity of a large variety of C-H bonds
toward the phenyl radical, a classic that has become a standard in the field of
homolytic substitution (63). This was Part XXIII in the Directive Effects in
Aliphatic Substitution series. Sadly, Bridger died suddenly after a short time at
the Mobil Research Laboratories in Princeton, NJ. Dale Hendry, another early
member of the group, did extensive studies of the solvent effects in oxidation
and chlorination reactions (64, 65). He also passed away at a young age while
working at Stanford Research Institute. Additional work in this area was also
carried out by Russell student Roger Williamson (66, 67).
Russell reviewed much of the work on free radical substitution reactions in
a comprehensive chapter in the book Free Radicals, edited by Jay Kochi (68),
who had worked with Hammond in the same labs that Russell used at Iowa State.
The chapter is titled “Reactivity, Selectivity, and Polar Effects in Hydrogen Atom
Transfer Reactions” and contains 220 references to research done through 1971.

121
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The NBS Story
As mentioned earlier, in the fall of 1960 the Russell group was all male. I
(then Kathleen Desmond) arrived in Ames that fall with the intention of getting
a Masters degree. All the forms had only two categories for MS students, MS
Prerequisite for those in academic difficulty, MS Terminal for those who failed
courses or prelims, and would not be allowed to go on for the Ph.D. I crossed
those out and added MS By Choice whenever I could. The idea of five years in
Iowa was not appealing. Fifty-five years later I am still in Ames.
I was assigned to Russell as an advisee and chose to stay in his group.
When it was time to start a research project, Russell suggested that a study of
the reactivity of the succinimidyl radical (NS), the presumed chain carrier in free
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

radical bromination by NBS (N-bromosuccinimide), would be an appropriate

masters project and would fit in with the other radical reactivity studies being done
in the group. At the same time a bright undergraduate, Chuck DeBoer, had signed
up to do research with Russell. He was placed in a different lab and given another
well-defined problem to work on, the relative reactivities of benzylic hydrogens
in photobromination, also a free radical substitution reaction. At that time, the
mechanism of the NBS reaction was widely assumed to involve abstraction of an
allylic or benzylic hydrogen by the succinimidyl radical as the rate determining
step, with chain propagation following by the aralkyl radical reacting with NBS
to give the brominated product and a new NS radical.
At my small undergraduate school, Emmanuel College in Boston, there had
not been an opportunity to do laboratory research, so my experimental skills were
not strong. Initially my “big brothers” in the lab would check to see if my clamps
were tight enough or my rate of heating things was appropriate, but they soon
relaxed and let me work on my own. The NBS reaction was very simple. The
reaction is initiated by visible light, so all I needed was an ordinary light bulb. The
solvent we used, methylene chloride, boils at 40 C., and refluxing in it provided
a constant temperature. The insoluble, solid NBS is more dense than the solvent,
and the insoluble, solid succinimide is less dense than the solvent. Consequently,
the progress of the reaction is visually apparent as the solid on the bottom dissolves
and a new layer of solid appears on the top. The relative rates of disappearance
of reactants in competitive reactions were calculated using data obtained from gas
chromatography. A planimeter was used to manually measure the areas of the
peaks before and after reaction, and these were related to areas from standard
solutions to get initial and final concentrations of the competing hydrocarbons.
It sounds primitive now, but the results were highly reproducible. The biggest
challenge was sharing the one planimeter in the group.
DeBoer and I used the same experimental approach to our respective
problems, i.e., the study of the relative reactivities of toluene, ethylbenzene,
cumene, and diphenylmethane in competitive reactions, but we worked in
separate laboratories and had limited contact with each other. It was customary
for students to rotate presentation of research results at weekly group meetings.
When the time came for Desmond and DeBoer to present, much to our surprise,
we reported relative reactivity numbers that were essentially the same for both
NBS and bromine. This meant that the rate determining step in both cases

122
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 probably involved hydrogen abstraction by the bromine atom. Apparently even
traces of moisture or HBr, which is very hard to remove from the NBS, can lead to
the formation of very small amounts of molecular bromine, and photobromination
is then initiated. The HBr formed in each H abstraction can then lead to more
bromine and propagate the chain reaction.
My college background may not have been heavy in experiment, but I was
taught the value of a good literature search. So, I looked back at earlier studies
of NBS bromination. The succinimidyl radical pathway had been proposed by
Bloomfield (69) as a reasonable explanation of the reaction. It looked quite
sensible on paper, but there was no direct evidence for a succinimidyl radical.
A bromine atom mechanism had been suggested by Goldfinger (70), but most
researchers and textbook writers ignored that possibility and continued to assume
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

that the succinimidyl radical mechanism was correct. Our conclusion made
even more sense in light of studies which showed that in reactions of alkenes
with bromine in the presence of light, very low concentrations of bromine
favored allylic substitution, a probable free radical substitution, while higher
concentrations lead to addition (71, 72).
At a meeting, Russell discovered that J. C. Martin, a young assistant professor
at Illinois, was doing a study of the relative reactivities of a series of substituted
toluenes, with NBS and bromine. His results were also leading to the conclusion
that bromine atom was the hydrogen abstracting agent. Russell and Martin agreed
to submit articles for simultaneous publication. Communications to JACS were
published in early 1963 (73, 74). The full JACS papers later that year were part of a
series where the lead article was by Cheves Walling and co-workers, who were also
studying NBS reactions (75–78). Walling stated that their results agreed with other
recent work in supporting the conclusion that a Br atom chain is the most plausible
mechanism for NBS bromination. In a footnote in his paper, Walling said, “We
wish to acknowledge that Martin’s and Russell’s results were made available to
us prior to publication and led to revision of some of our tentative conclusions.”
Since that time, all major organic chemistry textbooks have presented the Br atom
mechanism as the way to explain benzylic free radical bromination of aralkyl
hydrocarbons by NBS.

Oxidation of Carbanions
Russell’s second and third published papers were submitted from GE and dealt
with the oxidation of anions (79, 80). The second of these was a study of the
oxidation of 2-nitropropane in basic solution. There had been suggestions that
some carbanion oxidations had a free radical nature, but they often occurred too
rapidly for mechanistic studies. Russell thought that use of a resonance stabilized
anion would moderate the rate and allow for closer study. He concluded that
the reaction did involve a free radical mechanism proceeding by an ion-radical
chain with a hydroperoxide intermediate. He observed autocatalysis, an induction
period, and inhibition by standard free radical inhibitors, all hallmarks of free
radical chain reactions. His suggested mechanism proposed a one electron transfer
123
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 between a peroxy radical and the resonance stabilized anion to start the chain
process (Scheme 1).

Scheme 1. Russell’s suggested chain reaction.

Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Tony Moye was Russell’s first graduate student at Iowa State, and he took
up the study of the oxidation of resonance stabilized carbanions along with post-
doc Nagpal (81). It was difficult research, because radical anions were rapidly
quenched in most solvents; and the evidence for free radicals was indirect. By the
time Moye was leaving in 1960, two remarkable advances had become available:
one in instrumentation, ESR; one in solvent chemistry. DMSO, otherwise known
as dimethyl sulfoxide.
Edward Janzen continued the study of base catalyzed oxidations after Moye’s
departure and had constructed a device to maintain constant oxygen pressure
(82). Ed relates that Russell had returned excited from a conference in California,
where Donald Cram reported spectacular results with a new solvent called DMSO
(dimethyl sulfoxide) (83). The compound was not yet commercially available,
but Ed remembered seeing an advertisement from Crown Zellerbach offering free
samples of the compound. When the compound arrived and was used by Janzen,
his oxidations went way too fast to be followed with his equipment. He finally
moderated the process by using a mixture of 80%-DMSO/20%-t-butyl alcohol,
with potassium t-butoxide as the base. This allowed for kinetic measurements,
and this mixture was used by many Russell students for ESR experiments. Some
of this oxidation work appeared first as a communication (84). This publication
also reported initial work with the anion derived from DMSO. The methyl sulfinyl
carbanion had been reported by Corey and Chaykovsky (85), but the carbanion
had been synthesized independently and likely earlier by Janzen. Much of this
early carbanion oxidation work plus some electron transfer work appeared in
1965 in a long article in the ACS Advances in Chemistry series (86). This topic
was revisited in 1966 (87), 1967 (88), and in another long ACS Advances in
Chemistry article in 1968 (89).

The Semidione Story

The radical anions for which Russell coined the name “semidione”, anion
radicals of alpha diketones, were studied as part of Russell’s program on electron
transfer reactions. Electron transfer deserves a section of its own (see below), but
so do semidiones. I (Tom) made my big discovery in August, 1963, just when I was
124
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 finishing up the writing of my thesis. As a consequence, my thesis was expanded
by about a third.
The electron transfer reactions studied involved transfer of an electron from
an anion or dianion to an electron acceptor. The progress of electron transfer was
monitored by ESR by measuring the growth of electron acceptor anion radical .
Using present day language, you would want your electron acceptor to have a low
energy LUMO and for the ensuing radical to be stable for quite a while. Quinones
would certainly be expected to be good electron acceptors in view of the stability
of their one electron reduction products, semiquinones (Figure 16). However, I
was against using the alpha diketone cousin of the quinone as an electron acceptor,
because of a prejudice I had developed during my literature search. I found
in a paper that researchers had reduced diacetyl (2,3-butanedione) under basic
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

conditions and found they had formed 2,5-dimethylbenzosemiquinone, clearly

from a dimeric condensation. I jumped to the conclusion that any anion radical of
a 1,2-diketone that had alpha hydrogens would probably undergo condensation
reactions. Nevertheless, I did synthesize a number of acyloins (alpha hydroxyl
ketones). It seemed that oxidation of the acyloins in DMSO might give the anion
radical of the alpha diketone as an intermediate. I was unsuccessful in most of
those attempts; and I still don’t know why, because we now know that acyloins
are the prime intermediate from which to form semidiones. Following my belief
that alpha hydrogens should be avoided, I did synthesize the appropriate acyloin
lacking alpha hydrogens, 4-hydroxy-2,2,5,5-tetramethyl-3-hexanone. Oxidation
in the 80%-DMSO/20%-t-butyl alcohol solvent gave the anion radical of the
di-t-butyl alpha diketone (Figure 17). Later on, I was able to optimize conditions
to synthesize purely alkyl semidiones containing alpha hydrogens (90).

Figure 16. One electron reduction of p-benzoquinone to give the semiquinone.

Figure 17. Oxidation of the acyloin to give the di-t-butyl semidione.

125
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 If one wants to shun alpha diketones with alpha hydrogens, then the obvious
thing to do is to substitute them with benzene rings. There were precedents to
indicate that such compounds might undergo electron transfer reactions to give
anion radicals. Michaelis and Fetcher had clearly shown that the purple colored
intermediate in benzoin-benzil mixtures was the benzil anion radical (91). Buried
in the experimental details from an early Fraenkel group publication was the
information that they had synthesized the benzil anion radical (92). About the
time of our work, the Fraenkel group published an ESR paper on benzil anion
radical (93), while Luckhurst and Orgel published ESR data on benzil anion
radical and the di-t-butyl semidione (94). My literature search indicated that
the heteroaromatic analogues probably could be oxidized to give free radicals.
Fischer found that 2,2′-furoin gave a blue color (95), while Cardon and Lankelma
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

observed that 2,2′-thenoin gave a green color (96). When I carried out ESR
experiments on these systems, the anion radicals from these two systems were
observed (Figure 18). This work was published several years later as Part II of
the Russell Semidione series (97).

Figure 18. Anion radicals (semidiones) of 2,2′-furil (left) and 2,2′-theril (right).

I finally overcame my prejudice against alpha hydrogens when Aldrich

Chemical Co. offered a new compound for sale that was an acyloin,
2-hydroxycyclohexanone. I reasoned that the structure of the likely product,
the 1,2-dione, was unlikely to give an alpha condensation, so it was worth a
try. I used our normal solvent mixture of 80% DMSO/20% t-butyl alcohol with
potassium t-butoxide as the base, and I was rewarded by seeing a 1:4:6:4:1
quintet, with a 9.82 gauss splitting, The five peaks clearly were from the four
alpha methylene hydrogens in the anion radical of 1,2-cyclohexanedione. I
bought 4-t-butylcyclohexanone to see what would happen if the ring were locked.
In this case, since there was no longer an oxygen alpha to the carbonyl, I bubbled
oxygen into the system to ensure a surplus of oxygen. The ESR spectrum obtained
consisted of seven peaks from two kinds of alpha hydrogens, with splittings of
13.10 gauss from two hydrogens and 6.55 gauss from two hydrogens. Note that
their average is equal to the splitting from the unsubstituted compound. These
two spectra are shown in Figure 19. Any reasonable picture of hyperconjugation
dictates that the larger splitting must come from the pseudo-axial protons and the
smaller splitting from the pseudo-equatorial protons. Of course, the t-butyl group
removes the symmetry element from the radical, so those two axial-type protons
can’t be exactly equal nor can the two equatorial-type protons. However, the
126
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 resolution of the system wasn’t sufficient to pick up these differences. I applied
the Heller-McConnell equation (98) to calculate the angle of these protons with
the free radical site, getting a value of 13 degrees for the axial hydrogens. This
was close to the value calculated for the axial bond angle by nmr for cyclohexene
(99), suggesting that the cyclic semidiones might model the corresponding cyclic
alkenes.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 19. ESR spectra of (a) 1,2-cyclohexanesemidione and (b)

4-t-butyl-1,2-cyclohexanesemidione.

Because of this encouraging result, I ordered every cyclic ketone that

Aldrich had on hand. I studied the five, seven, eight, nine, ten, twelve, and
fifteen membered rings. All of them appeared to give semidione radicals. They
all seemed to have alpha methylene splittings, and some, not all, showed two
kinds of alpha protons, consistent with one favored conformer. This work was
published as a communication with the radicals called alicyclic semiquinones
and with the conformational aspects emphasized (100). Russell later called this
his first semidione publication, although he had not yet coined the name. I also
had found that acyl benzenes could be oxidized to give phenyl alkyl semidiones.
The paper from that work received the designation Semidiones I (also Reactions
of Resonance Stabilized Anions XXI) (101). The series reached its climax with
paper No. 44, published in 1985 (102). Some of the early ways of forming these
radicals are summarized in Figure 20. (The symbol BP stands for potassium
t-butoxide.) Russell reviewed much of this work at a fairly early stage (103, 104).
We will not cover all those 44 publications. Instead we will briefly mention
four interesting paths. Any simple HMO calculation for a semidione will give a
significant bond order for the carbons in the LUMO. Russell and Stephens found
that aliphatic semidiones occurred in both cis and trans forms (105). As would be
expected, the trans was favored, but the balance could be shifted toward the cis via
ion pairing.
127
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 20. Early synthetic methods for forming semidiones.

With the importance of many steroidal ketones, this was an area ripe
for exploration by ESR. A very talented post-doc named Erach Talaty joined
the group. His research was part of a new Russell series called “Application
of Electron Spin Resonance Spectroscopy to Problems of Structure and
Conformation.” Russell and Talaty explored simpler fused ring systems such
as decalones as well as steroids (106, 107). They were able to obtain useful
information on the A/B junction, and they later investigated aspects of the D-ring
ketones (108).
Bicyclic semidiones proved an additional fruitful area for research. Russell’s
student K.-Y. (Leo) Chang initiated this work (109, 110). In my original
communication, I had reported a 2.66 gauss splitting from the oxidation of
camphor, which appeared to come from three protons. I guessed it was from the
7-syn-methyl group, but I was wrong. Leo found that this splitting had come from
the single bridgehead proton plus the two exo protons from across the ring, all
three protons accidently having the same splitting constant. The surprising result
was the large splitting from the exo protons. There are three single bonds between
exo protons and the radical site. The exo protons, however, are perfectly aligned
for a W or zigzag arrangement, which transmits the spin information across the
gap. Figure 21 shows the results from the simpler bicyclo[2.2.1]heptane-2,3-
semidione (109). Note the large 6.35 gauss splitting constant from the 7-anti-H,
also in a W alignment with the radical site. Russell extended this work to many
different bicyclic systems.

128
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 21. Proton splitting in bicyclo[2.2.1]heptane-2,3-semidione.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

The conditions under which I synthesized that first group of cyclic semidiones
left much to be desired. My use of large amounts of oxygen caused line
broadening when the paramagnetic oxygen molecule underwent spin exchange
with the semidione. My results for the 5- and 6-membered rings hold up fairly
well, but things become more complicated from the 7-membered ring on up.
Long range interactions can start to show up, and for very large rings both cis
and trans semidione isomers can exist. For example, consider the 7-membered
ring. I found splittings of 6.70 gauss and 1.97 gauss from two pairs of protons.
I interpreted the spectrum as coming from two axial-type alpha hydrogens and
two equatorial-type alpha hydrogens in a single preferred conformation. Well,
the large splitting came from two axial-type alpha hydrogens all right, but later
work showed that the 1.97 gauss splitting came from two equatorial-type beta
hydrogens! Go figure. Russell soon realized the potential complications present
in the conformations of large rings, so he started the systematic reinvestigation of
all of these systems. Russell, Keske, and coworkers looked at the 7-membered
ring system again (111, 112). When they carried out deuterium exchange reactions
on the alpha hydrogens, they found that the 1.97 gauss splitting (now refined
to 2.05 gauss) remained. They deduced that this splitting came from the beta
equatorial hydrogens, now aligned in a trans arrangement which allowed for a
strong W-type coupling. Their conclusion was that the seven-membered ring
system had the pseudo-chair conformation.
Russell and coworkers next investigated the 9-membered ring semidione (113,
114). They found that the parent compound existed as the cis isomer with a pseudo-
chair conformation. However, for some of the derivatives, the trans compound
became an important competitor. Russell, Osuch, and Suleman then examined the
macrocylic rings from C-11 to C15 (115, 116). Here at last the trans isomer could
be observed as well. With potassium ion present, the cis isomer was favored, but
with a cryptand to complex the potassium ion the trans isomer was readily detected.
Elsewhere in this volume, Keith Ingold mentions the various descriptions of
free radicals, discussing the terms “stable,” “transient,” and “persistent.” Ingold
felt that the word “stable” should be confined to radicals that could be “put in a
bottle.” By this criterion, semidiones could be described as persistent if not stable.
Although there were exceptions, the semidiones generally stayed around in the
129
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 ESR cell for many minutes. In view of the synthesis a few years ago of potassium
benzophenone ketyl and potassium naphthalenide (117), we would hate to say that
semidiones could not be made in a stable form. However, just in their persistent
form they provided a source of chemical riches for the Russell group for 20 years.

Electron Transfer
Fortunately, Russell did write down an overview of his early work on electron
transfer sometime after 1982 plus more material about his later work in Jan., 1993,
but these write-ups are very terse (118). Consequently, our discussion of this work
will rely mainly upon the recollections of Ed Janzen (119) and Tom Strom along
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

with our analysis of the early publications in this area.

Russell himself considered his 1954 paper on the autoxidation of
2-nitropropane to be his first paper concerning electron transfer (80). However,
this work really got started in 1962. Ed Janzen was working on the oxidation of
carbanions, while I (Tom Strom) was working on making various anion radicals.
Russell thought that the oxidation process possibly entailed the transfer of an
electron to the oxygen molecule. Janzen felt that proof could be obtained by
using an organic catalyst to accept the electron. Ed suggested using nitrobenzene,
as its radical anion was stable and could be readily detected by ESR. Russell
was dubious about this, but Ed went ahead and did the experiments anyway,
using the hydrocarbon fluorene as the carbanion source after ionization with
potassium t-butoxide. The ESR experiments showed that nitrobenzene anion
radical was present, and oxidation rate experiments with fluorene demonstrated
that nitrobenzene greatly enhanced the uptake of oxygen. The publication from
part of this work has been referred to before (82). Probably it should have been
referred to as “Electron Transfer. II,” but Russell labeled it as “Reactions of
Resonance Stabilized Anions. II.”
Janzen followed up this work with a truly spectacular example of electron
transfer. When a solution of p-nitrotoluene was treated with potassium t-butoxide
in t-butyl alcohol under a nitrogen atmosphere, first a yellow color was observed,
possibly the carbanion, followed by the development of a blood red color in
tandem with the growth of the ESR spectrum of p-nitrotoluene anion radical.
The conclusion seemed inescapable that the carbanion transferred an electron to
unionized p-nitrotoluene. This work was published as a communication in 1962
(120) with the full paper following years later (121).
That same year I (Tom) had been drafted to work on electron transfer
reactions. It looked like an interesting area which would use my growing skills in
ESR, so I was glad to participate. It appeared that the electron transfer process as
studied thus far would be irreversible, but Russell thought that reacting di-anions
from a dihydro derivative of an unsaturated compound with the unsaturated
compound, for example, benzoin/benzil, hydrazobenzene/azobenzene, would
give two of the anion radical intermediates in an equilibrium process, with the
equilibrium possibly lying to the right. The two kinds of processes are shown
below in Figure 22. Ed and I studied a variety of the dihydro π compound/π
compound systems, the type shown in Figure 22b. We found that for a number
130
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 of the systems the ESR experiments showed that the equilibrium was well to the
right (122). The experimental procedure involved making up solutions of the
dihydro compound and its π parent in separate reservoirs, purging the reservoirs
of their oxygen, sending the separate solutions through flow tubes to a mixing
chamber, and then sending the mixture to the cell in the ESR cavity. Russell was
pleased with these results, and he wanted us to delineate the scope of the reaction
using mono carbanions, the system described in Figure 22a. He wanted us to use
lots of carbanions and lots of π acceptors. Our previous system had used copious
amounts of chemicals and even more DMSO. We needed to speed things up and
scale them down.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Figure 22. Electron transfer to unsaturated systems (П) with (a) mono-anions
or (b) di-anions.

Working together we designed the H cell (119). The cell consisted of two
small glass chambers connected at the top with a glass bridge. The top of each
chamber had a small 6 mm opening which could be plugged with rubber septa. A
small pierced glass joint at the top of the bridge was made to fit the top of the ESR
cell. The desired chemicals were weighed and along with solvent were transferred
into the two chambers through the small openings, the rubber septa closed the
openings, the septa were both pierced with hypodermic needles attached to tubes
from a cylinder of pre-purified nitrogen, and nitrogen was slowly bubbled through
the two chambers, with the gas being vented through the bottom of the ESR cell
attached through the glass joint. When the system had been sufficiently purged of
oxygen, the needles were pulled out of the septa, the bottom of the ESR cell was
plugged, and the contents of the two chambers were mixed and shaken down into
the ESR cell. It was possible to observe ESR spectra within a very few minutes
after mixing.
To keep track of the progress of the work, we made a chart similar to a
Sunday School attendance chart, with electron donors on one axis and electron
acceptors on the other. Those experiments that were successful got a gold star.
Tom remembers that Russell took his farewell photo of Tom with that chart.
The results of this work were published in 1964 (123), and a few years later
the Institute for Scientific Information designated it as a Citation Classic, one of
the most cited papers ever in its area. However, the article was almost turned
down. We had covered so many systems, about 40 carbanions and about a dozen
π acceptors, the latter ranging from ketones to azo compounds to nitro compounds
to hydrocarbons, that the paper was almost all data and little interpretation. The
referees felt the paper was “too diffuse” and were equally balanced between
acceptance and refusal. Fortunately, Russell had given a talk on this work at the
University of Rochester during that time, and JACS editor Marshall Gates was
in the audience. After the talk, Gates came up to Russell, telling him, “Now I
131
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 know what you’ve been doing.” Consequently, the editor came down on the side
of acceptance.
A later paper previously mentioned in connection with hydrocarbon oxidation
also had a large component involving electron transfer to nitro compounds (88).
Shortly thereafter, Russell published an additional article in which azo compounds
and their dihydro derivatives were the π acceptors and di-anion donors (124).
When Janzen and Strom received their degrees and left Iowa State, additional
significant electron transfer research was carried out by Ed Geels (125, 126). He
found that a deoxygenated mixture of nitrosobenzene and phenylhydroxyamine
gave an essentially quantitative yield of nitrosobenzene anion radical in the
80%DMSO-20%-t-butyl alcohol, with potassium t-butoxide as the base. If the
experiment was carried out in ethanol, up to 96% of azoxybenzene could be
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

isolated.
This next example of electron transfer research should be prefaced by an
incident Tom remembers from a group meeting sometime in 1963. Joseph
Schoeb was giving a talk to the group on Nathan Kornblum’s work on O- versus
C-alkylations. Russell got up during the talk and very animatedly pointed out
that the presence of a nitro substituent in any of those reactions might mean that
an anion radical was operative in those mechanisms. After the meeting Joe said
he had thought “Russell has lost his marbles.” No, Russell hadn’t! He had had
a flash of insight, which he shared with the group, albeit in a loud voice with
rapid speech. The timing is significant, because in 1964 and 1965 the Kornblum
group at Purdue published work about the involvement of radical anions in the
carbon alkylation of nitroparaffin salts (127, 128). Clearly Russell had come up
with the idea independently. He put Wayne Danen to work on the problem, and
two significant publications followed (129, 130). He later returned to several
studies of this type, which he later classified as examples of Joe Bunnett’s SRN1
mechanism.
Russell looked at other electron transfer processes throughout his career.
He examined the reactions of alkyl lithiums with alkyl halides, of titanium
tetrachloride with alkyl lithiums, of alkyl mercury chlorides with nitro-substituted
carbanions, and more. The study of electron transfer for him was a lifelong quest,
but we will stop our analysis of his quest here.

Research on Organic Synthesis

The basic DMSO system also had synthetic utility, and post-doc Hans
Dieter-Becker and graduate student Gary Mikol, later joined by others, made
many contributions in this area, examining condensations between the DMSO
anion and aldehydes, ketones, esters, etc. (131). Condensations with aromatic
esters gave beta keto sulfoxides. In mild acids these underwent the Pummerer
rearrangement to hemimercaptals of alpha keto aldehydes, which behaved like
typical glyoxals (132). This sequence provided a short, inexpensive route to
ninhydrin from ethyl phthalate (133). Hans Dieter-Becker characterized his
discovery in this way. “I got a little of the white product on my hands, and I saw
ninhydrin!” Ninhydrin gives a purple color on contact with peptides and proteins.
132
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 A variety of conversions were studied where a sulfur functional group was
retained (134). A series of transformations giving sulfur-free products provided
a method for extending the carbon chain by alkylation at the alpha carbon atom
(135). A number of papers in 1969 and 1970 described additional chemistry of the
beta keto sulfoxides (136–142).

The Exploration Comes to an End

After the death of Russell, his last graduate student, Chen Wang, completed
his degree with the guidance of Walter Trahanovsky. The last publication with
Russell as a coauthor appeared in 2001.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

We chose for the first part of the chapter title the phrase “Exploring Free
Radicals,” and what a glorious 50 years of exploration it was. Remembering
we were writing a chapter and not a book, we tried to be selective and not
comprehensive. We dealt mainly with Russell’s early themes of research, and
usually we did not follow those themes too far forward in time. Perhaps some
future chemical historian will explore these research themes fully. We were
grateful for the opportunity to learn how to do research from Glen Russell, and
physical organic chemistry is the richer for Russell’s body of work.

Acknowledgments
We were grateful to be able to provide a chapter on Glen Russell as part of this
volume on great physical organic chemists, a group to which Glen Russell clearly
belongs. First and foremost, we thank Susan and June Russell, his daughters, for
sharing their memories, photos, and other memorabilia with us. We acknowledge
the utility of the Russell Archive in the Special Collections Department of the Iowa
State University Library. The archive was also the source of that fine early photo
of Russell that served as our Figure 1. The archive allowed us to read important
correspondence written to and by Russell. We thank Vera Mainz for her invaluable
assistance in formatting this chapter and in preparing the figures to be included.
We are grateful for Edward Janzen’s reading and commenting on all sections
of this chapter as it was being prepared. Janzen’s memoir of his time in the Russell
group was both a source and an inspiration for us. We are pleased to learn that
volume 3 of his memoirs has been completed and is now in the publishing process.
We thank Keith Ingold and Henry Shine for carefully reading this chapter and
pointing out the errors they found. Any remaining errors are clearly our fault. For
answering our questions, sharing recollections, and assisting in other ways, we
thank the following people: Al Bemis, Ron Blankespoor, Ed Geels, John Gerlock,
Keith Ingold, Ed Janzen, Roy King, George Olah, Kirk Schmitt, Henry Shine,
Andy Streitwieser, Naushadalli Suleman, James Tanko, and Walter Trahanovsky.
We are grateful to Tim Marney of ACS Books for agreeing to the submission of this
chapter and to Bob Hauserman of ACS Books for facilitating the rapid completion
of this chapter.
133
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Last but by no means least, we thank our respective spouses, Walter
Trahanovsky and Charlotte Strom, for their encouragement and patience during
our completion of this chapter.
The task of a historian of chemistry is to somehow help the reader make sense
of an area of chemical history and learn about the chemists involved. We hope this
chapter will enlighten the reader about the history of free radical chemistry. We
also hope that the reader will be convinced that one of the seminal figures of 20th
century free radical chemistry was Glen A. Russell.

References
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

1. Janzen, E. G. I Chose the 1964 Ford Galaxy 500. My Graduation Present.

A Memoir My Life in Chemistry. Book II. Trips and Rides; Gentle Edge
Publishing Co., Ltd.: Guelph, Ontario, Canada, 2009. This book is available
from Dr. Edward Janzen, 37 Norma Crescent, Guelph, Ontario, Canada N1E
7M4.
2. Gomberg, M. J. Am. Chem. Soc. 1900, 22, 757–771.
3. Goldschmidt, S.; Renn, K. Berichte 1922, 55, 628–643.
4. Koelsch, C. F. J. Am. Chem. Soc. 1932, 54, 3384–3389.
5. Koelsch, C. F. J. Am. Chem. Soc. 1957, 79, 4439–4441.
6. Al’tshuler, S.; Zavoisky, E.; Kozyrev, B. Zh. Eksp Teor. Fiz. 1944, 14,
407–409.
7. Kochelaev, B. I. In Encyclopedia of Nuclear Magnetic Resonance, Volume 1.
Historical Perspectives; Grant, D. M., Harris, R. K., Eds.; Wiley and Sons:
Chichester, U.K., 1996; pp 435−437.
8. Eaton, G. R.; Eaton, S. S.; Salikhov, K. M. Foundations of Modern EPR;
World Scientific: Singapore, 1998.
9. Brown, H. C.; Russell, G. A. J. Am. Chem. Soc. 1952, 74, 3995–3998.
10. Russell, G. A.; Brown, H. C. J. Am. Chem. Soc. 1955, 77, 4025–4030.
11. Russell, G. A.; Brown, H. C. J. Am. Chem. Soc. 1955, 77, 4031–4035.
12. Russell, G. A.; Brown, H. C. J. Am. Chem. Soc. 1955, 77, 4578–4582.
13. Kharasch, M. S; Brown, H. C. J. Am. Chem. Soc. 1939, 61, 2142–2150.
14. Ash, A. B; Brown, H. C. Rec. Chem. Prog. 1948, 9, 81–94.
15. Schmitt, K. Private Communication to E. T. Strom, July 21, 2015.
16. Ingold, K. U. Private Communication to E. T. Strom, June 19, 2015.
17. Janzen, E. G. I Chose the 1964 Ford Galaxy 500. My Graduation Present.
A Memoir My Life in Chemistry. Book II. Trips and Rides; Gentle Edge
Publishing Co., Ltd.: Guelph, Ontario, Canada, 2009; pp 41−43, 173−174.
18. Russell, G. A. J. Am. Chem. Soc. 1956, 78, 1035–1040.
19. Russell, G. A. J. Am. Chem. Soc. 1956, 78, 1041–1044.
20. Russell, G. A. J. Am. Chem. Soc. 1956, 78, 1044–1046.
21. Russell, G. A. J. Am. Chem. Soc. 1956, 78, 1047–1954.
22. Russell, G. A. J. Am. Chem. Soc. 1957, 79, 2977–2978.
23. Letter from H. C. Brown to Henry Gilman, Mar. 14, 1958. Iowa State Russell
Archives.

134
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 24. Letter from Charles A. Goetz to Glen A. Russell, May 8, 1958. Iowa State
Russell Archives.
25. King, R. W. Private Communication to E. T. Strom, July 18, 2015.
26. Message from Vladimir Haensel presumably to Russell, undated, Iowa State
Russell Archives.
27. Letter from Robert Taft to Glen A. Russell, Mar. 17, 1965, Iowa State Russell
Archives.
28. Letter from George Hammond to Glen A. Russell, Mar. 24, 1965, Iowa State
Russell Archives.
29. Letter from Harold Hart to Glen A. Russell, Mar. 22, 1965, Iowa State
Russell Archives.
30. Letter from George Olah to Glen A. Russell, Mar. 27, 1965, Iowa State
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

Russell Archives.
31. Letter from George Pimentel to George Olah, Mar. 30, 1965, Iowa State
Russell Archives.
32. Olah, G. A. Private Communication to E. T. Strom, July 1, 2015.
33. Streitwieser, A. Private Communication to E. T. Strom, June 3, 2015.
34. Ingold, K. U. Private Communication to E. T. Strom, June 2, 2015.
35. Russell, G. A.; Wang, L.; Yao, C.-F. J. Org. Chem. 1995, 60, 5390–5395.
36. Letter from Orville Chapman to Walter Trahanovsky, Jan., 1998,
Trahanovsky Private Files.
37. Janzen, E. G. Private Communication to E. T. Strom, July 30, 2015.
38. Bemis, A. G. Private Communication to K. M. Trahanovsky, Mar. 18, 2015.
39. Geels, E. J. Private Communication to E. T. Strom, June 20, 2015
40. Tanko, J. M. Private Communication to E. T. Strom, Aug. 4, 2015.
41. Trahanovsky, W. S. Private Communication to E. T. Strom, Sept. 15, 2015.
42. Blankespoor, R. L. Private Communication to E. T. Strom, June 9, 2015.
43. Hogan, J. P.; Banks, R. L. In History of Polyolefins; Seymour, R. B., Cheng,
T., Eds.; D. Reidel: Dordrecht, Holland, 1986; pp 103−115.
44. Schmitt, K. Private Communication to E. T. Strom, July 29, 2015.
45. Ingold, K. U. Private Communication to E. T. Strom, April 27, 2015.
46. Gerlock, J. L. Private Communication to E. T. Strom, Mar. 19, 2015.
47. Russell, S.; Russell, J. Private Communication to E. T. Strom, Aug. 11, 2015.
48. Knight, D. W.; Russell, G. A. Encyclopedia of Reagents for Organic
Synthesis; John Wiley and Sons: Hoboken, NJ, 2001.
49. Russell, G. A.; Rhee, J. L.; Baik, W. Heteroatom Chem. 1998, 9, 201–208.
50. Morrison, R. T.; Boyd, R. N. Organic Chemistry, 1st ed.; Allyn and Bacon,
Inc.: Boston, 1959; p 259.
51. Morrison, R. T.; Boyd, R. N. Organic Chemistry, 2nd ed.; Allyn and Bacon,
Inc.: Boston, 1966; pp 128−130.
52. Russell, G. A. J. Am. Chem. Soc. 1955, 77, 4583–4590.
53. Miller, A. A.; Mayo, F. R. J. Am. Chem. Soc. 1956, 78, 1017–1023.
54. Mayo, F. R.; Miller, A. A. J. Am. Chem. Soc. 1956, 78, 1023–1034.
55. Russell, G. A. Chem. Ind. 1956, 1483.
56. Russell, G. A. J. Am. Chem. Soc. 1957, 79, 3871–3877.
57. Russell, G. A. J. Am. Chem. Soc. 1957, 79, 2977–2978.
58. Russell, G. A. J. Am. Chem. Soc. 1958, 80, 4987–4996.

135
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 59. Russell, G. A. J. Am. Chem. Soc. 1958, 80, 4997–5001.
60. Russell, G. A. J. Am. Chem. Soc. 1958, 80, 5002–5003.
61. Russell, G. A. J. Org. Chem. 1959, 24, 300–302.
62. Russell, G. A. Tetrahedron 1960, 8, 101–106.
63. Bridger, R. F.; Russell, G. A. J. Am. Chem. Soc. 1963, 85, 3754–3765.
64. Russell, G. A.; Ito, A.; Hendry, D. G. J. Am. Chem. Soc. 1963, 85,
2976–2983.
65. Hendry, D. G.; Russell, G. A. J. Am. Chem. Soc. 1964, 86, 2371–2374.
66. Russell, G. A.; Williamson, R. C. J. Am. Chem Soc. 1964, 86, 2357–2364.
67. Russell, G. A.; Williamson, R. C. J. Am. Chem. Soc. 1964, 86, 2364–2367.
68. Russell, G. A. In Free Radicalsl; Kochi, J., Ed.; Interscience: New York,
1972; pp 275−331.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

69. Bloomfield, G. F. J. Chem. Soc. 1944, 114–120.

70. Goldfinger, P.; Gosselain, P A.; Martin, R. H. Nature 1951, 168, 30–32.
71. Sixma, F. L. J.; Riem, R. H. Koninkl. Ned. Akad. Wetenschap., Proc. 1958,
61B, 183–187.
72. McGrath, B. P.; Tedder, J. M. Proc. Chem. Soc. 1961, 80–81.
73. Russell, G. A.; DeBoer, C.; Desmond, K. M. J. Am. Chem. Soc. 1963, 85,
365–366.
74. Pearson, R. E.; Martin, J. C. J. Am. Chem. Soc. 1963, 85, 354–355.
75. Walling, C.; Rieger, A.; Tanner, D. J. Am. Chem. Soc. 1963, 85, 3129–3134.
76. Russell, G. A.; DeBoer, C. J. Am. Chem. Soc. 1963, 85, 3136–3139.
77. Russell, G. A.; Desmond, K. M. J. Am. Chem. Soc. 1963, 85, 3139–3141.
78. Pearson, R. E.; Martin, J. C. J. Am. Chem. Soc. 1963, 85, 3142–3146.
79. Russell, G. A. J. Am. Chem. Soc. 1953, 75, 5011–5013.
80. Russell, G. A. J. Am. Chem. Soc. 1954, 76, 1595–1600.
81. Russell, G. A.; Moye, A. J.; Nagpal, K. J. Am. Chem. Soc. 1962, 84,
4154–4155.
82. Janzen, E. G. J. Chem. Educ. 1964, 41, 281.
83. Janzen, E. G. I Chose the 1964 Ford Galaxy 500. My Graduation Present.
A Memoir My Life in Chemistry. Book II. Trips and Rides; Gentle Edge
Publishing Co., Ltd.: Guelph, Ontario, Canada, 2009; p 81.
84. Russell, G. A.; Janzen, E. G.; Becker, H.-D.; Smentowski, F. J. J. Am. Chem.
Soc. 1962, 84, 2652–2653.
85. Corey, E. J.; Chaykovsky, M. J. Am. Chem. Soc. 1962, 84, 866–867.
86. Russell, G. A.; Janzen, E. G.; Bemis, A. G.; Geels, E. J.; Moye, A. J.; Mak, S.;
Strom, E. T. Oxidation of Hydrocarbons in Basic Solution. Adv. Chem. Ser.
1965 (51), 112–171.
87. Russell, G. A.; Bemis, A. G. J. Am. Chem. Soc. 1966, 88, 5491–5497.
88. Russell, G. A.; Moye, A. J.; Janzen, E. G.; Mak, S.; Talaty, E. R. J. Org.
Chem. 1967, 32, 137–146.
89. Russell, G. A.; Bemis, A. G.; Geels, E. J.; Janzen, E. G.; Moye, A. J.
Oxidation of Carbanions. Oxidation of Diarylmethanes and Diarylcarbinols
in Basic Solution. Adv. Chem. Ser. 1968, 75, 174–202.
90. Strom, E. T.; Russell, G. A.; Stephens, R. D. J. Phys. Chem. 1965, 69,
2131–2132.
91. Michaelis, L.; Fetcher, E. S. J. Am. Chem. Soc. 1937, 59, 1246–1249.

136
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 92. Venkataraman, B.; Fraenkel, G. K. J. Am. Chem. Soc. 1955, 77, 2707–2713.
93. Dehl, R.; Fraenkel, G. K. J. Chem. Phys. 1963, 39, 1793–1802.
94. Luckhurst, G. R.; Orgel, L. E. Mol. Phys. 1963−64, 7, 297–299.
95. Fischer, E. Ann. 1882, 211, 214–232.
96. Cardon, S. Z.; Lankelma, H. P. J. Am. Chem. Soc. 1948, 70, 4248–4249.
97. Strom, E. T.; Russell, G. A.; Schoeb, J. H. J. Am. Chem. Soc. 1966, 88,
2004–2007.
98. Heller, C.; McConnell, H. M. J. Chem. Phys. 1960, 32, 1535–1539.
99. Smith, G. V.; Kriloff, H. J. Am. Chem. Soc. 1963, 85, 2016–2017.
100. Russell, G. A.; Strom, E. T. J. Am. Chem. Soc. 1964, 86, 744–745.
101. Russell, G. A.; Strom, E. T.; Talaty, E. H.; Weiner, S. A. J. Am. Chem. Soc.
1966, 88, 1998–2004.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

102. Russell, G. A.; Law, W. C.; Zaleta, M. J. Am. Chem. Soc. 1985, 107,
4175–4182.
103. Russell, G. A.; Strom, E. T.; Talaty, E. R.; Chang, K. Y.; Stephens, R. L.;
Young, M. C. Record Chem. Progr. 1966, 27, 3–35.
104. Russell, G. A. Science 1968, 161, 423–433.
105. Russell, G. A.; Stephens, R. D. J. Phys. Chem. 1966, 70, 1320–1321.
106. Russell, G. A.; Talaty, E. R. J. Am. Chem. Soc. 1964, 86, 5345–5346.
107. Russell, G. A.; Talaty, E. R. Science 1965, 148, 1217–1218.
108. Russell, G. A.; Talaty, E. R.; Horrocks, R. H. J. Org. Chem. 1967, 32,
353–360.
109. Russell, G. A.; Chang, K. Y. J. Am. Chem. Soc. 1965, 87, 4381–4383.
110. Russell, G. A.; Chang, K. Y.; Jefford, C. W. J. Am. Chem. Soc. 1965, 87,
4383–4384.
111. Russell, G. A.; Keske, R. G. J. Am. Chem. Soc. 1970, 92, 4458–4461.
112. Russell, G. A.; Keske, R. G.; Holland, G.; Mattox, J.; Givens, R. S.;
Stanley, K. J. Am. Chem. Soc. 1975, 97, 1892–1900.
113. Russell, G. A.; Osuch, C. E. J. Am. Chem. Soc. 1978, 100, 5979–5981.
114. Russell, G. A.; Osuch, C. E.; Senatore, G.; Morita, T.; Yamashita, M. J. Org.
Chem. 1979, 44, 2780–2783.
115. Russell, G. A.; Osuch, C. E. J. Am. Chem. Soc. 1982, 104, 3353–3358.
116. Russell, G. A.; Suleman, N. K. J. Am. Chem. Soc. 1985, 107, 1717–1720.
117. Scott, T. A.; Ooro, B. A.; Collins, D. J.; Shatruk, M.; Yakovenko, A.;
Dunbar, K. R.; Zhou, H.-C. Chem. Commun. 2009, 65–67.
118. Russell, G. A. Unpublished manuscripts: undated; Jan. 1993, Iowa State
Russell Archives.
119. Janzen, E. G. I Chose the 1964 Ford Galaxy 500. My Graduation Present.
A Memoir My Life in Chemistry. Book II. Trips and Rides; Gentle Edge
Publishing Co., Ltd.: Guelph, Ontario, Canada, 2009; pp 85−87, 105−107.
120. Russell, G. A.; Janzen, E. G. J. Am. Chem. Soc. 1962, 84, 4153–4154.
121. Russell, G. A.; Janzen, E. G. J. Am. Chem. Soc. 1967, 89, 300–308.
122. Russell, G. A.; Janzen, E. G.; Strom, E. T. J. Am. Chem. Soc. 1962, 84,
4155–4157.
123. Russell, G. A.; Janzen, E. G.; Strom, E. T. J. Am. Chem. Soc. 1964, 86,
1807–1814.

137
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 124. Russell, G. A.; Konaka, R.; Strom, E. T.; Danen, W. C.; Chang, K.-Y.;
Kaupp, G. J. Am. Chem. Soc. 1968, 90, 4646–4653.
125. Russell, G. A.; Geels, E. J. J. Am. Chem. Soc. 1965, 87, 122–123.
126. Russell, G. A.; Geels, E. J.; Smentowski, F. J.; Chang, K.-Y.; Reynolds, J.;
Kaupp, G. J. Am. Chem. Soc. 1967, 89, 3821–3828.
127. Kerber, R. C.; Urry, G. W.; Kornblum, N. J. Am. Chem. Soc. 1964, 86,
3904–3905.
128. Kerber, R. C.; Urry, G. W.; Kornblum, N. J. Am. Chem. Soc. 1965, 87,
4520–4528.
129. Russell, G. A.; Danen, W. C. J. Am. Chem. Soc. 1966, 88, 5663–5665.
130. Russell, G. A.; Danen, W. C. J. Am. Chem. Soc. 1968, 90, 347–353.
131. Russell, G. A.; Becker, H.-D. J. Am. Chem. Soc. 1963, 85, 3406–3410.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch006

132. Becker, H.-D; Mikol, G. J.; Russell, G. A. J. Am. Chem. Soc. 1963, 85,
3410–3414.
133. Becker, H.-D.; Russell, G. A. J. Org. Chem. 1963, 28, 1896.
134. Russell, G. A.; Sabourin, E. J.; Mikol, G. J. J. Org. Chem. 1966, 31,
2854–2858.
135. Russell, G. A.; Mikol, G. J. J. Am. Chem. Soc. 1966, 88, 5498–5504.
136. Russell, G. A.; Sabourin, E. J. J. Org. Chem. 1969, 34, 2336–2339.
137. Russell, G. A.; Sabourin, E. J.; Hamprecht, G. J. Org. Chem. 1969, 34,
2339–2345.
138. Russell, G. A.; Ochrymowycz, L. A. J. Org. Chem. 1969, 34, 3618–3624.
139. Russell, G. A.; Ochrymowycz, L. A. J. Org. Chem. 1969, 34, 3624–3626.
140. Russell, G. A.; Ochrymowycz, L. A. J. Org. Chem. 1970, 35, 764–770.
141. Russell, G. A.; Ochrymowycz, L. A. J. Org. Chem. 1970, 35, 2106–2108.
142. Russell, G. A.; Hamprecht, G. J. Org. Chem. 1970, 35, 3007–3012.

138
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Chapter 7

Norbornyl Cation Isomers Still Fascinate

Paul von R. Schleyer,1,4 Vera V. Mainz,*,2 and E. Thomas Strom3
1Department of Chemistry, University of Georgia, Athens, Georgia 30602
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

2Departmentof Chemistry, University of Illinois at Urbana-Champaign,

600 S. Mathews Avenue, Urbana, Illinois 61802
3Department of Chemistry & Biochemistry, University of Texas at Arlington,

Box 19065, 700 Planetarium Place, Arlington, Texas 76019

*E-mail: [email protected]
4Deceased Nov. 21, 2014

The proposal by Winstein and Trifan that the 2-exo-norbornyl

cation has a bridged, non-classical structure initiated a number
of investigations into this system. In 1962, H. C. Brown
entered into the fray by proposing a pair of rapidly-equilibrating
cations and steric effects as alternative explanations. The
2-norbonyl cation was investigated via rate studies, nmr
experiments, theoretical calculations, and other methods over
a number of years by a number of workers. While the X-ray
determination of the 2-exo-norbornyl structure by Scholz et
al. (Science, 2013, 341, 62) brought long overdue closure to
the vituperative structure controversy, this chapter summarizes
other remarkable issues currently engaging research groups
worldwide. Unexpectedly, Duncan et al.’s gas phase protonation
of norbornene gives the 1,3-dimethylcyclopentyl cation, the
C7H11+ global minimum. Merino et al.’s molecular dynamics
simulations reveal many acyclic as well as monocyclic
intermediates along the reaction pathway. Finally, the
2-endo-norbornyl cation also has a bridged minimum, but
“leakage” to the 2-exo cation has a very low barrier. This
bridging of the 2-endo cation explains products found from the
reaction of α-pinene with acetic acid. Thus, even at this late
date, norbornyl cation isomers still fascinate.

© 2015 American Chemical Society

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Introduction
We truly wish that this Introduction were being written by Paul Schleyer
(Figure 1), but, sadly, Paul passed on before completely transforming his talk from
March, 2014, at the Dallas Symposium on the Norris Award into a written chapter
for this book. We had provided Paul a written transcript of his talk with which to
work. In his message to us of July 15, 2014, he noted that his talk “needs extensive
stylistic editing to refine oral into written English.” In his last message to us on Nov.
19, 2014, he stated that he planned to finish work on it the very next week; but then
he passed on two days later. He certainly had begun the process of revision, but
there was still more to be done. We have undertaken to finish the effort that Paul
started. Elsewhere in this book you can read Paul’s unfinished memoir as edited
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

by Andrew Streitwieser and decide for yourself whether we have captured Paul’s
authentic writing style.

Figure 1. Paul von R. Schleyer. (Photograph courtesy of István Hargittai).

140
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Our revisions included constructing an abstract, correcting non-fluencies,
inserting a sentence or so to improve clarity, providing figure captions, and
finding the references Paul cited incompletely in his text. We have done our
best to capture the essence of Paul’s analysis of this fascinating norbornyl cation
research. In a very early e-mail message to us (April 12, 2014), Paul stated “My
talk reported on research primarily by others (my participation was minor)---”.
We think otherwise. We think Paul was a significant actor in the norbornyl cation
problem. This chapter may not be the last word on the topic, but it is Paul’s last
words on the matter and as such is well worth reading.

Overview
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

It is appropriate that somebody should talk about the norbornyl cation on the
50th anniversary of the founding of the James Flack Norris Award. A search of
Web of Science under “norbornyl” (Figure 2) shows the yearly progression of the
citations. Astonishingly, today there are still a thousand citations per year for the
word norbornyl. These citations include the cation and a few other things.

Figure 2. Web of Science: Number of “norbornyl” citations per year.

If you search for “norbornyl cation” you will also find a goodly number
of citations (Figure 3). More than half of the hits for norbornyl are due to the
norbornyl cation. It is still with us.

141
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 3. Web of Science: Number of “norbornyl cation” citations per year.

The first papers on the norbornyl cation were written by Saul Winstein (1)
and Jack Roberts (2). Some high points of the investigations are marked in Figure
3.The first production of the stable norbornyl cation was by Olah (3) in 1964.
In 1962, Brown began to criticize Winstein’s interpretation (4) of the σ-bridged
2-norbornyl cation (and other σ-bridged carbocations). This controversy
continued very heatedly – it was most vituperative (5). Brown withdrew from the
fray in 1986 (6), after over a dozen years of keeping things at a boil, calling for
experimental evidence. He was still unconvinced as only theoretical evidence had
been presented to date, but in particular, the X-ray structure was not available. An
X-ray crystal structure is an obvious way of solving a structural problem and was
not available. Many people had tried but failed to solve the X-ray structure due to
disorder. I’ll come back to this point later. Note that at this point the structure in
the gas-phase had not been realized either, but other experimental evidence and
definitive theoretical calculations had established the bridged structure reliably.

Historical Background
The story, however, goes back a lot further, to the paper by Wagner in 1899
(7), which proposed carbonium ions as intermediates in terpene rearrangements.
The mechanism of the rearrangement of camphene hydrochloride into isobornyl
chloride was further elucidated by Meerwein and van Emster in 1922 (8), see
Figure 4. Proving the structures of these products and starting materials was very
difficult. And we should remember that these types of rearrangements were hardly
known at the time. People expected the least structural change in moving from
starting materials to product. The first formulation of a non-classical ion is usually
attributed to Christopher Wilson (9), although Ingold (10, 11) has some claim for
it, too. These terpenes, of course, have three more methyl groups than norbornyl.

142
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 4. (Top) Representation of the camphenyl cation to isobornyl cation

rearrangement (8). Wilson (9) and Ingold (10) representations of the camphenyl
cation. [(Reproduced with permission from refn. (8). Copyright (1922)
Wiley-VCH Verlag GmbH & Co. KGaA); (Reproduced with permission from
refn. (9). Copyright (1939) The Royal Society of Chemistry); (Reproduced with
permission from refn. (10). Copyright (2001) American Chemical Society)]

The chronology of the 2-norbornyl cation controversy is summarized in

Table 1. We start our history with Winstein and Trifan (1). Trifan was a
faculty member at Princeton at the time I was there, so I knew him personally.
They proposed the bridged, non-classical structure. The first refutation of this
by Brown was in 1962 (4). In 1977, Brown published The Nonclassical Ion
Problem (12), with my comments included. I talked about the stable ion, the
rearrangements, the nonclassical vs. classical carbocation problem. And then
a number of important works appeared supporting the nonclassical structure.
NMR, in particular Yannoni’s 13C 5K NMR data (13) failed to find equilibration
of localized (classical) ions. Saunders’s isotopic perturbation method (14) was
applied to the 1H NMR spectrum of the norbornyl cation and the model supported
the nonclassical structure. In 1983, Fritz Schaefer and I published the first
computations that got the right answer for the right reason (15). Prior to 1983,
many people had published calculations on this system but they were not reliable
because you need to have a large enough basis set and explicitly include electron
correlation corrections in the calculation. There was considerable debate about
the interpretation of the ESCA spectra (16–21). And as I said already, Brown
withdrew from the fray, unconvinced (6). Efforts to get the crystal structure
went on in many laboratories (22, 23), probably more than I’ve listed here, but
143
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 their efforts failed due to disorder. The X-ray crystal structure was finally solved
in 2013 by German groups under Meyer and Krossing (24). Mike Duncan,
my colleague at Georgia is still seeking the gas-phase IR spectrum and I’ll say
something about that later.

Table 1. The 2-Norbornyl Cation (2NB+) Controversy: A Chronology

1949, 1952 Winstein and Trifan propose the bridged, non-classical structure of
2NB+ (1)
1962 H. C. Brown refutes their evidence; proposes a pair of
rapidly-equilibrating classical cations and steric effects as alternative
explanations (4)
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

1964 Schleyer, Olah, Saunders, et al. report preparation of the stable 2NB+
and interpret its temperature-dependent NMR spectra - rearrangements
(3)
1977 book “The Nonclassical Ion Problem,” H. C. Brown (with Schleyer
commentary) (12)
1982 Olah, Prakash, Anet variable temperature NMR and isomerization
barriers (25)
1982 Yannoni, Mhyre 5 K MAS NMR fails to find evidence of equilibration
(13)
1983 Saunders, Kates isotopic perturbation method applied - nonclassical
(14)
1983 Schaefer, Schleyer, et al. first adequately high-level computations
- nonclassical (15)
1982 - 1985 Olah, Prakash, further support the nonclassical 2NB+ structure (26),
but considerable debate continues (e.g., ESCA spectra interpretation)
(16–21)
1986 H. C. Brown, his major objections never answered directly, withdraws
unconvinced from the debate, calling for decisive experimental
evidence (6)
1960s - now R. Bau (unpublished results), J. C. Huffman (22), T. Laube (23), M.
Goldstein (unpublished results), others. Many attempts to solve the
X-ray structure of 2NB+ failed due to disorder problems
2013 Scholz, Himmel, Heinemann, Schleyer, Meyer, Krossing (24) finally
solve the 2NB+ crystal structure - a tour de force!!

The Winstein evidence for bridging in the beginning was rather flimsy (1)
but was convincing to most people. The high 2-norbornyl exo/endo solvolysis
rate (Figure 5) and product ratios were the basis for Winstein and Trifan’s non-
classical ion formulation. The bornyl, norbornyl, and isobornyl structures have
very high exo/endo rate ratios which Winstein and Trifan interpreted as indicating
participation of the neighboring C-C bond. Brown argued (4) that steric effects
are responsible since the tertiary analogs also have high exo/endo rate ratios. His

144
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 rebuttal, inter alia, in his 1977 book with me as commentator (12), pointed out
many imperfections in interpretations based on nonclassical ion theory, which still
remain unanswered even to this day. So to some extent the controversy still goes
on.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 5. Solvolysis rates for norbornyl compounds vs. cyclohexyl

Two Classical Ions in Rapid Equilibrium or Symmetrical

Bridged Structure
The issue was the decision between the double-well minimum, two classical
ions in rapid equilibrium, or a symmetrical bridged structure (Figure 6). This is not
unlike the benzene problem, whether benzene oscillates back and forth between
two Kekule structures or has a single minimum. Winstein and Trifan (1) suggested
both, but favored bridging. Solutions of such double- vs. single-well potential
problems are difficult!

Figure 6. Nonclassical ion vs. rapid equilibrium model.

The Winstein evidence (27) was not just based on high exo/endo rate ratios
but also on the product (Figure 7). The products were exclusively exo starting
from both endo and exo starting materials. But there was something funny from
the endo starting material, because if optically active starting materials were used,
the products were not completely racemized.

“In anchimerically unassisted solvolysis of endo-2-norbornyl-X chemical

capture of an unsymmetrical norbornyl cationic species is quite evident,
however. While most of the initially formed classical species becomes
nonclassical before giving rise to product an appreciable fraction does
145
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 give optically active inverted product. In the more nucleophilic solvents
this amounts to 13%.” --- Winstein (27) (Reproduced with permission
from refn. (27). Copyright (1965) American Chemical Society)
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 7. Exo vs. Endo 2-Norbornyl Solvolyses [Endo-Product Complexities]

So it doesn’t go completely through the nonclassical ion. It was supposed that

some sort of classical ion intervenes in the endo case – and I’ll come back to that,
later on.

Direct Observation of the 2-Norbornyl Cation

In 1962, Olah gave his memorable lecture at the Ninth Organic Reaction
Mechanism Conference held at the Brookhaven Natural Laboratories and
announced that he was able to get stable norbornyl ions in solution. We agreed
then to attack the norbornyl problem. He pointed out that he worked for Dow
and they didn’t have a preparative laboratory, and asked if I would make starting
materials for him. My research group then made these various starting materials
and sent them to George and he converted them to the same norbornyl ion. We
published these results (28) in 1964 (Figure 8).

146
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 8. Excerpts from Schleyer, Watts, Fort Jr., Comisarow, and Olah, 1964
(28). (Reproduced with permission from refn. (28). Copyright (1964) American
Chemical Society)

Olah and I enlisted Marty Saunders’s help, as he had better NMR facilities
at the time, and he found evidence (Figure 9) for three types of internal
rearrangements: the Wagner-Meerwein, (which is at least the nonclassical ion),
the 6,2-hydride shift, and the 3,2-hydride shift.

147
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 9. Excerpts from Saunders, Schleyer, and Olah, 1964 (3). (Reproduced
with permission from refn. (3). Copyright (1964) American Chemical Society)

In Jack Robert’s isotope-labeling study of Wagner-Meerwein and 6,2-hydride

shifts in the 2NB+ (29), Roberts anticipated the 1964 NMR observations of
complete scrambling, when the slower 3,2-hydride shifts also took place. Jack
pointed out in this paper that if the 3,2-hydride shift also went fast enough it
would lead to scrambling of all the ions.

“The observed near-equilibrium distribution rules out any 2,3-hydrogen

shifts which would be associated with such an intermediate as shown [M.
J. S. Dewar, Ann. Reports (Chem. Soc.), 121 (1951)], since cations of this
type in combination with III and XIII would lead to ‘scrambling’ of all of
the atoms.” --- Roberts (29) (Figure 10). (Reproduced with permission
from refn. (29). Copyright (1954) American Chemical Society)

And in fact, that scrambling is what the NMR spectra show at higher
temperatures (Figure 11), where you see only a single hydrogen peak. At lower
temperatures, the rapid rearrangements slow down and you see a more highly
resolved spectrum.
So this is the nub of the problem once again: does the 2NB+ structure have CS
symmetry or two C1 symmetry enantiomers which rapidly equilibrate (Figure 12).

148
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 10. Excerpts from Roberts, Lee, and Saunders, 1954 (29). (Reproduced
with permission from refn. (29). Copyright (1954) American Chemical Society)
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 11. 1H NMR spectra of the 2-norbornyl cation at various temperatures

(3). (Reproduced with permission from refn. (3). Copyright (1964) American
Chemical Society)

149
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 12. The 2-norbornyl cation: nonclassical (A) (1) vs. classical structure
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

(B) (4). [(Reproduced with permission from refn. (1). Copyright (1949) American
Chemical Society); (Reproduced with permission from refn. (4). Copyright
(1962) The Royal Society of Chemistry)]

Adventures in X-Ray Crystallography

Much evidence, computational and experimental, supports the nonclassical

structure, but the X-ray structure was strangely missing. The problem is that the
norbornyl cation is more or less spherical and rotates rapidly in the crystal, so
therefore it causes disorder. You can try to freeze it out but no one had succeeded.
The final solution came in 2013 (24) by two German associates of mine. Karsten
Meyer at Erlangen-Nürnberg and Ingo Krossing at Freiburg recognized that not
only was there disorder due to the rotation in the crystal but there was also disorder
due to the rapid rearrangements (Figure 12), and you had to freeze both types of
disorder out. In order to do that, 20-fold annealing, from about 86 K to about 40
K had to be carried out. The experiment cost about 10,000 euros in liquid helium
– so it was a tour de force to say the least. I won’t go into this, but at 86 K the
half-life is still pretty fast. Figure 13 summarizes this data.
In our publication of the X-ray structure (24), we summarized the situation as
follows:

“After decades of vituperative debate over the classical or

nonclassical structure of the 2-norbornyl cation, the long-sought X-ray
crystallographic proof of the bridged, nonclassical geometry of this
prototype carbonium has finally been realized. Exceptional treatment
was required. Crystals of the solvated [C7H11]+[Al2Br7]– • CH2Br2 salt,
obtained by reacting norbornyl bromide with aluminum tribromide in
CH2Br2, undergo a reversible order-disorder phase transition at 86 K
150
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 due to internal 6,1,2-hydride shifts of the 2-norbornyl cation moiety.
Cooling with careful annealing, data collection and refinement at 40
K revealed three similar independent 2-norbornyl cations structures
in the unit cell. All agree very well with ab initio computations at the
MP2(FC)/def2-QZVPP level.” -- Scholz, Himmel, Heinemann, Schleyer,
Meyer, and Krossing, 2013 (24). (From refn. (24). Reprinted with
permission from AAAS)

The left side of Figure 14 shows what the norbornyl cation looks like to the
X-ray diffractometer. When it is disordered you just have a jumble of atoms for the
norbornyl cation. But after the annealing process the unit cell has three norbornyl
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

cations in it (right, Figure 14), so you get three for the price of one. And in fact,
fortuitously and coincidentally, the cover of Brown’s book (12) has those three
norbornyl cations (see Figure 15). So, somehow, this result was anticipated!
The three cations in the tripled unit cell have roughly the same bond lengths
- the average agrees with good ab initio calculations and without a doubt the
structure has the bridge form and CS symmetry. This data is summarized in Figure
16.

Figure 13. Disorder in 2-norbornyl cation due to rapid 6,2-H shift.

151
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 14. Asymmetric unit of the [C7H11]+[Al2Br7]- • CH2Br2 crystal structure.

Figure 15. The cover from The Nonclassical Ion Problem by Herbert C. Brown
(12). (Reproduced with permission from refn. (12). Copyright (1977) Plenum
Press. With permission of Springer Science+Business Media)

152
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 16. Unsubstituted 2-norbornyl cation XRD Structure (24) showing

the data for the three independent cations in the unit cell. (From refn. (24).
Reprinted with permission from AAAS)

Classical vs. Nonclassical Structure via Infrared Spectroscopy

Another way this problem might be solved would be to compare the
experimental infrared spectrum with those that have been computed for the
classical and nonclassical alternatives (Figure 17). So which one does it fit? The
solution spectra that George Olah and others have reported are not very sharp; the
lines are broad in solution. Getting a spectrum in the gas phase would be better
and my colleague Mike Duncan at Georgia has an apparatus that does exactly
that. So one has to decide if the experimental spectrum fits the nonclassical model
or the classical model.

Figure 17. Another potential experiment solution: the gas phase 2NB+ structure
could be revealed by infrared photodissociation spectroscopy.
153
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The apparatus that Duncan uses is shown in Figure 18. It is very neat. The
cation is generated here and it is standard procedure for him to use H3+ as his
standard proton source. This proved to be a poor choice, but I have to tell the story
the way it happened and this is how he generated it. What you see in the normal
mass spectrum, since this is generated in Ar, is not only the various peaks due to
hydrocarbon cations, but also Ar tagged species. This is very important because
one can use a mass gate in the reflectron mass spectrometer to select this ion or
that set of ions. With the mass gate on, you irradiate them with a tunable dye laser
and scan the spectrum. And whatever the spectrum absorbs, you knock off the Ar.
You then detect the ions on the bottom part of the reflectron. Whenever there is a
peak in the infrared you get a peak in this spectrum. By plotting the intensities of
the laser vs. the frequency you get a nice IR spectrum.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 18. Duncan’s IR-PD apparatus and illustration of technique.

The IR spectrum obtained is shown in Figure 19. It is very sharp and very
good quality. Duncan has published a number of these. The first thing you do is
compare the spectrum with that computed for the 2-norbornyl cation, the bridged
form. And much to your horror, it doesn’t fit at all. In particular, the peak at
1525 cm-1, one of the strongest peaks in the experimental spectrum, is completely
missing in the calculated spectrum. Either you haven’t made the norbornyl cation
in this experiment or it doesn’t have the nonclassical structure.

154
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 19. Experimental spectrum of C7H11+ from protonation of norbornene

with H3+ vs. calculated nonclassical spectrum.

When you compare the experimental spectrum with the spectrum calculated
for the classical 2NB+ cation and it doesn’t match either (Figure 20). The
conclusion is that this experiment, the protonation of norbornene, did not yield
the 2-norbornyl cation. What is this the spectrum of?
People have gotten so used to the 2-norbornyl cation that they may imagine
that it must be the global minimum. It isn’t. There are other isomers of C7H11+ that
are more stable, and in particular the dimethyl allyl, which is the global minimum
(Figure 21).
Let’s compare the spectrum computed for the 1,3-dimethylcyclopentenyl
cation with the experimental spectrum (Figure 22). These peaks match almost
perfectly, and that is what the structure is. Instead of getting the norbornyl cation
from norbornene Duncan had gotten the 1,3-dimethylcyclopentenyl cation.

155
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 20. Experimental spectrum of C7H11+ from protonation of norbornene

with H3+ compared to the calculated spectrum for the 2-norbornyl cation - both
classical and nonclassical models.

Figure 21. Six low energy isomers, A - F, of C7H11+ (30). (Reproduced with
permission from refn. (30). Copyright (2014) Wiley-VCH Verlag GmbH & Co.
KGaA, Weinheim)
156
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 22. Experimental spectrum (top) vs. computed spectra for

1,3-dimethylcyclopentenyl (middle) and 2-norbornyl (bottom) (31). (Reproduced
with permission from refn. (31). Copyright (2014) Wiley-VCH Verlag GmbH &
Co. KGaA, Weinheim)

The next question is how on earth do you rearrange the norbornyl cation,
which is probably formed first, into this cyclopentenyl cation? Well, I’m sure all
of you think you are good carbonium ion pushers – I dare you to try it! You’ll see
that there is no easy way of doing it and so I enlisted the help of Gabriel Merino
in Mexico to help solve this problem.
The question again is how does the norbornyl cation rearrange into the
dimethylcyclopentenyl cation (Figure 23)? Merino (30) used a Born-Oppenheimer
molecular dynamics (BOMD) program that searches for minima, which generates
the potential energy surface going from the norbornyl cation bridge going to
dimethylcyclopentenyl cation. There are some 16 minima, each corresponding to
a different isomer, and a corresponding number of transition states in this process.
It is amazing what happens. First, the bicyclic ring opens to a monocycle, and
then it goes to an acyclic structure which closes back down to a monocycle, and
various shifts occur along this pathway. I won’t go into this in any more detail,
but it isn’t something you develop by trial and error.

157
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 23. Born-Oppenheimer Molecular Dynamics (BOMD) simulation of the

2-norbornyl cation conversion to the 1,3-dimethylcyclopentenyl cation (30).
(Reproduced with permission from refn. (30). Copyright (2014) Wiley-VCH
Verlag GmbH & Co. KGaA, Weinheim)

The 2-Endo-Norbornyl Cation

The next problem I want to talk about is the nature of the 2-endo-norbornyl
cation which I believe has gotten lost in the shuffle. It has been assumed that it has
a classical structure but if you try to calculate it with a classical geometry you never
find a minimum. The new insight is that the 2-endo-norbornyl cation also favors a
bridged structure. Now, that’s hardly new as there has been much evidence hinting
at 2-endo norbornyl bridging since 1900, but not so interpreted! The problem in
finding it experimentally arises because the 2-endo bridged ion, which I’ll show
you in a moment, has a very small transition state to go to the more stable 2-exo
form. And therefore, what Winstein calls leakage occurs quite easily (Figure 24).
For example, bridged ion depictions have been related to 2-bicyclo[3.1.1]heptanes
(the norpinane mesonomers) instead. A few examples follow.
This again is Winstein’s interpretation (33), in which the reaction goes through
some sort of species which he didn’t try to pin down (Figure 25).

“In anchimerically unassisted solvolysis of endo-2-norbornyl-X chemical

capture of an unsymmetrical norbornyl cationic species is quite evident,
however. While most of the initially formed classical species becomes
nonclassical before giving rise to product an appreciable fraction does
give optically active inverted product. In the more nucleophilic solvents
this amounts to 13%.” --- Winstein (33) (Reproduced with permission
from refn. (33). Copyright (1965) American Chemical Society)

158
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 24. Intrinsic reaction coordinate (IRC) for endo- to exo 2NB+ “leakage.”
IRC: First computations by MN12L (32).

Figure 25. Exo vs. endo 2-norbornyl solvolyses showing endo- product
complexities (33). (Reproduced with permission from refn. (33). Copyright
(1965) American Chemical Society)

Kirmse has written five or six papers, as well as a review article, on this general
topic, which doesn’t get much publicity and is summarized in Figure 26.

“The ionization of exo-2-norbornyl derivatives occurs with the

participation of the C1-C6 bond. (Winstein, 1952) For endo-2-norbornyl
derivatives corresponding participation of the C1-C7 bond should
likewise be possible. However, the ring strain (Maier/ Schleyer, 1981)
thereby increases.” --- Kirmse (34). (Reproduced with permission from
refn. (34). Copyright (1993) Wiley-VCH Verlag GmbH & Co. KGaA)

159
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 26. Exo vs. endo 2NB+ solvolysis (34). (Reproduced with permission
from refn. (34). Copyright (1993) Wiley-VCH Verlag GmbH & Co. KGaA)

If we invoke trans-anti-peri-planar arrangements, the exo would give the

bridged cation that we know. But the endo should give this bridged cation,
which is related to pinene, and has sometimes been called the pinyl cation in the
literature.
Figure 27 shows one of Kirmse’s reactions, endo with a methoxy, where the
product is a ketone. You can see the bridged ion in this classical representation.

Figure 27. Norpinyl product via an endo-2-norbornyl cation (35). (Reproduced

with permission from refn. (35). Copyright (1986) American Chemical Society)

Now, you recall from terpene chemistry, that addition of HCl to camphene
gives isobornyl chloride. That’s the exo product, as seen in the top reaction in
Figure 28. That’s the reaction that Meerwein (8) studied so brilliantly. But if I
160
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 write it as a classical ion, it looks like the representation in the bottom reaction in
Figure 28. The α and β pinenes react with HCl not to give isobornyl but bornyl
chloride, stereospecifically endo. If you write it as a classical ion and you turn it
around, these two are identical. You can’t get two different products from the same
intermediate ion, so there has to be some difference between these intermediates.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 28. Puzzling Wagner-Meerwein product stereochemistry.

So the difference must be in the nature of the bridging (Figures 29 and 30).
Wilson explained the exo case first, as I mentioned already. The exo case has
bridging as shown in Figure 29 and the endo case has bridging as shown in Figure
30.

Figure 29. I. Clarifying Wagner-Meerwein reaction products.

161
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 30. II. Clarifying Wagner-Meerwein reaction products.

Henry Rzepa has looked into this problem (Figure 31). In the reaction of
camphene + HCl we need to watch the chlorine. The first thing that happens is that
the methyl group rotates and the chlorine moves, giving the transition state. The
chlorine then moves to its new position. If both those movements in the counter
ion occur simultaneously we end up with bornyl chloride. This was studied by
Meerwein in 1922 (8).

Figure 31. Pinene-HCl to bornyl chloride (endo) rearrangement. (Images

courtesy of Prof. Henry Rzepa, 2015)

Henry has looked at some of the other possible transition states, see Figure
32. The barrier values indicated are free energy barriers.

162
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 32. 2-Pinyl to 2-endo-norbornyl dyotropic rearrangement. Barrier values

are free energy barriers. DataDOIs for the calculations are rs4, 645 and 646,
resolved as e.g. http://doi.org/rs4. (Images courtesy of Prof. Henry Rzepa, 2015)

Interestingly, Winstein found one case like this but didn’t realize it
(36). With Holness he had prepared what he thought were the nopinyl
p-bromobenzenesulfonates and reported that they reacted anomalously slowly;
that is, did not show the expected anchimeric assistance. Expansion of the
adjacent four membered ring with relief of ring strain apparently also did not
occur. This work was challenged by the Foote-Schleyer relationship, which
shows that the rate is anomalously slow. We pointed out that these results
would not be anomalous if Winstein’s starting structure was wrong. It turned
out that Winstein had not studied the nopinyl p-bromobenzenesulfonates at
all! They had rearranged to dimethylnorbornyl isomers (α-nopinyl reacted
with p-bromobenzenesulfonyl chloride in pyridine to form endo-camphenilyl
p-bromobenzenesulfonate while β-nopinyl reacted under the same conditions to
form apobornyl p-bromobenzenesulfonate) by internal return before the rates
were measured!
I told Winstein of our results. The compounds from the previous work (36)
were still available. They were re-examined and spectroscopically were shown
to have the structures as reported by Schleyer. In fact, internal return had taken
place and what he had solvolyzed was an endo- norbornyl derivative instead of the
nopinyl derivative that he had imagined. The two papers that resulted from these
investigations were published back-to-back (37, 38). The beginning of Winstein’s
paper (37) is shown in Figure 33.
163
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 33. Excerpts from Friedrich and Winstein, 1964 (37). (Reproduced with
permission from refn. (37). Copyright (1964) American Chemical Society)

The first computed structure of the endo-norbornyl cation (39) is shown in

Figure 34. There was an artifact in MINDO-3 which makes it an inadequate theory
level. In fact, these authors commented,

“Greatly to our surprise, [the bridged 2-endo-norbornyl cation] not only

turned out to be a stable species but more stable than any of the other
2-norbornyl isomers.” --- Dewar (39).

Figure 34. Excerpts from Dewar (39). Computational “discovery” of the bridged
2-endo-norbornyl as a MINDO-3 artifact. (Reproduced with permission from
refn. (39). Copyright (1977) American Chemical Society)

The endo-norbornyl cation is more stable than the exo-norbornyl cation

according to this inadequate calculation. But not more stable at refined ab
initio levels (32), where the minimum structure of the “2-norpinyl cation” was
confirmed.
164
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 A paper by Valkanas in 1976 (40), studied the reaction of α-pinene with
carboxylic acids in order to investigate the transition states necessary to achieve
the various products of the reaction (Figure 35).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Figure 35. Excerpt from Valkanas (40). 2-Endo bridging explains products from
reaction of α-pinene with acetic acid. (Reproduced with permission from refn.
(40). Copyright (1976) American Chemical Society)

Figure 36 shows some more endo-norbornyl cations that I’ve calculated.

These are the cations with methyl groups in various positions. All of these are
minima at good calculational levels, including the endo-2-norbornyl cation itself.

Figure 36. L2-endo-norbornyl (norpinyl) cations with computed bridged minima

(also see Kirmse (35).). (Reproduced with permission from refn. (35). Copyright
(1986) American Chemical Society)

165
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Again, the reason these endo structures have not been found experimentally is
that the endo bridged ion leaks rapidly to the exo bridged ion, as shown in Figure
24, and so you don’t detect it very easily.

Conclusions
The structure of the 2-norbornyl cation, the most vituperative controversy
of the 20th century involving numerous leading chemists, stimulated many
methodological developments and refinements to distinguish single- and
double-well potentials. Brown did not “kill physical organic chemistry”, but his
criticisms pointed to many important, but poorly understood problems.
The verification of sigma-bridging established the validity of “pentavalency”
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

for carbon and its wider acceptance as a bonding possibility. Note that nitrogenase,
a rather important enzyme, has a six coordinate carbon!

References
1. Winstein, S.; Trifan, D. S. The Structure of the Bicyclo[2,2,1]2-Heptyl
(Norbornyl) Carbonium Ion. J. Am. Chem. Soc. 1949, 71, 2953–2953.
2. Roberts, J. D.; Lee, C. C. The Nature of the Intermediate in the Solvolysis of
Norbornyl Derivatives. J. Am. Chem. Soc. 1951, 73, 5009–5010.
3. Saunders, M.; Schleyer, P. v. R.; Olah, G. A. Stable Carbonium Ions. XI.
The Rate of Hydride Shifts in the 2-Norbornyl Cation. J. Am. Chem. Soc.
1964, 86, 5680–5681.
4. Brown, H. C. Strained Transition States. Spec. Pub.-Chem. Soc. 1962, 16,
140–162.
5. Walling, C. An Innocent Bystander Looks at the 2-Norbornyl Cation. Acc
Chem. Res. 1983, 16, 448–454.
6. Brown, H. C. The 2-Norbornyl Cation Revisited. Acc. Chem. Res. 1986, 19,
34.
7. Wagner, G.; Brickner, W. Ueber die Beziehung der Pinenhaloïdhydrate zu
den Haloïdanhydriden des Borneols. Chem. Ber. 1899, 32, 2302–2325.
8. Meerwein, H.; van Emster, K. Über die Gleichgewichts-Isomerie zwischen
Bornylchlorid, Isobornylchlorid und Camphen-chlorhydrat. Chem. Ber.
1922, 55, 2500–2528.
9. Nevell, T. P.; de Salas, E.; Wilson, C. L. Use of Isotopes in Chemical
Reactions. Part 1. The Mechanism of the Wagner-Meerwein Rearrangement.
Exchange of Radioactive Chlorine and of Deuterium between Camphene
Hydrochloride and Hydrogen Chloride. J. Chem. Soc. 1939, 1188–1199.
10. Brunelle, P.; Sorensen, T. S.; Taeschler, C. The Historic Camphenyl Cation:
A Detailed Structure Evaluation Including Solvation Energy Calculations. J.
Org. Chem. 2001, 66, 7294–7302.
11. Ingold, C. K. Structure and Mechanism in Organic Chemistry; Cornell
University Press: Ithaca, NY, 1953; pp 482−494, 520−523.
12. Brown, H. C. The Nonclassical Ion Problem with commentary by Schleyer,
P. v. R.; Plenum Publishing: New York, NY, 1977; pp 83−100, 123−150.
166
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 13. Yannoni, C. S.; Macho, V.; Myhre, P. C. Resolved 13C NMR Spectra of
Carbonium Ions at Cryogenic Temperatures. The Norborynl Cation at 5 K.
J. Am. Chem. Soc. 1982, 104, 7380–7381.
14. Saunders, M.; Kates, M. R. Isotopic Perturbation Effects on a Single
Averaged NMR Peak: Norbornyl Cation. J. Am. Chem. Soc. 1983, 105,
3571–3573.
15. Raghavachari, K.; Haddon, R. C.; Schleyer, P. v. R.; Schaeffer, H. F.,
III. Effects of Electron Correlation on the Energies of 2-Norbornyl Cation
Structures. Evaluation of the Nonclassical Stabilization Energy. J. Am.
Chem. Soc. 1983, 105, 5915–5917.
16. Olah, G. A.; Liang, G.; Mateescu, G. D.; Riemenschneider, J. L. Stable
Carbocations. CL. Fourier Transform 13C Nuclear Magnetic Resonance and
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

X-ray Photoelectron Spectroscopic Study of the 2-Norbornyl Cation. J. Am.

Chem. Soc. 1973, 95, 8698–8702.
17. Olah, G. A. The σ-Bridged 2-Norbornyl Cation and Its Significance to
Chemistry. Acc. Chem. Res. 1976, 9, 41–52.
18. Brown, H. C.; Liu, K. T. Additions to Bicyclic Olefins. VII. Electrophilic
Additions of Hydrogen Chloride and Deuterium Chloride to Norbornene,
2-Methylenenorbornane, and Related Bicyclic Olefins. Evidence for a
Carbonium Ion Process and the Capture of Unsymmetrical (Classical)
2-Norbornyl Cations. J. Am. Chem. Soc. 1975, 97, 600–610.
19. Kramer, G. M. The Norbornyl Cation: A Reappraisal of its Structure Under
Stable Ion Conditions. Adv. Phys. Org. Chem. 1975, 11, 177–224.
20. Goetz, D. W.; Schlegel, H. B.; Allen, L. C. Ab Initio Electronic Structure
Calculations for Classical and Nonclassical Structures of the 2-Norbornyl
Cation. J. Am. Chem. Soc. 1977, 99, 8118–8120.
21. Clark, D. T.; Cromarty, B. J.; Colling, L. A Theoretical Investigation of the
Core Hole States of the 2-Norbornyl Cation. J. Am. Chem. Soc. 1977, 99,
8120–8121.
22. Montgomery, L. K.; Grendze, M. P.; Huffman, J. C. The Molecular
Structure of a Substituted 2-Norbornyl Cation, 2-Methoxy-1,7,7-
trimethylbicyclo[2.2.1]hept-2-ylium Fluoroborate. J. Am. Chem. Soc. 1987,
109, 4749–4750.
23. Laube, T. X-ray Crystal Structure Analysis of a Symmetrically Bridged
Bishomocyclopropenyl Cation: 2,3-Dimethyl-7-phenyl-2-norbornen-7-
ylium-Hexafluoroantimonate(V). J. Am. Chem. Soc. 1989, 111, 9224–9232.
24. Scholz, F.; Himmel, D.; Heinemann, F. W.; Schleyer, P. v. R.; Meyer, K.;
Krossing, I. Crystal Structure Determination of the Nonclassical 2-Norbornyl
Cation. Science 2013, 341, 62–64.
25. Olah, G. A.; Prakash, G. K. S.; Arvanaghi, M.; Anet, F. A. L. High-Field 1H
and 13C NMR Spectroscopic Study of the 2-Norbornyl Cation. J. Am. Chem.
Soc. 1982, 104, 7105–7108.
26. Olah, G. A.; Prakash, G. K. S.; Saunders, M. Conclusion of the
Classical-Nonclassical Ion Controversy Based on the Structural Study of the
2-Norbornyl Cation. Acc. Chem. Res. 1983, 16, 440–448.
27. Winstein, S.; Clippinger, R. H.; Vogelfanger, E. The Nonclassical Norbornyl
Cation. J. Am. Chem. Soc. 1965, 87, 376–377.

167
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 28. Schleyer, P. v. R.; Watts, W. E.; Fort, R. C., Jr.; Comisarow, M. B.;
Olah, G. A. Stable Carbonium Ions. X. Direct Nuclear Magnetic Resonance
Observation of the 2-Norbornyl Cation. J. Am. Chem. Soc. 1964, 86,
5679–5680.
29. Roberts, J. D.; Lee, C. C.; Saunders, W. H., Jr. Rearrangements in Carbonium
Ion-type Reactions of C14-Labeled Norbornyl Derivatives. J. Am. Chem.
Soc. 1954, 76, 4501–4510.
30. Jalife, S.; Martínez-Guajardo, G.; Zavala-Oseguera, C.; Fernándex-
Herrera, M. A.; Schleyer, P. v. R.; Merino, G. Mechanistic Elucidation of
the 2-Norbornyl to 1,3-Dimethylcyclopentenyl Cation Isomerization. Eur.
J. Org. Chem. 2014, 35, 7955–7959.
31. Mosley, J. D.; Young, J. W.; Agarwal, J.; Schaefer, H. F.; Schleyer, P. v. R.;
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch007

Duncan, M. A. Structural Isomerization of the Gas-Phase 2-Norbornyl Cation

Revealed with Infrared Spectroscopy and Computational Chemistry. Angew.
Chem., Int. Ed. 2014, 53, 5888–5891.
32. Sieber, S.; Schleyer, P. v. R.; Vančik, H.; Mesíc, M.; Sunko, D. E. The
Nature of the 7-Norbornyl Cation and its Rearrangement into the 2-Norbornyl
Cation. Angew. Chem., Int. Ed. 1993, 32, 1604–1606.
33. Winstein, S.; Clippinger, E.; Howe, R.; Vogelfanger, E. The Nonclassical
Norbornyl Cation. J. Am. Chem. Soc. 1965, 87, 376–377.
34. Kirmse, W.; Minkner, D. Stereochemistry of Norbornyl → Norpinyl
Rearrangements. Angew. Chem., Int. Ed. Engl. 1993, 32, 385–387.
35. Kirmse, W. Metastable Norbornyl Cations. Acc. Chem. Res. 1986, 19,
36–41.
36. Winstein, S.; Holness, N. J. Neighboring Carbon and Hydrogen. XVIII.
Solvolysis of the Nopinyl p-Bromobenzenesulfonates. J. Am. Chem. Soc.
1955, 77, 3054–3061.
37. Friedrich, E. C.; Winstein, S. Carbonium Ion Behavior of Nopinyl
Derivatives. J. Am. Chem. Soc. 1964, 86, 2721–2722.
38. Schleyer, P. v. R.; Watts, W. E.; Cupas, C. Estimation of
Nonassisted Solvolysis Rates. The Structures of the “Nopinyl
p-Bromobenzenesulfonates”. J. Am. Chem. Soc. 1964, 86, 2722–2723.
39. Dewar, M. J. S.; Haddon, R. C.; Komornicki, A.; Rzepa, H. Ground States of
Molecules. 34. MINDO/3 Calculations for Nonclassical Ions. J. Am. Chem.
Soc. 1977, 99, 377–385.
40. Valkanas, G. N. Interaction of α-Pinene with Carboxylic Acids. J. Org.
Chem. 1976, 41, 1179–1183.

168
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Chapter 8

From the Ivy League to the Honey Pot

Paul von Rague Schleyer1,3 and Andrew Streitwieser*,2
1Dept. of Chemistry, University of Georgia, Athens, Georgia 30602
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

2Dept. Of Chemistry, University of California, Berkeley,

California 94720-1460
*E-mail: [email protected]
3Deceased Nov. 21, 2014

Paul Schleyer writes of his early life and research, his use
of computational chemistry and his move from Professor at
Princeton University to Professor in Germany at the University
of Erlangen-Nurnberg. His draft was edited and completed by
Andrew Streitwieser.

Introduction
A few decades ago almost two dozen organic chemists were chosen to have
their chemical autobiographies published by the American Chemical Society as
part of a series, “Profiles, Pathways and Dreams” edited by Jeffrey I. Seeman. Paul
Schleyer (Figure 1) was one of those chosen and for this purpose he wrote a draft
that he called “From the Ivy League to the Honey Pot”. Jeff Seeman critiqued
this early draft and suggested, as he did with everyone in this series, numerous
changes, mostly expansions and clarifications. Paul never got back to this project
and his autobiography was never published. When I last had dinner with him, a few
months before his death, he expressed regret that he never finished this work. After
Paul’s death it seemed appropriate, given his extensive contributions to physical
oganic chemistry, to include a memorial to him. What better memorial than one
in his own words? I’ve taken Paul’s unfinished autobiography, made editorial
changes here and there and added appropriate references and figures. I could
not respond to all of Jeff Seeman’s queries but I have included some additional
clarifying material in brackets. The result emphasizes what Paul thought important
and I hope it is a fitting tribute to this remarkable scientist. He and I were friends
for many years. He has stayed in my house in Berkeley on a number of occasions
and I have stayed at his house in Rangen, a hamlet near Erlangen. One disclaimer:

© 2015 American Chemical Society

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 the opinions expressed are those of Paul Schleyer and do not necessarily reflect
my own.
Andrew Streitwieser
University of California, Berkeley
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

Figure 1. Paul Schleyer in 1990 in Erlangen. Photo courtesy of Inge Schleyer

Early Years
I received a small Gilbert chemistry set on my fifth birthday. My mother
overestimated my mental abilities but I do remember heating iron filings together
with sulfur in a spoon over the kitchen gas range. The fact that the iron sulfide
produced was no longer attracted by a magnet interested me less than the blue
170
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 color and the pungent smell of burning sulfur. Since the experiment was described
in the manual, I was delighted to perpetrate this nuisance.
When I was 12, my earnings as a newspaper delivery boy enabled me to buy a
larger Chemcraft set. But I was especially intrigued by the catalog, which showed
beakers, Erlenmeyer flasks, and similar “real” equipment not supplied with the set.
But what could be done with such apparatus? I started reading chemistry books
from the public library, one after the other, voraciously. Like the classical novels
I had been borrowing from the Cleveland Public Library, I reasoned that the older
the book, the better it must be. It soon became obvious, however, that the newer
chemistry texts had more information.
My basement laboratory grew and grew. When the time arrived for high
school, there was no point in taking the junior-year chemistry course, so my
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

mother arranged for me to try the final examination. My grade of 99.5% annoyed
me because I had missed a question. So I started with the senior advanced
chemistry course as a sophomore. This Senior chemistry course was intensive,
3 hours for five days a week, much of it laboratory. We were being trained in
analytical chemistry, and I suspect that my interest in the quantitative aspects of
science originated here. I also graduated from my paper route to jobs in analytical
laboratories. The first was with a water testing outfit (I titrated hard water samples
with standard soap solution until the suds remained after shaking) and then with a
cast iron foundry (I measured the water content of moulding sand by the acetylene
gas pressure developed when CaC2 was added), but I liked the determination of
manganese better, because of the lovely purple color produced.
Having taken just about all the academic courses West Tech had to offer, I
graduated a semester early in the winter of 1946-47 expecting to enter Oberlin with
a scholarship. However, reversion to Fall-only admission led to reconsideration of
my choice of college. My mother thought another part of the country would be a
good experience for me, and she arranged a last minute interview and application
to Princeton (I had never even heard of the “College Boards”). Being high school
valedictorian got me into Harvard as well. I was told that Princeton had the better
chemistry department (wrong) and paid more attention to undergraduates (right).
I decided even then to wait until graduate school for Harvard and set off in the Fall
of 1947 for “the happiest years of my life”. This proved to be no exaggeration:
The Princeton undergraduate years were wonderful and it was a thrill to be able to
return later as an Instructor.

Undergraduate Years at Princeton (1947-1951)

Through some stupidity on the part of my advisor, I did not take chemistry
in my Freshman year. The honors section of Freshman chemistry was a “Gut”
as a sophomore, but material I knew well bored me. I finished the qualitative
analysis laboratory in record time and N.F. Furman awarded me a book as prize.
Although elected President of the ACS, he was a tedious lecturer but Clark Bricker
made analytical chemistry exciting. He and R. B. Woodward were the two most
inspiring teachers that I have had. I did my senior research thesis with Bricker,
171
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 even though the project, which involved the degradation of various hexoses, first
to 5-(hydroxymethyl)furfural and then to levulinic acid, had been suggested by
Eugene Pacsu, a carbohydrate chemist. Despite Bricker’s influence, I saw more
future in organic chemistry which I mastered easily, at least at the level then taught
in Princeton. Only later at Harvard did I learm about “resonance”, that reactions
have mechanisms, and that orientation effects in (1, 2) benzene substitutions have
a logical explanation.
Hugh Scott Taylor had been called from Glasgow to head Princeton’s
Chemistry Department in 1923. His leadership built a center of physical chemistry
which had an established reputation in the 1930’s. Henry Eyring’s departure
for Utah shortly after WWII, and the aging of an inbred faculty led to a sharp
decline. Taylor never had had any use for organic chemistry, and this division
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

was weak. The leading organic chemist, Everett S. Wallis, had extended Lauder
Jones’ demonstrations that migrating groups in various rearrangements retain
their configuration (1,2). He also, as a consultant, contributed to the Merck total
synthesis of steroids (carried out by his former student, L. H. Sarrett (3) ). The
discrepancy between his high opinion of himself as a scientist and reality was
embarrassingly apparent when I returned to Princeton a few years later. However,
that Wallis had not kept up was less important for his academic products. Wallis
was impressive as a lecturer and raconteur. His advanced course on “the 25 ways
to make a carbon-carbon bond” (although given until the end of his life from
the same yellowed note cards) contributed to my decision to become an organic
chemist.
My nemesis was physical chemistry. R.N. Pease’s dry, cramped lectures
spoiled the subject for me. I did well in courses I liked – these required little effort
- but relatively poorly in those I didn’t. I have never developed the self-discipline
to study subjects I don’t find interesting.
Besides their written theses Princeton seniors had to take comprehensive
examinations in four chemical branches: of these, analytical and organic were
easy for me. Although I had not taken a course in inorganic chemistry at
Princeton, my youthful background augmented by further reading sufficed. A
poor grade in physical chemistry reduced my graduation to “magna cum laude”,
but I won the prizes in analytical chemistry and for the best Senior Thesis.
The three top Princeton chemistry seniors of the class of 1951 were accepted
for graduate work to all of the schools to which we had applied, and all three of us
chose Harvard. I was influenced by L.H. Fieser’s organic text (4) , which was the
best of the day, and his book on steroids (5) which was also famous. I learned when
I arrived in Cambridge in September 1951, that other members of the chemistry
faculty were perhaps even more highly regarded by my contemporaries.

Graduate Student at Harvard (1951-1954) and the Return to

Princeton as Instructor (1954)
I arrived back in Princeton as Instructor in the Fall of 1954 without my degree
in hand, or even my thesis well under way. The Harvard faculty at this time was
172
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 lax requiring theses to be finished before departure, and many candidates took
years to complete this task. I did not set the negative record in this respect, but
came close. Although I had only spent 3 years at Harvard, it took nearly that
long again to produce the six hundred page opus I finally presented to the faculty,
essentially a complete review of the bicyclo[2.2.l]heptane literature. I actually
think that Bob Woodward, the second reader, perused the entire manuscript
judging from the number of his detailed comments and corrections. He paid me
the compliment of suggesting that I publish my efforts as a book, but I lacked the
self-confidence to even consider this possibility. Years afterwards, people were
remarking that I had the thickest thesis on the Harvard Chemistry Library shelves,
and I gather that others made good use of my review. Indeed, my only carbon
copy (this was long before the days of Xerox machines) was lent to H.C. Brown.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

He still has it! Although it nearly cost me my job as instructor at Princeton,

my thesis and its preparation taught me a lot. I attempted to treat all aspects
of norbornane chemistry, terpenes, Diels-Alder reactions, physical studies, etc.,
in a more general context, as well as emphasizing the behavior of norbornyl
cations and the beautiful and often highly complex rerangements they undergo.
This was the theme of my research work with Bartlett, a study of the cations
associated with dicyclopentadiene in the various possible stereochemistries. I
never finished the project he assigned, unfortunately. The reason was partly due
to my lack of perseverance, but the project would have been greatly facilitated
by gas chromatography and NMR spectroscopy, both of which had not yet been
invented. Indeed, years afterward, others, who were not aware of my efforts and
those of my predecessor on the problem, published the essential features of what
we were investigating. Bartlett had given me considerable latitude in research,
and I chose to determine the accelerating effect of 1-methyl substituents on the
solvolysis rates of 2-exo and 2-endo-norbornyl systems. I wrote up these results
after arriving back in Princeton, but I had no experience in the construction of a
scientific manuscript. My fellow students at Harvard had the idea that our work
was being carried out, not for publication in the ordinary literature, but for a higher
authority, namely the Harvard faculty who might deign to complete a manuscript
years afterwards. Times were different then. There was less publishing and less
perishing.
I submitted the manuscript to J. Am. Chem. Soc.; the referees were critical
but kind. Nevertheless, I was so discouraged that I never tried to revise the paper
and instead I let it die. I now know that getting the paper accepted would have
been relatively easy, but I had no experience then. I strongly feel that coworkers
should play a direct role in the publication process while they are still in residence
and direct discussion is possible. This is necessary training for future scientists,
who must write proposals, reports, and primary publications. The first chapter of
most of my students’ theses give a thorough review of the background of their
topic; many of these have appeared in the literature. Professors will read the thesis
of their students more carefully, when each chapter is to be submitted as a joint
scientific paper!
Although it is nice to be able to start one’s academic career at age 24, I was
immature and inexperienced. Paul Barlett, rather unwittingly, had gotten me the
job at Princeton. As a member of the Chemistry Department Advisory Board, he

173
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 had berated Princeton for its ingrown faculty. But when asked to recommend an
instructor candidate in the field of physical organic chemistry, he mentioned my
name without remembering my undergraduate Princeton background. Privately,
he had told me that Princeton was pretty much in the chemical doldrums and
that I should accept a postdoctoral opportunity that had arrived from Cal Tech.
I should have. The incumbent of the honorific Fellowship I was offered was given
independence, but I would have had much contact with my scientific hero, Jack
Roberts, whom I had greatly admired from seeing him in action during his MIT
days. But the lure to my alma mater was too great, and I doubted if my car, a
rickety MG TC, would have made it across the country with my wife Barbara and
my first daughter Betti, who had been born at the beginning of my last year at
Harvard. [Paul married Barbara Anne Kinne in 1952. They had three daughters,
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

Betti (b. 1953), Laura (b. 1955) and Karren (b. 1963). They divorced in 1969
shortly before Paul married Inge Venema. Barbara died in 1994.]
The Korean War was also responsible for the chance to return to Princeton.
R. K. Hill, the incumbent instructor responsible for running the undergraduate
organic laboratory, was about to be drafted. After I was hired, Dick told the good
news of his deferment to his chairman, R.N. Pease, A crestfallen face was the
reply. What was the faculty going to do with an extra, unnecessary instructor?
Dick, with whom I had played tennis when we were both at Harvard, and I
always got along well. We never felt that we were in competition for a single slot,
despite evidence to the contrary. In fact, we both achieved tenure. The leisurely
pace up the academic ladder of the Eastern universities in the 1950’s can hardly
be imagined today. After five years as an Instructor, one might be promoted
to Assistant Professor, which was expected to last an additional six years. A
few selected individuals might then become tenured Associate Professors.
Evidently, this was honor enough as many faculty members retired at this rank.
Eventual promotion to a Full Professor was not a forgone eventuality. Princeton’s
Chemistry Department finally had to give up the luxury of an initial Instructor
appointment due to competition from “Middlewestern” universities who actually
started people off as Assistant Professors. My early junior colleagues at Princeton,
F. C. (Sherry) Rowland and E. C. (Ted) Taylor, were more aggressive than I
and applied for research grants even though the senior faculty strongly advised
otherwise. Sherry Rowland left Princeton because of this conservatism but Ted
Taylor suffered the envy of the tenured staff, who also would have liked to have
had postdocs of their own.
Princeton’s Chemistry Department was not an active place in the middle
1950’s. The only outside speakers were supported by the local ACS section,
which meant one organic visitor per semester. Dick, Ted, and I instituted an
organic seminar program with student speakers and whatever faculty we could
get to come without expenses. We regularly attended symposium series held by
the North Jersey ACS Section in Seton Hall and in Philadelphia and transported
carloads of graduate students with us to experience the prominent chemical
personalities of the day. This led to admonishment by the senior faculty whose
students complained about this “requirement”.

174
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 From Adamantane to Dodecahedrane: Cage Hydrocarbons
by Rearrangement
My knowledge of the bicyclo[2.2.l]heptane literature led to an early reward
which helped to establish my career. An obscure paper (6) published in Dutch in
1903 reported the sulphuric acid isomerization of tetrahydrodicyclopentadiene, 1.
A cis-trans isomerization was proposed but the author used an incorrect structure
for 1. Alder and Stein’s elucidation of the structure of dicyclopentadiene in
1931 (7) allowed speculation about the nature of the rearrangement. Until Derek
Barton’s (8) “conformational analysis” trained organic chemists to consider the
geometrical consequences of molecules even bridged ring systems were all too
commonly regarded in the same way they were written: as planar representations.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

Barton had preceded me at Harvard by only a year, but three-dimensional thinking

had taken complete hold by the time I arrived. The new insights into the behavior
of molecules were actively discussed and explored during my graduate student
days. On the basis of such considerations a cis-endo-1 (1a) to cis-exo-1 (1b)
isomerization seemed likely (Scheme 1), and verification of this explanation was
one of the first projects I undertook after my arrival at Princeton.

Scheme 1

Sulphuric acid treatment of 1a gave 1b just as I had expected, but the

isomerization was slow. Aluminum chloride was known to be a much more
effective catalyst for alkane transformations, and indeed led to a vigorous reaction
when added to 1a. Although I could hardly have anticipated it at the time, this
experiment proved to be the beginning of three decades of involvement with cage
hydrocarbon rearrangements.
Princeton’s underequipped Chemistry Department did not even possess a
gas chromatograph in 1955, and tedious fractional distillation, e.g., through a
carefully packed and temperature controlled two meter column, was the only way
to analyze hydrocarbon mixtures. It was more convenient to work on a several
hundred gram scale, even though this meant that distillations would take over
a week and constant supervision was necessary to ensure the maintenance of
equilibrium conditions. My garret laboratory-office in Frick (now condemned
for this purpose by safety regulations) afforded an ideal arrangement since the
column could be set up right next to my desk. At the end of the long run, after
the lower boiling isomer 1a had distilled, a solid formed and clogged the take-off
tube. Expecting this material to be the less stable, higher boiling exo-isomer 1b,
I flamed the take-off capillary with a bunsen burner. Nothing happened. More
175
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 intense heat. Still no results. Direct application of the flame caused the solid to
pop in the tube, but it wouid not melt. The distillation pot was essentially empty,
but several grams of this high melting substance were obtained by washing the
column with ether after it had cooled. I realized that tetrahydrodicyclopentadiene
and adamantane (2), with its notoriously high melting point of 270°, were
isomers, but such a deep-seated rearrangement had no precedent and seemed
quite improbable. Transformation into mechanistically closer lying but more
symmetrical C10H16 isomers seemed much more likely, but how was the structure
to be elucidated? I started by determining the melting point of the unknown
substance in sealed capillary. Amazingly, at well over 200°, the sample still had
not melted, but had sublimed completely onto the cooler upper end of the tube.
I had had much experience in working with volatile solids, but this was unique.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

When the entire sealed part of the capillary was immersed in the heating oil,
a melting point could be determined. This approached the literature value for
adamantane (2) so closely as to constitute a secure identification. But would this
suffice for publication purposes?
Adamantane had been discovered in the early 30’s by Landa and Machacek
(9) and assigned the symmetrical structure 2 primarily on the basis of its
exceptionally high melting point and the unusual cubic crystalline habit. Their
sample was isolated from Czech petroleum. Adamantane is now known to
be present, if only in minute quantities, in virtually every petroleum found
worldwide, but no adamantane samples were available in 1955. A repetition
of Preiog’s rather direct but low yield total synthesis (10, 11) was out of the
question at this point. Fortunately, I was able to locate infrared (12) and mass
spectra of authentic adamantane and my brief report on “The Simple Preparation
of Adamantane” appeared as my second publication in 1957 (13).
I should not really complain about the lack of research facilities at Princeton
in the 50’s. If we had had a gas chromatograph, the adamantane rearrangement
might not have been discovered. Many GC peaks are apparent in the crude
product obtained from A1Cl3 and 1, but none have any guidemarks that would
have encouraged further exploration. I take no pride at having been lucky.
Many scientists have benefited from chance discoveries! However, I believe
chemists should strive to direct and eventually to gain control over what they are
doing to an extent which precludes nature springing unexpected surprises. All
of my subsequent syntheses in the cage hydrocarbon area were deliberate. The
discovery of the antiviral properties of adamantylamine at DuPont was based,
for example, on logical extrapolation of existing leads. This compound and its
relatives, along with adamantane-based high temperature lubricants, provide
the major commercial applications at present. Adamantane is in principle a
cheap chemical. Industrial conditions for the quantitative isomerization of 1a
have been developed in Japan. Unfortunately, I never patented the adamantane
rearrangement. This possibility had actually been suggested to a consulting firm
but their evaluation failed to reveal any reasonable uses, and they declined the
offer. I have been rewarded sufficiently. Every fine chemical catalogue now
lists numerous adamantane derivatives, all derived from my original synthesis.
Perhaps there is an even higher scientific compliment: being taken for granted.
Hardly any current paper dealing with adamantane cites even one of my four

176
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 dozen papers and reviews in the area. The writer of a recent letter wondered if I
could be the same Schleyer who had worked on cage molecules a generation ago!
A new graduate student, Robert Nicolas, exploited adamantane chemistry
successfully, both physical organic and synthetic aspects. Some physical organic
applications will be discussed below. The synthesis of adamantane was quickly
generalized. Various C11H18 and C12H20 precursors provided the next examples of
the adamantane rearrangement by leading to methyl- and dimethyl-adamantanes.
For example, as summarized in Scheme 2, the Diels-Alder product 3 of norbornene
and butadiene (14) was hydrogenated to the saturated hydrocarbon 4, which on
standing with aluminum chloride, gave a quantitative yield of substantially pure
1-methyladamantane, 5 (Scheme 2) (15).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

Scheme 2

“Congressane” was our next achievement. The name originated from the
1963 International Congress of Pure and Applied Chemistry held in London
which featured 6 as a decoration; e.g., on the covers of abstracts, programs, and
publicity material. The foreword to the Handbook explained, “The Congress
Emblem 6 represents a beautifully symmetrical molecule which has not, so far,
been described in the literature. If adamantane 2 is regarded as an “adamantalog”
of cyclohexane then the congress emblem is an adamantalog of adamantane. The
hypothetical process of adamantalogous expansion would provide a family of
compounds all of which contain part of the diamond lattice. Indeed, diamond is an
infinite adamantologue of cyclohexane. The synthesis of the Congress Emblem....
is suggested as a challenging objective for the participants in the Congress.”
To meet this challenge, a C14 pentacyclic precursor was needed. During
a consulting trip to the Union Carbide laboratories in South Charleston, West
Virginia, David Trecker described his successful [2+2]photodimerization of
norbornene in the presence of a copper catalyst. Although the four-membered
ring in this product, 7, was not ideal (too much strain in rearrangement precursors
normally leads to ring opening and tar formation), I took a generous sample
of 7 back to Princeton and suggested to Chris Cupas that he investigate the
rearrangement. The first experiment failed but in the second experiment he
noticed sublimed crystals on the cooler portion of the flask that proved to be
congressane, 6 (Scheme 3). The simple IR and NMR spectra were indicative, but
did not constitute a structure proof. I had heard that Dr. J. Karle and his wife, I. L.
Karle had developed a method of X-ray analysis that did not require the presence
of a heavy atom. They kindly determined the structure of congressane and their
report (16) appeared simultaneously with our paper (17) in 1965.
177
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

Scheme 3

The synthesis (18) of the next homologue, 8, shortly thereafter caused a

nomenclature problem. We proposed the trivial name, “triamantane” (Scheme 3),
which allowed elaboration to still higher members of the series. Following the
same reasoning, “congressane” was redubbed “diamantane”. Sandy Balaban and
I later systematized the whole diamond hydrocarbon family (19). These studies
on diamondoid and related caged systems are just a part of what has turned out
to be a whole research area that soon also involved a number of other research
groups. The subject has been the topic of many reviews only a few of which are
cited here (20–23).

Carbocation Rearrangement Mechanisms

“Arrow pushing” as a speculative means to describe reaction mechanisms had
reached its heyday while I was at Harvard. Some of my contemporaries bragged
that they could write a plausible mechanism for any organic reaction. Indeed, we
searched for the most complicated examples in the literature to test our friends
and to present at the Thursday night Woodward seminars. My speciality was
the really wild carbonium ion transformations in cage ring systems, particularly
those in terpenes. Later, the adamantane rearrangement afforded a challenge
from my own research work. Although, in principle, only three CC bonds have
to be broken/made in order to convert 1a to adamantane, these would involve
unlikely 1,3-carbon shifts. The many other mechanisms one could imagine only
serve to emphasize the complexities one could expect. At first, it seemed unlikely
that the actual mechanism would ever be elucidated, but my early pessimism
proved to be unfounded. In a neat application of graph theory, Whitlock and
Siefkin at Wisconsin systematized the skeletal rearrangement possibilities
in what has been described as a “map of adamantaneland” (24). By actual
count, there are 2,897 different possible pathways to go from 1a to adamantane
2. Some way of narrowing these down to a more manageable number was
178
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 needed. This was accomplished by an early application of empirical force-field
(molecular mechanics) calculations. Alkane isomerizations proceed reversibly
via intermediate carbocations and, hence, are thermodynamically controlled. By
calculating the heats of formation of the possible C10H16 isomers on Whitlock
and Siefkin’s graph, and by considering the strain energies of the carbocation
intermediates, we showed that most of the possible pathways are energetically
unlikely. We then predicted the most likely routes from the various isomers to
adamantane (25). Almost all of these routes have now been confirmed by studies
of the behavior of the various C10H16 isomers synthesized separately. Indeed,
all isomerize to adamantane via the isomers postulated as shown, for example,
by following the reactions gas chromatographically. More recent investiagtion
with carbon labelled materials, carried out by my group and that of the late Prof.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

Camille Ganter [Deceased 2014], ETH, Zurich (23), reveal that the processes are
only somewhat more complex than originally envisaged, partly due to degenerate
rrearrangements that lead to label scrambling.
Our first attempt to learn something about the adamantane mechanism, by
starting with methyl-substitution precursors, failed to give any useful information
because of the equilibration of the two possible methyladamantanes under the
rearrangement conditions. The 1,2-shift of a methyl group from the 1- to the
2- position of the adamantyl cation appears superficially to be a straightforward
carbocation process. As shown in Scheme 4 the rearrangement is accompanied
by the change from a secondary to a tertiary carbocation, albeit at a bridgehead (I
show below that the bridgehead cation is actually quite stable). However, as also
shown in Scheme 4, the dihedral angle between the methyl group (bold bond) and
the vacant orbital is highly unfavorable – the methyl bond lies virtually in the nodal
plane of the empty p-orbital. Instead, the interconversion of the 1- and 2-methyl
carbocations occurs by a skeletal rearrangement in which the methyl and attached
carbon move as a unit. Protoadamantane (so named because it is just before
adamantane on the Whitlock - Siefkin graph) intermediates were demonstrated to
be involved by carbon-labeling experiments. These considerations led to a rather
extensive study of protoadamantane chemistry which is enriched by the lack of
symmetry and the large number of positions each exhibiting a different chemistry
and often degenerate isomerizations (26, 27).

Scheme 4

The advent of superacid chemistry afforded a new way of examining

such rearrangements, by viewing the cationic intermediates directly. The
179
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 1-adamantyl cation is a species that George Olah generated as part of our
long-lasting collaboration (28). The adamantane rearrangement takes place
under such stable ion conditions and both the Sorensen (29) and my group have
observed intermediate ions with different carbon skeletons. Nevertheless, the
1-adamantyl cation refused to undergo even degenerate rrearrangement at the
highest temperatures we could use. Thus, Martin Saunders and I found the barrier
to 1,2-hydride shifts to be at least as high as 30 kcal/mole (30), in sharp contrast to
the usual facile rearrangements and neglible barriers of systems that can achieve
more favorable orbital overlap.
These results show that not just thermodynamic factors control these
reactions. There can be mechanistic bottlenecks. In other words, some of the
needed rearrangement steps are inhibited by barriers too high to be overcome
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

under the reaction conditions, even when the overall reaction is favorable
thermodynamically.

The Move to Erlangen

I have often been asked, “Why did you move back to Germany?” or “How
do you like being back in Germany?” “Back?” Perhaps if my genes had a
memory, I could answer these questions. The implication appears to be that no
native American would leave a chaired Professorship at Princeton for a post in
a little known provincial German university. Although 28% of Americans have
predominantly German ancestry, the assumption that I must have been born in
the old world rather than in Cleveland, Ohio, seems to have taken a persuasive
hold. After my move, I suddenly became “von Schleyer” on both sides of the
Atlantic. It is a not uncommon practice in the United States to honor a relative
(my grandmother, in my case) by christening a child with a family name as middle
name. I never have used “von Rague”, except professionally, but I have been
consistent in employing it or an abbreviation on all my publications, from my
first onwards. Although the “von Rague” has nothing to do with the “Schleyer”,
Germans like titles and it is impossible to convince them to call me simply by my
last name. But to my amazement, “Paul Schleyer” became “Paul von Schleyer”
also to his American colleagues after his move “back” to Germany. I have
complained to my wife, but she responded, “Give the people what they want”.
Inge is a genuine Hessian, but emigrated at the age of 26 never expecting to return.
I accepted the “call” to Erlangen despite, rather than because of her wishes.
Computer time was the main reason. It was literally impossible to develop my
interest in computational chemistry at Princeton because of the restrictive policies
of the administration. Computer time was sold, and at high hourly rates, in order
to make the computer center as self-supporting as possible. Through grant money,
I could perhaps have purchased 10 CPU hours a year. I needed more like 10 hours
per day. At the beginning of the computer age, Princeton treated its facility like a
library, and computer time was free. But nothing ever seems to be free in American
Universities. This meant that costs for the computer were an indirect charge, levied
at some percent of every research grant. Government auditors forced a change to
direct charging, arguing that libraries can be assumed to be used by every research
180
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 project, but not necessarily computers. Kurt Mislow, while he was Chair, took
care of me by allocating a good share of the computer “funny money” allocated to
the Chemistry Department. While this sufficed (but only barely!) for molecular
mechanics calculations, the entire time could hardly make a dent into my emerging
interesting ab initio calculations.
During the 1974-1975 sabbatical year in Munich, many members of my
research group were able to accompany me. Although we were only guests, we
were able not only to use the University of Munich computer, but also the IBM
employed for administrative purposes at the Bavarian Ministry of Culture and
Education on nights and weekends. For the first time, we could almost calculate
“to our hearts’ content” and to use the computer to explore basically new ideas in
chemistry, rather than merely calculating what was essentially known already.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

It was during this time that the offer to come to Erlangen was made. I am a
realist and knew quite well that the inherent advantages of the German system with
regard to the positions and research support provided could not be matched even
by a wealthy American university. The institutional structures simply are different.
My only significant request - two hours of free computer time a week - was turned
down flat by the Princeton’s Dean of the Faculty, even though it was obvious to
my students that the machine was often idle on nights and weekends. What might
have been an agonizing decision thus became quite simple. So I left my Alma
Mater, where I had spent a quarter century as student and teacher, with little regret,
the children from my first marriage, my aged father and a lovely Bucks County
farm behind with considerable regret and moved to Erlangen with a very unhappy
Inge, several cats, and a few members of my small research group. Mainly due to
the elimination of the NIH Postdoctoral Fellowships Program, which had allowed
the recipients to work with whom they wished, my group had shrunk from the
high of about 30 around 1970 to only a handful. I had reckoned that the research
set-up at Erlangen might in its totality be about one and a half times better than at
Princeton. I quickly discovered that my estimate was far wrong. The new situation
was far, far better than I had expected. The universities in Germany support their
Professors rather than vice versa with positions and general funds whereby a rather
large group can be supported without need for any outside research grants. The
infrastructure of a German institute, with many supporting personnel in permanent
state positions, is formidable. The Institute budget should, in principle, cover
research as well as teaching costs, although outside grants are now needed to
supplement this amount.
I had been promised one shift, 40 hours of computer time a week, on the
TR440, albeit that this Telefunken machine was slower than the IBM in use in
Princeton. Eight hours a day was possible because the Erlangen computer center
was only running their computer during the regular work week, so that someone
could always be in attendance. After our shift was added, it was discovered locally
that computers really could be allowed to run unattended over night, and eventually
it was found that this was also possible on weekends. As a consequence, we soon
were using over 70% of .the total University CPU time. We had the lowest priority,
but nobody else was using the machine. Computational chemists can never get
enough CPU time - their demands are like a black hole - but this seemed pretty
near to being in heaven.

181
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Inge had hoped that I would be disappointed with Erlangen, and would
soon seek to move back to Princeton. But the magnitude of my new set-up was
immediately apparent to her. She described the situation to her friends, “He
has fallen into a honey pot”. Indeed, a German Professorship (especially in
Bavaria) is still a formidable institution. My colleagues complain, but I doubt
if they know how well off they are. A recent Chemical and Engineering News
reported on the “unanimity”, of the “bitterness” and “discouragement” expressed
by American chemistry professors with regard to the funding process. It might
be harder for younger scientists to build a reputation in Germany, but established
Professors can still run substantial research operations without the need for
outside funding. German Science Foundation research grants are relatively small,
but are also relatively easy to obtain. The proposals are short and can even be
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

drafted (also the reports) by senior associates. The Chemical Industry Fund is like
the Petroleum Research Fund in the US, but distributes funds to most chemistry
professors automatically, without the need to write any proposals or reports.
There are no overhead charges whatsoever! Grants are partly administered by the
granting agencies, partly by the Institutes themselves, and only to a small extent
by the central university administrations. In Germany, these have not grown
out of proportion as in American institutions. Since all German universities are
supported, it is felt that charging overhead would only put money from one pocket
into another. The German system is too inflexible, but the advantages far outway
the deficiencies.
I can actually work in my office creatively for hours during week-days in the
middle of the semester without being disturbed. A few Americans have moved to
academic positions in Germany but I wonder why more don’t get the idea and
apply. It is true that other factors made the move relatively easy in my case.
I had spent both my 1964-65 and 1974-75 Sabbatical leaves with Rolf Huisgen
in Munich and had been Guest Professor there as well as in Würzburg, Münster
and other European universities. I have always taken the opportunity to travel
professionally and have given an average of about 25 outside lectures annually
over my career. The neat and orderly society of Germany might be boring to some
but not to those who enjoy working in a smoothly functioning and well supported
environment. My publication rate documents the result. The average over the
21 Princeton years was 11 per year, or 18 in the last decade after I became a Full
Professor. The Erlangen average is 25, but I am more proud of the ideas and quality
and breadth of these papers than their number.
My transition to Erlangen was remarkably smooth. My predecessor was
Prof. Gerhard Hesse, a Meerwein student who had made notable contributions
on chromatography during the last part of his career. In the late 1950’s, Hesse
had received the call to Cologne to be Alder’s successor. Erlangen was able to
keep Hesse by offering him a new building for the Organic Institute, which was
finished about a decade before I arrived. Hesse’s carefully thought out design
resulted in an attractive laboratory, with a remarkable number of shops just for
organic chemistry and instruments. Upon my arrival, I inherited Hesse’s generally
well-functioning organisation, and his capable administrator. Dr. Armin Haag.
The frictions which had built up between the personnel assigned to the two
“chairs” dissipated through friendship with my new organic colleague, Hans

182
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Bestmann, who had played a major role in convincing me to come to Erlangen.
I admire Hans for his enthusiasm for research and his interdisciplinary interests.
The move was also facilitated by the good advice of another new colleague, Hans
Hofmann, and by E.D. Jemmis, a member of my Princeton research group, who
had accompanied me to Munich and remained in Germany to set up the computer
programs we were using in Erlangen during the Fall of 1975, my last, “lame
duck” semester in Princeton. Since 1975, Tim Clark was indispensable, first as
a particularly productive member of my research group, and then as a colleague
responsible for the rather complex computer operation which had developed over
the years. I was additionally helped later when I hired Roswitha Völkner in 1989.
With her ability to communicate in German, English and French, she handled
much of my correspondence, travel and papers. I was also able when questioned
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

about administrative matters to frequently reply “Ask Rosi”.

The Organic Institute benefitted from a rich flow of international and national
colleagues most of whom stayed with Inge and me in the “Wagner Villa”, a
spacious mansion decorated with many pictures of conductors, opera singers,
and Wagner family members who appeared annually at the Bayreuth Wagner
Festivals. Inge and I were fortunate to be able to rent this house shortly after
our arrival in Erlangen. Our visitors helped compensate for the geographical
separation of 4 km from our inorganic, physical, and theoretical chemistry
colleagues who were housed in a newer campus-like area in the south part of
Erlangen but transportation fortunately is quite easy. My group had somewhat
different interests than those of my theoretical chemistry colleague, Janos Ladik,
but his many visitors, seminar programs, and fine library were considerable assets.
Otherwise, the library in Erlangen was pretty bad, especially when compared
to Princeton’s superb facility. Books and journals were housed in different
locations; the budget was inadequate and there was no effective coordination;
each Institute decided what to purchase for itself. Indeed, there was no “Chemistry
Department” at Erlangen. A “chemistry group” held meetings, but these were
informal and were only advisory. The organization consisted essentially of the
semi autonomous institutes. Some dechromatization has taken place (less in
Bavaria than in Northern Germany), but the Institutes still tend to be dominated
by the Full Professors.
When I moved to Erlangen, I noticed that my office, although quite large
and pleasant, had a major problem: the door exited directly to the hallway rather
than to my secretary’s office. Without this buffer, I expected from my American
experience that there would be constant traffic into my room and an intolerable
number of interruptions. I instisted that a sign, “Nicht storen”, be installed, which
I could light from a switch at my desk. I never used it. There was no need. Even
during semester workdays, hours could pass before someone came to see me, or
the telephone rang. On nights and weekends when the laboratory was essentially
empty, there is no unwanted interruption at all. The reasons reveal the sociology
of the universities as well as German life in general.
German society tends to be family-oriented and immobile. Most students
go to the nearest university and then settle in the same area where their parents
and earlier ancestors had lived. My foreign colleagues could not believe that
there is no “pecking order” of German universities; with the possible exception

183
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 of some of the newest ones all universities are considered to be equally good.
There is no competition for undergraduate or graduate students. In fact, there
are no undergraduates at all! German secondary education, although essentially
exclusively public, generally achieves an average standard rivalled only by the
better schools in other countries. Students complete their general education in
these Gymnasia at about age 20 having had the equivalent of the first two years
of College. Upon entering a University, a German student specializes in his
major subject immediately. The first degree, the Diplom (or Master’s Degree), is
achieved after four years of study. Many students take longer, even much longer,
and about a year is devoted to the Master’sThesis. Afterwards, most chemistry
students continue to the Doctor’s level with the same advisor working on the same
topic. Although a German Doctor’s Degree is considered to be a higher level
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

than the Ph.D. they are really equivalent. However, it certainly takes longer! Too
many students reach 30 before finishing. The system does not force students to
complete their studies early, and many take advantage of what was once intended
to be intellectual freedom, to draw out the process excessively.
However unfortunate the result may be for a coworker, there are obvious
advantages for a Professor’s research program in having competent and
experienced associates around for a longer time. In addiion, each institute has
a number of Instructor positions (Assistantships), a few of them permanent,
which can be used to support graduate students, postdocs, or even senior research
personnel. In addition, my group benefitted from a number of Habilitation
candidates, i.e., aspirants for the teaching profession. These were something like
nontenured assistant professors, except that a Habilitation (a “superthesis” with
no equivalent in the US) was prepared from independent research results, and no
promotion to a tenured position at the same university was possible.
I have always liked to teach, but the first years at Erlangen were pretty
awful. I lectured in German from the beginning, but continued to write on the
blackboard or on overheads in English to enhance the students’ exposure to
what has become the common language of science. I have never been any good
at languages, and it was a real struggle in the beginning to lecture in German
with a limited vocabulary and range of expression. The chemistry courses at
Erlangen did not have final examinations and were not graded. (I think this is
quite inefficient, since reading assignments are unheeded and most students do
not bother taking notes. However, my overheads were xeroxed and circulated
to those interested.) Hence, if the lecturer does a poor job and loses attention,
students start conversations which creates a background buzz which can become
almost intolerable. As a consequence, my German colleagues tended to put a
great deal of effort into their lectures and, from a didactic viewpoint these might
be better than those of their American colleagues.
Much emphasis was placed on the practical laboratories, which were
conducted in inorganic, organic, and physical chemistry with standards and
duration that are unknown in the United States. The assistants in charge of these
practica had at least a Master’s Degree, and the labs were supervised by an
Instructor with a Doctor’s degree. All of these teaching personnel had their own
research work space directly in the laboratory, so as to be generally available
to the students, not only to answer questions about the experiments but also

184
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 concerning the Professors’ lectures. Although these practica, as well as the
accompanying mini-courses on spectra interpretation, mechanisms, etc., received
a grade, these were not part of the official record, which consisted exclusively
of the marks obtained on set oral hour examinations in the major chemistry
divisions at the Master’s level. Each student was examined by a Full Professor
and a second member of the teaching staff. There was also a similar, but shorter
“pre-examination” about two years earlier. This also was not a part of the final
record, but did help to orient the student (and the faculty) regarding his level of
achievement and promise.
In research, Germany students and the German training system were at their
best. By the time they began research, German students had much laboratory
experience and considerable self-confidence. Since no differentiation was made
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

until the Master’s Thesis, all students were trained equally in organic, physical,
and inorganic chemistry. They took pride in solving experimental problems by
themselves. Repeating an experiment just to get a higher yield than the literature
was typical.

Applied Spectroscopy and Hydrogen Bonding

Early in his career Woodward formulated rules for the interpretation of UV
spectra that enabled the interpretation of chromophores in conjugated systems.
Applied infrared spectroscopy, which had its first major success in the penicillin
project in WWII, was a more general method, since every organic compound has
an infrared spectrum. Detailed structural information could be gleaned from the
exact position of stretching vibrations in particular, but what was possible went far
beyond the generalities found in the literature.
I actually took my first infrared spectrum while an undergraduate at Princeton,
but the cumbersome single channel apparatus was almost useless. It took a day
to run. One had to subtract the spectrum of the solvent, determined separately,
by hand. By my return, Princeton had acquired a Perkin-Elmer model 21, the
standard infrared spectrometer of the day, which, along with a Beckman, I had had
much experience in using at Harvard. Every modern inexpensive replica grating
spectrometer has better resolution, despite the presence of an impressively large
rock salt prism in the older apparatus. In the 3,000 cm-1 region, good resolution
could be obtained with an LiF prism. Luckily, the Chemical Engineering
Department not only possessed such a prism, but also a PE 21 instrument which
was hardly used. Daniel Trifan had made some remarkable observations with
this machine. The infrared spectrum of a ferrocene derivative having an alcohol
side chain showed evidence in the OH stretching region for two new types of
hydrogen bonding: to the iron atom and to the π-electrons of the aromatic ring.
As was typical of the early post WWII period, Dan had had difficulty in finding
a regular academic job, and had settled for a position as an organic chemist in
Princeton’s “plastics lab”. Dan had also obtained his Ph.D. with Paul Bartlett
(before my time), and, as a postdoc, did the original work with Saul Winstein
on the norbornyl cation. Indeed, it was his discovery (31) that the polarimetric
racemization rate of the 2-exo derivative exceeds the solvolysis rate that led to
185
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Winstein’s ion pairing concepts (32). We engaged in joint investigations of the
extent of OH...π hydrogen bonding. It was possible to show that other polarizable
systems, such as cyclopropane rings, also formed hydrogen bonds with proton
donors, and to demonstrate the power of this method as a tool for analyzing
structural details. This soon became one of my major areas of research interest
leading eventually to some 30 papers.
A major contribution was the demonstration that hydrogen bonding was
a much more wide-spread phenomenon than had been realized. For example,
Pauling’s “Nature of the Chemical Bond” states that hydrogen bonds only are
possible with the most electronegative elements, nitrogen, oxygen, and fluorine
(33). Water, ammonia, and HF have abnormally high boiling points whereas their
higher molecular weight congeners, PH3, SH2, and HC1, are gases. My third
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

scientific paper (34), with Bob West of Wisconsin, showed that the elements P, S,
Cl, as well as As, Se, Te, Br, and I, as their alkyl derivatives, function perfectly
well as proton acceptors. It is their inferior proton donor ability that differentiates
them from their first row counterparts. In a later paper with West’s group
(35), these conclusions were extended by thermochemical hydrogen bonding
measurements.
In my group, Adam Allerhand showed that strong hydrogen bonds form
to halide anions (36), and that carbon can function both as proton acceptor
(in isonitriles) and as proton donor (37). At the time, the only C-H hydrogen
bonding known was with chloroform and with acetylenes. We demonstrated that
this phenomenon is much more widespread (38); the presence of two or more
electronegative atoms or substituents can “activate” phenyl and vinyl derivatives,
or heterocyclic rings, for hydrogen bonding. Many implications are still being
studied now because of the obvious importance for solvation
Intramolecular hydrogen bonding, typically between hydroxy groups and
a second function in the molecule, provide an effective way to study structural
details, for example, in the influence of substituents on the populations of
different conformers as well as on the ease of ring formation. In many cases the
configurations could be assigned readily although this might have been difficult
by other methods. The ability to form hydrogen bonds depends on the orientation
of the functional groups; “mapping” of proton acceptor sites could be achieved
in rigid systems in which the OH probe could only reach certain positions of
the proton acceptor (e.g., the π– but not the σ-lone pairs of a carbonyl group).
Fine detail is revealed by this method, even the steric environment of free OH
(i.e., not hydrogen bonded) groups by the number of stretching vibrations and
by their exact frequencies. The “mapping” of proton acceptor sites by means
of intramolecular hydrogen bonding is a forerunner of the modern electrostatic
potential representations of functional groups. Rather than the tedious and limited
experimental procedure of synthesizing many molecules with (more or less)
fixed geometries, it would be much simpler and effective today to carry out such
hydrogen bonding “mapping” computationally. Furthermore, this would also give
information about energies which could be inferred only indirectly from the OH
spectral shifts.
In the 1960’s NMR became an indispensible spectroscopic tool. One of my
best early contributions in this area was with my first post-doctoral coworker,

186
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Pierre Laszlo. Our paper on the NMR spectra of norbornene derivatives (39)
became one of my “citation classics”. We were among the first to obtain 13C-
H coupling constants from proton spectra on compounds with carbon in natural
abundance. This enabled the determination of proton-proton coupling constants
between otherwise identical nuclei (which is why the norbornene paper attracted
so much attention). We employed this method more widely, e.g., to study the
effects of substituents on coupling constants, especially in symmetrical systems.

Solvolysis Reactions
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

From the work of Ingold and Hughes on nucleophilic substitution reactions

in the 1930’s until approximately the death of Saul Winstein in 1969, the
solvolysis reaction engaged the attention of many, indeed, most of the leading
organic chemists of the day. So many insights concerning mechanism and the
nature of electronic interactions in ions and molecules were revealed. It became
fashionable in the 1950’s and 1960’s, particularly in the United States, to introduce
a tosylate group onto an interesting carbon skeleton and measure the reaction
rate. The remarkable results, an enormous range of reactivities often influenced
by seemingly small changes in structure, complicated rearrangements, and the
extension of carbon bonding theory to higher coordination, more than justified
such widespread interest. Substrates with arylsulfonate leaving groups could be
prepared readily from alcohols of known structure. The solvent of choice in the
early days, acetic acid, could be purified easily and gives esters as substitution
products. These in turn could be converted to alcohols which often were easy to
identify.
Although significant work on solvolysis continues today, this area no longer
engages the rapt attention of former years. The reasons are several. The first is
the natural turn to newer fields of interest, the second is the ability of studying
carbocations directly in superacid solution as well as in the gas phase rather
than as fleeting intermediates, and the third is an unfortunate aftermath of the
“norbornyl ion controversy”. If highly regarded chemists could not agree about
the interpretation of the behavior of such a seemingly simple system, what was to
be learned by further investigation of yet more examples? To these three reasons,
a fourth can now be added: the ability of ab initio calculations to provide detailed
information about the geometries, energies, and electronic structures not only of
carbocation minima, but also of transition structures.
Based on salt effects and the effects of substituents, many solvolysis reactions
of sulfonates in acetic acid were thought to have significant carbocation character
and to involve carbocation ion pairs as intermediates (40). My initial studies in this
area were designed to develop criteria for the direct ionization to such carbocation
intermedates without involvement of nucleophilic displacement by solvent; i.e.,
the so-called limiting solvolyses. Solvolyses involving phenonium ions appeared
to be such systems. Cram’s superb early work had established that tosylates with β-
aryl groups solvolyze to give substitution products with retention of configuration
at the reaction center, but with inversion at the rearrangement site, i.e., exactly
187
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 the stereochemistry expected on the basis of bridged phenonium ion intermediates
(41, 42). While participation of the phenyl group was believed to be simultaneous
with ionization, there was a major problem. A participation mechanism requires
rate enhancement, but the introduction of a β-phenyl group often led to a decrease
in the rate. A “numbers game” developed whereby possible steric and inductive
corrections were applied to attempt to explain this discrepancy. Brown (43) argued
that these corrections were small, whereas Cram (44) suggested corrections that
proved to be too high.
We solved this problem in a 1969 set of papers in three different ways
(45–48). The most general approach employed graded Hammett sets of β-aryl
groups. But instead of a single straight line, two intersecting straight lines were
observed. The one of higher slope clearly corresponded to aryl participation since
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

the charge was brought more into the vicinity of the substituent in the phenyl ring,
while the line of lower slope corresponded to the aryl-unassisted reaction with
the solvent. The key to the interpretation was provided by the observation that
the β-aryl rate enhancement (obtained by subtraction of the extrapolation of the
line of lower slope), corresponded exactly to the product distribution. Olefins and
inverted substitution products arose from the reaction with the solvent, whereas
retention of the configuration indicates phenonium ion intermediates. As there is
no cross-over between these mechanisms, no free or weakly solvated carbocations
could intervene. Instead, there must be competition between two strongly assisted
processes, involving β-aryl group participation on the one hand and nucleophilic
solvent participation on the other.
This conclusion provided many new insights, not only with regard to the
phenonium ion problem and the interpretation of neighboring group effects in
general, but also with regard to the mechanism of solvolysis. It had become
apparent that acetic acid is quite far from limiting as a solvent, and that the
solvolyses of many secondary and virtually all primary substrates are probably
essentially more SN2 in character rather than SNl. Interpretations which had
assumed carbocation intermediates now required revision. Earlier observations
that pointed in this direction, such as Streitwieser’s finding of essentially complete
inversion of configuration in secondary (49) and deuterium-substituted primary
solvolyses (50), had been reconciled with the SN1 mechanism by assuming ion
pair intermediates. Those might still be present, but it was now apparent that
considerable bonding to the solvent must be involved. To demonstrate this
directly, a secondary substrate was needed which shows almost no tendency
towards participation of any kind. The 2-adamantyl system seemed ideal for
this purpose. The backside of the leaving group is strongly hindered, and no
participation by neighboring hydrogen (unfavorable torsional angles) or by
neigboring carbon (this would lead from an unstrained to a strained skeleton)
could be expected. Indeed, the solvolysis behavior of 2-adamantyl tosylate in a
variety of solvents paralleled that of tertiary substrates, particularly 1-adamantyl
where no solvent participation from the backside is possible. This constancy
contrasts sharply with 2-adamantyl/isopropyl rate ratio comparisons, where
changes over 107 were observed with changes in solvent. The 2-adamantyl system
also passed numerous other mechanistic tests. These results were presented as a
series of communications in 1970 (51–53). Although it has now been joined by

188
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 other similarly-behaving substrates, 2-adamantyl remains the primary mechanistic
standard for “limiting” solvolysis in secondary systems.
Another important contribution to solvolysis reactions was a modification of
the Grunwald- Winstein solvolysis equation (eq. 1) (54–56).

This equation contains both solvent ionizing power (mY) and solvent
nucleophi1icity (lN) terms, but the latter term was never evaluated. We did so by
means of the simple and reasonable assumption that CH3OTs has m = 0.3 (the
value in AcOH and HCOOH), and 1 = 1.0. The resulting scale was published
in two full papers in 1976 (57, 58). There is still debate over details, but the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

importance of nucleophilic solvent participation is firmly established.

The Interminable Norbornyl Controversy

When facts are in dispute, further experimentation can resolve the matter.
Even when the facts are clear, debates over interpretation may be considerably
more heated. They certainly can be harder to resolve. The recriminating scientific
altercations of 19th century chemists are famous. Definitive data was difficult
to obtain and professional reputations were considered to be of paramount
importance. Even though we pride ourselves on the sophistication of modern
science, the norbornyl cation controversy lasted for over a quarter century. At
least for some it never ended.
The problem can be stated as follows: In reactions of exo-norbornyl
compounds 9 (Scheme 5) expected to produce the carbocation 10a, rearrangement
to the enantiomeric cation 10b takes place presumably via structure 11. The
simple question is then: Is 11 a transition state between 10a and 10b or is it an
intermediate more stable than 10? Structures of the type 11 have been proposed for
a number of other carbocations and are referred to as σ-bridged or non-classical,
the latter because in such structures the number of bonds exceeds the number of
electron pairs available. In their initial paper on norbornyl, Winstein and Trifan
(31) proposed that 11 is relatively stable and explains why 9 is more reactive
than the endo-isomer, an effect that they called anchimeric assistance. H. C.
Brown, however, explained the exo-/endo ratio on steric grounds. Thus was born
the norbornyl controversy that involved a number of chemists and much of the
chemical literature.

Scheme 5

189
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Only part of the norbornyl controversy was exposed to the chemical public
(probably more than they were interested in), but this was a modicum relative
to the information flood that insiders received, in particular, by H.C. Brown’s
privately circulated Norbornyl News Letter. This, and my correspondence with
him, literally occupied a file cabinet of space. In the next century, historians
of science will have a treasure trove of detailed documentation. It is not my
intention to write this history here, but I would like to relate some of my own
involvement. My scientific experiences with Herb Brown on the phenonium ion
were quite positive. He considered my interpretation of the solvolytic behavior
of β-arylalkyl arenesulfonates as a competition between solvent and neighboring
group participation, and subjected it to further experimental test. This convinced
him, and we published the results together (59). However, the phenonium ion
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

can be written as a normal covalent intermediate with a tetracoordinate carbon.

For Brown, the norbornyl cation is in a different category, since five bonds or, at
least partial bonds to carbon are involved. Brown could not really object to such
pentacoordination on theoretical grounds. After all, myriad carboranes and metal
carbide carbonyl clusters have carbon coordination five, six and even eight, but
he was sceptical that such bonding arrangements for hydrocarbon ions could be
viable in solution.
Over the years an overwhelming amount of information built up that supports
Winstein and Trifan’s original contention, but Brown would not concede. In 1975,
Brown suggested that we might debate the norbornyl problem not merely by letter,
but in the form of a book in which all aspects could be aired. In “The Nonclassical
Ion Problem” (60), I was given the opportunity to comment on each of the chapters.
Earlier reviews of the norbornyl problem had all concluded that the ion is bridged,
but Brown not only summarized the data exhaustively, but presented his reasons
against coming to such conclusions. In turn, I tried to expose the weaknesses of
Brown’s arguments and interpretations.
It is not true that Brown locked me into a room next to his office and would
not let me out until my “Comments” were completed. But this rumour has some
foundation. Since my contribution was making slow progress Herb invited me to
come to Purdue to get it done. Herb himself brought me my lunch (a hamburger
from McDonalds) so that I could work through undisturbed. The resulting book
gives a reasonably complete view of the matter at the time of the mid-1970s; but
in the end Brown failed to show that effects other than anchimeric assistance and
bridging are responsible for the behavior of the parent 2-norbornyl system. I will
not go through the many research results since then but will suffice to refer to some
recent reviews of the entire subject (61–66) .
I have often heard the opinion that the interminable norbornyl controversy
“killed” physical organic chemistry. At least in the USA, funding went elsewhere.
But physical organic chemistry merely changed its address and its leading
practitioners have become scientists who are formally associated with other
disciplines. The negative aftermath of the greatest chemical controversy of our
time was outweighed by far by the scientific challenge and stimulation it provided.

190
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 My Beginnings as a Computational Chemist

The structures and energies of molecules have always intrigued me. The
appreciation of the relationship between the geometry of a molecule and its
behavior is the seminal contribution of “conformational analysis”, but the
awareness that “strained” molecules would be more reactive and could exhibit
exceptional behavior is now over a century old. For example, nonclassical
carbocations are associated, with few exceptions, only with strained systems. My
interest in cage molecules obviously required an appreciation of thermodynamic
quantities.
By the 1950’s, the heats of combustion of an extensive set of acyclic and
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

many representative cyclic hydrocarbons had been determined as part of an

American Petroleum Institute project. However, data for only a few polycyclic
hydrocarbons were available. I actually started in my early days at Princeton to
prepare large quantities of the carefully purified samples which were needed for
measurements in specialized laboratories. It quickly became apparent that less
tedious approaches to obtaining thermochemical data would be a blessing. The
same thought applied to the elucidation of the detailed structures of molecules.
I was particularly intrigued by the exact geometries revealed by microwave
spectroscopy, and followed the chemical physics journals where these papers
were published. Victor Laurie, a colleague in this area at Princeton, convinced
me that no rapid accumulation of data from this source could be expected. I am
sure most of my fellow chemists, needing to know a fundamental quantity for
their research, have often suffered the frustration of not being able to find the data
in the experimental literature. The solution to this problem is now obvious. Basic
infomation can now be obtained more readily, and perhaps even more reliably,
by computation than by experiment. But this was not the situation in 1965 when
my interest in computational chemistry began. Gerald J. Gleicher had joined
my group as a postdoc fresh from a similar stay with Michael Dewar in Austin.
Molecular mechanics programs were in their infancy, but held considerable
promise, especially since digital computers were becoming available. Ken
Wiberg kindly gave us his program, the first to include an automatic geometry
optimization routine. Gleicher developed it further and parametrized a force-field
for carbocations. This led to my first computational paper, published early in
1967 (67). The central application was the conformational analysis of bridgehead
carbocations and the program was ideal for this purpose. Paul Bartlett had shown
that l-halobicyclo[2.2.1]heptane derivatives are essentially inert (68), since the
carbocation intermediates would be high in energy because the preferred planar
geometries could not be achieved. Indeed, the difference in strain energies
between a bridged hydrocarbon and the corresponding bridgehead carbocation
provides a reactivity index. This gave a nicely linear plot with the experimental
rate data for the four systems that then were available.

191
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Good theories must be able to make difficult predictions and, to be effective
in chemistry, must be able to do so quantitatively. I would like to quote my 1967
paper: “Although the calculation scheme presented here is by no means in final
form, we have used it to predict the reactivities of other bridgehead systems.”
Starting with this 1967 paper, I have been confident enough not only to make,
but also to publish my predictions. I have tried deliberately to discover examples
which violate the accepted and most fundamental rules (or preconceptions)
involving structures and bonding. The increasing number of experimental
verifications of these predictions has been especially gratifying. The first of these
verifications involved several new solvolysis rates of bridgehead systems that
Gleicher and I had predicted. By 1971, the experimental data had been extended
to 16 bridgehead systems encompassing a relative rate span of nearly 1020. The
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

Princeton molecular mechanics programs had undergone further revision, and

Dick Bingham and I showed how well the new data could be correlated (69).
Further predictions were made and the applicability of the same type of treatment
to free radical reactivities was demonstrated. A little later, in collaboration with
Willy Parker, we showed that reactivities at bridgeheads could be enhanced, rather
than inhibited in middle ring polycyclic systems, since strain could be relieved
in going to planar carbocation or radical centers (70) . The same reasoning
eventually led to our computational discovery of hyperstable bridgehead olefins
which, contrary to Bredt’s rule, favor placement at bridgehead positions (71).
The predictions of bridgehead solvolysis were the first demonstration that
empirical force-field calculations could be applied successfully to reactivity
problems. This emphasized that chemists eventually would have easy-to-use
quantitative computational methods to help solve, and, even more significant, to
help find important problems. These predictions of reactivity have had important
and interesting sequels. The carbocation force-fields we derived were applicable
to secondary, as well as to tertiary systems. But the further development of this
approach has been carried out by others.

John Pople and My Introduction to Quantum Chemistry

In 1969, I had my first substantial contact with John Pople at the Mellon
Institute during an NSF Lectureship in which I described our work on
carbocations. John and his group were just developing the Gaussian 70 program,
and carbocation problems were attractive applications. My first participation in ab
initio calculations on such species actually had begun at Princeton in collaboration
with my colleague L. C. Allen, with Rainer Sustmann, a postdoctoral associate
who had carried out semi empirical computations in Michael Dewar’s group, and
with my first theory graduate student, Jim Williams. However, it soon became
apparent that Pople’s programs were much more efficient, and he was less limited
by computer time restrictions as well. Thus began our 70-paper collaboration
which continued until John Pople’s untimely death in 2004.
I never had any formal theoretical training. Even physical organic chemists
of my day saw little point in learning about quantum mechanics, since this was
unlikely to be anything they would ever use. I served largely as “carbonium ion
192
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 consultant” in the early collaborative work with John Pople. John had always had
an especial ability to couch his explanations in terms his discussion partner could
understand; high level for some, lower level for me. Unlike many theoreticians,
he knew and understood chemistry extremely well. But I was a busy person
with a large research group in the early 1970’s, and calculations were only one
of my interests. Quantum chemistry texts presented masses of equations that
quickly defeated my emerging interest. Like all organic chemistry teachers of
the day, we had acquired a superficial knowledge of qualitative molecular orbital
theory from the Woodward-Hoffmann rules, but these were largely confined to
π-systems. The picture book of Jorgensen and Salem (72) proved to be a revelation
for me. It is said that organic chemists can only understand pictures, rather than
equations. I doubt if this is correct, but the visual presentation of MO’s made
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

obvious and interesting what seemed obscure before. I spent sleepless nights in
extreme excitement literally thinking about molecular orbitals. How might I apply
them to problems in which I was interested? As a teacher, how could I provide
new explanations which heretofore had been couched in other formalisms? During
my sabbatical year in Munich, I developed a course of lectures on qualitative
molecular orbital theory for organic chemists based on slides from the Jorgensen-
Salem book. Wisdom was overlooked in my enthusiasm. There was a great deal
which I should have known, but did not understand.
There was one good way to learn: by calculating. I started to use molecular
orbital theory to design new arrangements of atoms, and to use computations as
an exploratory tool to discover new chemical structures. Ideas were tested in the
computer. If they did not work, there was something I had not understood. A new
job was submitted, and the next day a new result emerged that provided further
insights. Never in my life had I been able to have such rapid turn-around between
an idea and its test. My experience with experimental research was nothing like
this. One could wait for months until a student prepared the starting material
and the key measurement was carried out. Computational chemistry goes to the
heart of the problem directly and provides the key answers almost immediately.
The 1974-1975 sabbatical year was the turning point in my professional life.
Thereafter, theory and computations came first, and experiment followed. But this
is not quite accurate. Computations are also experiments, and I will never lose
the experimentalist’s insistence that results, from whatever source, be subjected
to critical scrutiny and be compared with as many other sources of information
as possible.
Roald Hoffmann’s ideas provided early inspirations. He had examined the
possibility that the electrical effects of substituents might help to stabilize planar
tetracoordinate carbon preferentially. He explored a number of possibilities
by extended Hückel calculations while recognizing the inherent deficiencies
of this method. His work indicated that π-acceptor groups and electropositive
substituents would reduce the energy difference between tetrahedral and planar
methane, but he did not succeed in locating any system that might violate van’t
Hoff stereochemistry. This seemed like an ideal problem for ab initio calculations.
My research group attacked this problem with excitement! What evolved was
the development of an understanding of the best strategies for solving problems
by computational means. We started by examining ideas that seemed attractive

193
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 from qualitative MO considerations. Thus, the isomer of cyclopentadiene with all
hydrogens in the plane would have six π-electrons which should enjoy aromatic
stabilization but it was far from enough to overcome the inherent preference for
the tetrahedral arrangement of the methylene group. This anecdotal approach
was hit or miss. In contrast, John Pople had been studying the effects of F, OH,
NH2 and CH3 (as well as other substituents) on basic organic functional groups.
We extended this approach for the first time to the more electropositive groups,
BH2, BeH, and Li, in order to complete the “first row sweep” of groups with
widely varying electrical properties. This systematic approach was applied to the
planar methane problem with startling results. The energy difference between
“tetrahedral” (C3V) CH3Li and the planar C2v arrangement was reduced by over
100 kcal/mol from the value for methane. When two lithiums were present, as in
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

CH2Li2 a practically planar species resulted. Not only that, but the singlet-triplet
separation also became quite small. Both results were unprecedented for such a
simple methane derivative. We had already ascertained that the planar-tetrahedral
energy difference was reduced significantly in strained systems. After all, planar
tetracoordinate carbon has 90° rather than 110° bond angles. Combining these
effects led to 1,1- dilithiocyclopropane, which was the first molecule actually
calculated to prefer a planar tetracoordinate carbon.
John Pople then made a key suggestion. In some respects, he argued,
cyclopropane and ethylene behave similarly. Perhaps the preferred geometry
of 1,1-dilithioethylene might also be turned around by 90°. Indeed, our
calculations showed the perpendicular form to be more stable than the planar,
even though both retain the C=C double bond! The calculational demonstration
that dilithioacetylene prefers a doubly bridged structure rather than the usual
linear geometry completed this basic series of rule-violating simple molecules.
Before these papers were published, Saul Wolfe invited me to lecture at
a 1978 stereochemistry conference in Kingston, Ontario, Canada. I thought
this would be a splendid opportunity to excite my organic colleagues by our
new findings. What could be more dramatic to stereochemists than a planar
tetracoordinate carbon, perpendicular ethylene, and doubly bridged acetylene?
I expected enormous interest and much discussion, especially as many of the
leaders in the field were in attendance. But I have never met with a colder even
hostile reception - the response was devastatingly negative. No one believed
what I had to say or even seemed interested. A decade later (and after many
confirmations of other of our theoretical predictions), the climate of acceptance
for computational chemistry is changing, but not completely. While Ph.D.’s
trained in computational chemistry are sought after for industrial and national
laboratory posts, academic positions are not available. Instead, these are reserved
for traditional experimentalists, particularly those working in fashionable areas.
Meanwhile, computers have become ever faster and more cost efficient, and
quantum mechanical and other chemical programs more effective. It is time to
train chemistry students for their future that includes computations, and not just
for our non-computational past.

194
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Epilogue
Paul Schleyer’s original manuscript ends at this point, about a quarter century
ago. He continued working in computational chemistry and in 1993 became the
driving force and director of the Computational Chemistry Center in Erlangen.
In 1998 he retired from Erlangen in accordance with German law, but he did
not fade away. Instead, he moved to the University of Georgia in Athens, GA,
as the Graham Perdue Professor and continued to work with graduate students
and post-docs and published prolifically. He collaborated there actively with the
theorist H. F. (Fritz) Schaefer. He continued to amass a number of awards but
the one he desired most was denied him: he was never elected to the National
Academy of Sciences, a fact I consider a discredit to that organization. His
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

sometimes agressive zeal probably turned off too many chemists but I believe
they should have separated their personal feelings from an appreciation of his
scientific accomplishments. Paul and I had our own deep differences in some
interpretations of theory but these did not affect our friendship (73). A photograph
of the two of us together is shown as Figure 2. Paul’s legacy as one of the pioneers
in computational organic chemistry is to be found in the now common use of
computational results even in papers whose main thrust is the synthesis of natural
products.
Paul Schleyer died of a heart problem on November 21, 2014.

Figure 2. Paul Schleyer and Andrew Streitwieser in 1989.

195
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Acknowledgments
I thank the many people who contributed their thoughts and interactions about
Paul Schleyer. I am especially indebted to Roswitha Völkner, Schleyer’s longtime
Erlangen secretary who provided me with a 60-page draft of Paul’s manuscript,
including many penciled in comments and questions by Jeff Seeman.
Andrew Streitwieser
University of California, Berkeley

References
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

1. Jones, L. W.; Wallis, E. S. J. Am. Chem. Soc. 1926, 48, 169–181.

2. Wallis, E. S.; Nagel, S. C. J. Am. Chem. Soc. 1931, 53, 2787–2791.
3. Sarett, L. H.; Wallis, E. S. Ann. Rev. Biochem. 1947, 16, 655–688.
4. Fieser, L. F.; Fieser, M. Organic Chemistry; D. C. Heath and Co.: Boston,
1944.
5. Fieser, L. F.; Fieser, M. Natural Profucts Related to Phenanthrene, 3rd ed.;
Reinhold Publishing Corp.: New York, NY, 1949.
6. Eijkman, J. F. Chem. Weekblad 1903, 1, 7–12 (Chem. Zentralblatt 1903 II,
1989).
7. Alder, K.; Stein, G.; Finzenhagen, H. Liebigs Ann. Chem. 1931, 485,
223–246.
8. Derek H. R. Barton (1918 - 1998) shared the 1969 Nobel Prize in chemistry
for his work on conformational analysis first published as a short paper in
Experientia 1950, 6, 316−320.
9. Landa, S.; Machacek, V. Collect. Czech. Chem. Commun. 1933, 5, 1–5.
10. Prelog, V.; Seiwerth, R. Ber. 1941, 74, 1644–1648.
11. Prelog, V.; Seiwerth, R. Ber. 1941, 74, 1769–1772.
12. Obtained from Dr. R. A. Dean of the British Petroleum Co., Ltd.
13. Schleyer, P. v. R. J. Am. Chem. Soc. 1957, 79, 3292–3292.
14. Alder, K.; Monch, J.; Wirtz, H. Liebigs Ann. 1959, 47–59.
15. Schleyer, P. v. R.; Nicholas, R. D. Tetrahedron Lett. 1961, 2, 305–309.
16. Karle, I. L.; Karle, J. J. Am. Chem. Soc. 1965, 87, 918–920.
17. Cupas, C.; Schleyer, P. v. R.; Trecker, D. J. J. Am. Chem. Soc. 1965, 87,
917–918.
18. Williams, V. Z., Jr.; Schleyer, P. V. R.; Gleicher, G. J.; Rodewald, L. B. J.
Am. Chem. Soc. 1966, 88, 3862–3863.
19. Balaban, A. T.; Schleyer, P. v. R. Tetrahedron 1978, 34, 3599–3609.
20. Fort, R. C. Adamantane : the chemistry of diamond molecules; M. Dekker:
New York, 1976.
21. Bingham, R. C.; Schleyer, P. v. R. Topics Curr. Chem. 1971, 18, 1–102.
22. Olah, G. A. Carbocations and Electrophilic Reactions of Cage Hydrocarbons.
In Cage Hydrocarbons; Olah, G. A., Ed.; Wiley: New York, 1990; pp
103−153.
23. Osawa, E.; Yonemitsu, O. Carbocyclic cage compounds : chemistry and
applications; VCH: New York, 1992; p 409.
196
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 24. Whitlock, H. W., Jr.; Siefken, M. W. J. Am. Chem. Soc. 1968, 90,
4929–4939.
25. Osawa, E.; Tahara, Y.; Togashi, A.; Iizuka, T.; Tanaka, N.; Kan, T.;
Farcasiu, D.; Kent, G. J.; Engler, E. M.; Schleyer, P. v. R. J. Org. Chem.
1982, 47, 1923–1932.
26. Lenoir, D.; Mison, P.; Hyson, E.; Schleyer, P. V. R.; Saunders, M. J. Am.
Chem. Soc. 1974, 96, 2157–2164.
27. Schleyer, P. V. R.; Lenoir, D.; Glaser, R.; Mison, P. J. Org. Chem. 1971, 36,
1821–1826.
28. Schleyer, P. v. R.; Fort, R. C., Jr.; Watts, W. E.; Comisarow, M. B.; Olah, G.
A. J. Am. Chem. Soc. 1964, 86, 4195–4197.
29. Sorensen, T. S.; Whitworth, S. M. The Superacid Route to 1-Adamantyl
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

Cation. In Cage Hydrocarbons; Olah, G. A., Ed.; Wiley: New York, 1990;
pp 65−101.
30. Vogel, P.; Saunders, M.; Thielecke, W.; Schleyer, P. v. R. Tetrahedron Lett.
1971, 1429–1430.
31. Winstein, S.; Trifan, D. J. Am. Chem. Soc. 1952, 74, 1154–1160.
32. Winstein, S.; Clippinger, E.; Fainberg, H.; Heck, R.; Robinson, G. C. J. Am.
Chem. Soc. 1956, 78, 328–335.
33. Pauling, L. The Nature of the Chemical Bond; Cornell University Press:
Ithaca, NY, 1960.
34. Schleyer, P. v. R.; West, R. J. Am. Chem. Soc. 1959, 81, 3164–3165.
35. West, R.; Powell, D. L.; Whatley, L. S.; Lee, M. K. T.; Schleyer, P. v. R. J.
Am. Chem. Soc. 1962, 84, 3221–3222.
36. Allerhand, A.; Schleyer, P. v. R. J. Am. Chem. Soc. 1963, 85, 1233–1237.
37. Allerhand, A.; Schleyer, P. v. R. J. Am. Chem. Soc. 1963, 85, 866–870.
38. Allerhand, A.; Schleyer, P. v. R. J. Am. Chem. Soc. 1963, 85, 1715–1723.
39. Laszlo, P.; Schleyer, P. v. R. J. Am. Chem. Soc. 1964, 86, 1171–1179.
40. Streitwieser, A. Solvolytic Displacement Reactions; McGraw Hill Book
Company, Inc.: New York, NY, 1962.
41. Cram, D. J. J. Am. Chem. Soc. 1949, 71, 3875–3883.
42. Cram, D. J. J. Am. Chem. Soc. 1952, 74, 2129–2137.
43. Brown, H. C.; Morgan, K. J.; Chloupek, F. J. J. Am. Chem. Soc. 1965, 87,
2137–2153.
44. Cram, D. J. J. Am. Chem. Soc. 1964, 86, 3767–3772.
45. Schleyer, P. v. R.; Lancelot, C. J. J. Am. Chem. Soc. 1969, 91, 4297–4300.
46. Lancelot, C. J.; Schleyer, P. v. R. J. Am. Chem. Soc. 1969, 91, 4291–4294.
47. Lancelot, C. J.; Harper, J. J.; Schleyer, P. v. R. J. Am. Chem. Soc. 1969, 91,
4294–4296.
48. Lancelot, C. J.; Schleyer, P. v. R. J. Am. Chem. Soc. 1969, 91, 4296–4297.
49. Streitwieser, A., Jr.; Walsh, T. D. Tetrahedron Lett 1963, 1, 27–34.
50. Streitwieser, A., Jr.; Wolfe, J. R.; Schaeffer, W. D. Tetrahedron 1959, 6,
338–344.
51. Fry, J. L.; Harris, J. M.; Bingham, R. C.; Schleyer, P. v. R. J. Am. Chem.
Soc. 1970, 92, 2540–2542.

197
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 52. Fry, J. L.; Lancelot, C. J.; Lam, L. K. M.; Harris, J. M.; Bingham, R. C.;
Raber, D. J.; Hall, R. E.; Schleyer, P. v. R. J. Am. Chem. Soc. 1970, 92,
2538–2540.
53. Schleyer, P. v. R.; Fry, J. L.; Lam, L. K. M.; Lancelot, C. J. J. Am. Chem.
Soc. 1970, 92, 2542–2544.
54. Grunwald, E.; Winstein, S. J. Am. Chem. Soc. 1948, 70, 846–854.
55. Winstein, S.; Grunwald, E.; Jones, H. W. J. Am. Chem. Soc. 1951, 73,
2700–2707.
56. Winstein, S.; Fainberg, A. H.; Grunwald, E. J. Am. Chem. Soc. 1957, 79,
4146–4155.
57. Bentley, T. W.; Schleyer, P. v. R. J. Am. Chem. Soc. 1976, 98, 7658–7666.
58. Schadt, F. L.; Bentley, T. W.; Schleyer, P. v. R. J. Am. Chem. Soc. 1976, 98,
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch008

7667–7675.
59. Brown, H. C.; Vander Jagt, D. L.; Schleyer, P. V. R.; Fort, R. C., Jr.; Watts, W.
E. J. Am. Chem. Soc. 1969, 91, 6848–6850.
60. Brown, H. C. The Nonclassical Ion Problem; Plenum Press: New York, NY,
1977; p 292.
61. Creary, X. Advances in Carbocation Chemistry; Jai Press Inc.: Greenwich,
CT, 1989; Vol. 1, p 253.
62. Coxon, J. M. Advances in Carbocation Chemistry; Jai Press Inc.: Greenwich,
CT, 1995; Vol. 2, p 273.
63. Prakash, G. K. S.; Schleyer, P. v. R. Stable Carbocation Chemistry; John
Wiley & Sons, Inc.: New York, 1997; p 587.
64. Olah, G. A.; Prakash, G. K. S. Carbocation Chemistry; John Wiley and Sons,
Inc.: New York, 2004; p 393.
65. Moss, R. A. J. Phys. Org. Chem. 2014, 27, 374–379.
66. Anslyn, E. V.; Dougherty, D. A. Modern Physical Organic Chemistry;
University Science Books: Sausalito, CA, 2006.
67. Gleicher, G. J.; Schleyer, P. v. R. J. Am. Chem. Soc. 1967, 89, 582–593.
68. Bartlett, P. D.; Knox, L. H. J. Am. Chem. Soc. 1939, 61, 3184–3192.
69. Schleyer, P. V. R.; Bingham, R. C. J. Am. Chem. Soc. 1971, 93, 3189–3199.
70. Parker, W.; Tranter, R. L.; Watt, C. I. F.; Chang, L. W. K.; Schleyer, P. v. R.
J. Am. Chem. Soc. 1974, 96, 7121–7122.
71. Maier, W. F.; Schleyer, P. V. R. J. Am. Chem. Soc. 1981, 103, 1891–1900.
72. Jorgensen, W. L.; Salem, L. The Organic Chemist’s Book of Orbitals;
Academic Press: New York, 1973.
73. Streitwieser, A. A Lifetime of Synergy with Theory and Experiment; American
Chemical Society: Washington, DC, 1996; p 164.

198
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Chapter 9

Lost in the Funhouse. Life in the Research

Laboratory of William von Eggers Doering
Ronald M. Magid*,1 and Maitland Jones, Jr.*,2
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

1The
University of Tennessee (retired), 7496 North Creek Loop NW,
Gig Harbor, Washington 98335
2Department of Chemistry, New York University, New York, New York 10003
*E-mail: [email protected]; [email protected]

William von Eggers Doering was one of the most influential

organic chemists of the 20th century. In this article, two of his
students from the late 50s-early 60s reflect on Doering’s research
accomplishments, while simultaneously describing the silliness,
pranks, and amusements that young adults are wont to engage
in.

One possible ending would be to have Ambrose come across another

person lost in the dark. They’d match their wits together against the
funhouse, struggle like Ulysses past obstacle after obstacle, help and
encourage each other.
John Barth (1968), Lost in the Funhouse

In many ways, college students of the 1950s were like poor Ambrose, fearful
of what lay ahead but, at the same time, eager to take on the challenges. Their
apprehension was real, but could be ameliorated by their identifying with others
who were in a similar situation. In this memoir, the authors (who have been
friends since 1955) relate some of their experiences as undergraduate and graduate
students of one of the most brilliant chemists of the 20th century.
William von Eggers Doering, Figure 1, (“Bill,” once one got brave
enough to call him that) received his undergraduate and graduate degrees at
Harvard; in his career, he served as a faculty member at Columbia, at Yale,
and at Harvard. Upon the occasion of his 90th birthday, a celebration of
some of his major research accomplishments was written by Frank-Gerritt

© 2015 American Chemical Society

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Klärner (with some input by the present authors) and published in 2009 (1).
Subsequently, Mait Jones and Ron Magid produced two additional documents
(http://web.utk.edu/~rmagid/index5.html) to provide some personal insights
into what it was like to have been associated with Bill during these incredibly
productive years. Those manuscripts (unpublished and, frankly, unpublishable)
serve as the basis for the chapter that you are now reading.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

Figure 1. Bill Doering at a reunion with students in Cambridge, 2009.

Photograph courtesy of Ronald Magid.

For incidents and stories that involved the two of us, the plural pronoun “we”
will be used. But for those that involved only one, we will take our guidance
from the current practice wherein professional athletes refer to themselves in the
third person ... and the most famous of whom are identified by only a first name
(LeBron, Peyton, Kareem, etc.). Thus, these sections will be third-person accounts
featuring either “Ron” or “Mait.”
It is not our intention to use this forum to give detailed accounts of Bill
Doering’s research accomplishments. Suffice it to say that in addition to the
topics covered by Klärner, et al. (non-benzenoid aromatic compounds, carbene
chemistry, and thermal rearrangements of hydrocarbons), Bill had the knack of
doing key experiments that would “open” a field ... and then move on to other
interests, leaving these new fields to be explored by others. The breadth of the
subjects that intrigued him is astounding. Included are Baeyer-Villiger oxidation;
asymmetric induction; the Doering-Zeiss mechanism for solvolysis; [2 +2] and
200
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 [4 + 2] cycloadditions; stabilization of closed-shell dienes, trienes, and polyenes;
the behavior of diradicals; and many more.
Mait and Ron are the co-authors of this chapter. Both of us were Yale
undergraduates in Bill’s introductory organic course in Spring-Fall, 1957, and
both did our graduate work under Bill’s supervision. Mait knew Bill well before
he even knew that chemistry existed, and he worked several summers at Hickrill
Chemical Research Laboratory (about which much more, shortly) in Katonah,
New York, an easy 40-mile drive north from Columbia’s campus. His parents
lived nearby, knew Bill socially, and were eager to find a job to keep their son off
the street, or at least off the tennis court. Bill was kind enough to take on a totally
inexperienced kid in a variety of probably make-work jobs. Much later, during
his years on the Princeton faculty, Mait maintained a closer relationship with his
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

mentor.
Most organic chemists will know of the absolutely fabulous work that came
from that place – much of it the result of Bill’s collaboration with the great,
underappreciated Lawrence H. Knox – the tropylium ion and much of the early
carbene work are only two examples. Hickrill had an ever-changing staff of
postdocs, many of them from Europe, all overseen by Doering and a resident,
“permanent” postdoc (the afore-mentioned Larry Knox). How could such a place
come into being? When Doering joined Columbia in his first faculty position,
there was, of course, a period of time when he was low man on the totem pole.
At that time there was an unusual graduate student at Columbia, a middle-aged
woman, Ruth Alice Weil, who had gone back to graduate school in chemistry.
She found that the world of science didn’t exactly embrace such people with
open arms, and wound up in the research group of the lowest-ranking person,
Bill Doering. At some point, at what we used to call a bull-session (ahem), the
question of “what would be the ideal home for research?” came up. When it
came Bill’s turn to speak, he replied that he thought a smallish institute, with 5-10
postdocs, a resident majordomo, and secure funding would be about right. Some
time after the meeting, she came up to him and asked, “Did you mean it? if so, I’ll
speak to my husband about it.” It turned out that they had zillions of dollars and
they did it. Now we are not sure what the take-home lesson is here, but we find
it delicious to imagine all of the big daddies of the time walking the halls of the
Columbia chemistry department smiting their foreheads. It’s a wonderful image.
Hickrill certainly was “state of the art” at the time, but seems remarkably
primitive now. Instrumentation consisted of a Beckman UV, an early infrared
spectrometer, a refractometer, and….that’s it. There was a wonderful library,
looted, as were some of the labs, during clandestine nighttime raids from the Yale
campus in New Haven, CT (just 60 miles from Katonah) after Hickrill closed.
Times were changing, though, and one of the early projects that Mait can recall
was the construction of a gas chromatograph – a preparative model, with a 2”
wide column – and essentially no resolving power at all. It could hardly compete
with the two-story fractional distillation column that ran from the basement to the
second floor – and was packed by Mait at Doering’s command, “One glass helix
at a time!” Mait cheated.
The “wet” chemistry at Hickrill – and Yale – was also far different from that of
today. For example, the scale of things was vastly larger. Need cycloheptatriene?

201
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 You couldn’t buy it in those days, and you needed more than the fractions of
milligrams we require today. The solution was to dissolve vast quantities of
diazomethane (explosive, carcinogenic, and poisonous) in 12 L of benzene and
blaze away with General Electric Sunlamps purchased at the hardware store. Low
tech it may have been, but it made the molecule. The occasional fires, fueled by
12 L of burning benzene, are best not mentioned. Diazomethane was made from
the notoriously dangerous and carcinogenic N-nitroso-N-methylurea (NMU) with
50% aqueous KOH (never NaOH, which was rumored to detonate diazomethane
– or was it the other way round?). NMU itself was required in kilo amounts,
and was synthesized in a metal garbage pail that was stirred with an old baseball
bat; the solid NMU would rise to the surface and would be scooped away with
a beaker in an ungloved hand. At Hickrill, it was Mait’s job to do that, as he
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

seemed not very sensitive to NMU, and he recalls bending over that hideous vat,
which was bubbling, spitting, and emitting who knows what noxious molecules.
At Yale, every new grad student was given this opportunity to achieve fame. Can
you imagine that today? Everyone would be in jail.
Mait’s role at Hickrill was something less than gopher, and he owes those
folks, Larry Knox in particular, a great deal for putting up with him. The effect
on a young person is easy to imagine. He didn’t know what anyone was really
doing, and he vividly recalls looking at the blackboard and trying to work out all
the crossing lines of norbornanes and what they meant. Nonetheless, the effect on
him was enormous – literally life-determining, and that may stand as a metaphor
for the impact Hickrill had on many people and, indeed, for the impact Bill Doering
has had on so many of his students and associates.
The “miracle that was Hickrill, alas, came to an ignominious end. At about
the time that Mait and Ron entered grad school, Ruth Alice Weil died and her
husband at that time decided that the community would be better served if the lab
were sold to The Harvey School, a local private school. (One is reminded of the
French judge’s comment as he denied Lavoisier’s appeal of his death sentence,
“The Republic has no use for savants.” But we digress.) We were participants
in one raid (of several) to see what could be “liberated” from the lab so that it
might serve a higher purpose than as a playground for the benefit of rich kids.
One of the items taken was the Perkin-Elmer Vapor Fractometer which had, if
one recalls correctly, serial number 0001. Our memory on this is fuzzy, but we
remember switching its serial number with one from a machine at Yale so that the
new owners would mistakenly believe that this instrument did not belong to them.
(We assume that the statute of limitations has long passed. At least we hope it
has.) We don’t recall what other mischief we did that day or what we might have
loaded into the back of Bill’s big Mercedes, but Ron has a very vivid recollection
of being petrified, not by our having skirted the law, but of riding “shotgun” in
the front seat (no seat belts, of course) with Bill driving on back-country roads,
holding a hamburger in one hand and a beer in the other, steering with his knees.
Obviously we survived, but there were definitely moments of doubt.
Like Hickrill, and of much broader impact in that it affected hundreds of
students, Doering’s sophomore organic course at Yale was another powerful
magnet attracting young people to organic chemistry, even those who were
somewhat turned off by their previous experiences with chemistry. Ron, for

202
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 example, was definitely put off by Yale’s introductory freshman course for chem
majors – it was classical, descriptive, lacking in any theory, and dull – there
was no excitement, no sense of passion whatsoever. He had been juggling the
possibility of majoring in physics or math or chemistry – by the end of his
freshman year, he was ready to commit to one of the first two, but Mait convinced
him to defer his decision for another semester or two. The first term of our
sophomore year featured quantitative analysis; as Yale had (and still has) no
analytical chemists on the faculty, the teaching of this course fell to the youngest
physical chemist on the staff, who enjoyed being in the classroom and lab almost
as little as we did. That solidified Ron’s decision to get out, but Mait, who knew
that Ron would be enthralled by Doering’s course the following semester, urged
him to stick it out. And so he did. And he fell in love – with the subject matter
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

and with the man who let it unfold in glorious technicolor (at least five different
colored pieces of chalk, along with an extra-long eraser, were employed every
lecture) with witty dialog and extraordinary insights. We think that Bill loved
teaching the undergrads as much as we loved being in his class. We soaked up the
lessons and stories, bordering in many cases on the libelous. We were treated to
remarkably candid comments about some famous organic chemists: “His head is
filled with homogenized goose feathers” was said about one chemist and “Thank
goodness he’s dead” referred to another, whom Doering also disparaged as “the
maker of the last of the nonanes.” We don’t recall who the lucky person was
whom Bill described as “a throw-back to a Neanderthal” and perhaps it’s just
as well. We ate it up, as undergrads will, and as it was intended to be. Many
alumni of Doering’s Chem 29/35 went on to become professors of chemistry
themselves; it is a near-certainty that their lecture notes, style of delivery, and
blackboard technique are patterned so closely after Doering’s that the fine line
between flattery and plagiarism may well have been breached.
As for Ron’s reluctance to choose to major in chemistry, it was the prodding of
Mait that convinced him: first, to endure the quantitative analysis semester in Fall
1956 and second, to get past the required textbook (Fieser’s Organic Chemistry)
for Bill’s Spring semester course. After a general chemistry course that had been
more engineering than science (who can forget the Bessemer steel process or the
Frasch process for extracting sulfur from the ground?), Ron recalls having opened
the organic text while at the bookstore and seeing pictures of industrial distillation
towers and chemical processing plants. Once again, he wanted to bail out. But
Mait was persistent and so both of us enrolled in Chem. 29 in Spring 1957. (It
was not the Yale chemistry department’s soundest pedagogical decision to split the
two-semester organic sequence between Spring and Fall semesters with a summer
term in between, but that’s what was done. So it goes.) Ron recalls the first class
meeting: a short, dark-complected, rather young man wearing a tie that was much
too long for him began to lecture without having introduced himself to the class.
Ron wondered, “This is Doering? Surely not.” It wasn’t until the second class
meeting that the lecturer sought approval from a somewhat older man, whose
presence I had not earlier noted, in the back of the room and asked, “Isn’t that
right, Bill?” Mait, of course, knew all along that the lecturer was not Doering. In
fact, it was a new assistant professor, Martin Saunders. At the third lecture, Bill
took over and led us the rest of the way. (When he told us that he would take a

203
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 mechanistic approach, Ron’s juvenile mind registered “mechanistic = mechanics
= industrial processes = engineering” and he almost bailed out again.) He regrets
that he no longer has his notes from that class – in their stead, he has the notes that
he took as a grad student, sitting in on that same undergraduate class, that became
the basis for his own organic chemistry lectures when he became a professor.
As for the Fieser text, it proved to be of more value as a reference work
than as a guide to learning. In truth, we have used it many times whenever we
needed a key fact about a compound or reaction. Bill’s rationale for requiring
it was “Usually the only thing that students get out of their organic course is
the textbook.” The book was irrelevant simply because Bill’s Chem. 29/35
courses were purely lecture and 100% mechanisms-based. It wasn’t until Ron
began teaching Rice University’s introductory organic course, in 1966, that he
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

discovered that the textbook by Morrison and Boyd mirrored many of Bill’s
approaches to topics – is it possible that he had a clandestine pre-publication copy
that he never shared with us?
Did we know that even in his late 30’s he was already one of the leading
organic chemists of the world? (Figure 2) Like other undergrads, most of us had
very little understanding of the nature of academic research. And, in striking
contrast to the open-door policy that Mait and Ron employed throughout their
teaching careers, we don’t recall ever having visited Doering in his office to ask
questions. We suppose that he had office hours, but for some reason none of us (so
far as we knew) took advantage of them. Ron doubts that Doering knew his name
or those of most of his classmates by the time the course ended.

Figure 2. The young Bill Doering in 1963. Photograph courtesy of Ruth Haines
Doering Zeiss.

204
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Ron’s other major interaction with him, prior to grad school, came during
a semester of senior research. In retrospect, he realizes that the project that
Doering entrusted to him (and that he was too inexperienced and inept to execute)
was not only very sophisticated but also significant – it was to produce the
parent cyclopropenylium ion, C3H3+, as a test of the Hückel prediction that a
2π-electron system would be aromatic. (At that time, no such compound had
been made.) The synthetic design was straight-forward: Diels-Alder reaction
of dimethyl acetylenedicarboxylate with 7-cyanotropilidene (reacting via its
norcaradiene valence tautomer) would produce a tricyclic adduct whose retro
Diels-Alder cleavage of different bonds would yield dimethyl phthalate and
3-cyanocyclopropene; reaction of the latter with silver ion should produce AgCN
and the desired ion. Ron writes, “I encountered several significant hang-ups.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

Neither Bill nor our TA (Tony Vellturo, a member of Gary Griffin’s group)
told me that the acetylenic diester was commercially available, so I synthesized
it from fumaric acid by (1) bromination, (2) double dehydrobromination, and
(3) methylation. According to the literature procedure that I had found, the
bromination was to be carried out in a round-bottomed flask using a Herschberg
stirrer – little did I know that the twisted Nichrome wire paddle would be
consumed by bromine, thus precluding any stirring action. Furthermore, although
I did not know it ahead of time, I soon discovered that the acetylenic diester
was a nasty lachrymator. But I soldiered on. Next, I was to beg the legendary
Horst Prinzbach, the first of Bill’s brilliant German post-docs at Yale, for a
sample of precious tropylium ion that I was to treat with cyanide ion to make the
cyanotropilidene; somehow (memory fails me on this) I was unsuccessful here,
so the project ground to a halt, but still I managed to get my B.S. degree.”
Mait recalls his own senior research project, “Does methylene react with the
carbon-carbon bonds of spiropentane?” Answer, “No.” But he would not like to
revisit the experiments he did to show that! Spiropentane was made by the zinc-
dust mediated debromination of pentaerythrityl tetrabromide (look it up), and that
process inevitably yields zinc bromide, which is Lewis acidic enough to rearrange
spiropentane to methylenecyclobutane, a sure precursor of the putative insertion
product of reaction of methylene with spiropentane. He supposes the fact that he
found no reaction is comforting. Years later, he went back and did this work better
in several respects. There was a special issue of Tetrahedon (July, 1997) dedicated
to Bill, and it seemed appropriate to revisit this reaction in a proper way. First
of all, spiropentane is not the best molecule to use as there is little strain relief in
expanding the three-membered ring to a four (well, there are those bent bonds....);
in its favor, however, spiropentane does possess the virtue of having the highest
ratio of C-C to C-H bonds of any hydrocarbon. Anyway Mait persuaded Guoxian
Wu to look at the reaction of methylene with spiro[3.3]heptane. He did the whole
reaction in a few weeks, and we put Bill’s name and the late Larry Knox’s on the
paper (2). Mait still feels good about that: “I owe Knox the world and was proud
to have another paper with him, even if he couldn’t know about it.”
In 1959, Mait and Ron moved on to graduate school, but the questions that
we had as undergrads could still be asked: What do graduate students know about
their advisor, their co-workers, or even their research projects? Answer: A lot
and not a lot. And much of what they “know” is wrong or on the fringes of the

205
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 ridiculous. We knew that Doering was a very important person in the world of
organic chemistry – his publications, his many absences to give symposia, his
friends in the scientific community, etc. told us that. We also knew about some
of our chemical “ancestors,” for example his earlier spectacularly outstanding
group of graduate students at Columbia, among whom were people whose
publications we were reading: Jerry Berson, Andy Streitwieser, Ken Wiberg,
and Chuck DePuy, among others. And we were surrounded by an extraordinary
group of post-doctorals, most of them German, who would go on to brilliant
independent careers when they left Yale; among these were Horst Prinzbach,
Gerhard Schröder, Wolfgang Roth, Wolfgang Kirmse, Wolfram Grimme, and
Gerhard Klumpp.
Our research group was far more homogeneous than those of today. True,
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

there were European – mostly German – postdocs, but very few from Asia, and
almost no women. In truth, the situation was little different from that encountered
by Ruth Alice Weil at Columbia! There were some exceptions: postdoc Ray
Okamoto came from Japan and gave his name to a unit used in the group. He
was such a wizard at fractional crystallization, that laborious and time-consuming
technique, that Peter Gaspar coined a term “the okamoto” to refer to the number of
such operations that Ray could accomplish in one day. The rest of us could aspire,
on our good days, to accomplish, at best, a milli- or a micro-okamoto. Jean Day
Lassila was the only female graduate student in our group, a couple of years ahead
of us, and not only married but pregnant. She delivered her daughter, Kathrin Day
Lassila, on a Saturday, immediately after passing her seventh cumulative exam,
and was rapidly back in the lab. There seems to have been no deleterious effect
on that daughter – she is now, four decades later, the editor of the Yale Alumni
Magazine. She wrote to Ron, “Now that I have two children of my own and a full-
time job, I often wonder how in earth she managed it. She always insisted I was
no trouble at all.” Jean would bring her infant daughter to departmental seminars;
the child would awaken and scream only when the speaker was notoriously bad.
We knew of Doering’s close friendships with R. B. Woodward, dating from
the quinine synthesis in the 1940s, and with Saul Winstein. Here are a few
personal memories. In the undergraduate course, Doering devoted a couple of
lectures to the story of the synthesis of quinine. We reveled in the personal
touches, such as “Woodward ran exactly one reaction” and “When I reported
that a key compound had the wrong melting point, Woodward went directly to
bed; this was on Good Friday. Purification by crystallization led to the right
melting point, which I reported to Woodward on Easter Sunday.” Woodward was
probably the only person about whom Doering never said a bad word. He was
in awe of Woodward’s encyclopedic knowledge and his imagination in synthetic
design. Ron recalls that Woodward came to Yale to give a lecture on his recently
completed synthesis of chlorophyll: “I happened to be walking down a hallway
when I glanced to my left and saw the great man, standing at the blackboard of
the room where the seminar was to be, painstakingly and meticulously drawing
the target structure on the board – chalk in one hand, wet sponge in the other (to
expunge any bond or group that was not perfect), cigarette dangling from his
lower lip.”

206
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 At the 1966 Sheffield Symposium on aromaticity, Woodward reported that
Doering had told him that heptafulvalene undergoes cycloaddition with TCNE.
“Now, when he called to tell me about the result, Doering, who is a very playful
fellow, said: ‘Of course it adds at the 2 and 2′ positions!’ Those of you who like
to count will have observed already that addition in that sense represents a [14 +
2] combination, and at first sight you will stigmatize that as bad. ... The important
fact which Doering withheld (until I had discerned the solution of the dilemma) is
that the adduct (3)” is formed by antara addition to the 14-electron component, an
allowed process.
Winstein spent a semester at Yale (Fall, 1960) and gave two lectures per
week: one on organic chemical topics (solvolysis mechanisms, non-classical
ions, and homoaromaticity – do any of these topics appear in today’s courses?)
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

alternating with one about the physical chemistry of ionic solutions. Yale’s
faculty happened to have some of the world’s leaders in this latter subject, most
notably Lars Onsager. At one point, someone asked Winstein a question, causing
him to go to the board to carry out a calculation. Just as he began, Lars (who had
probably been dozing) sprung to life and said “Vell, I tink de answer is 6,221.”
Some minutes later, Winstein (not having heard Onsager) finished his calculation
and announced, “There it is - 6,221!” Two memorable quotations from Winstein,
uttered more than once during that semester were, “Heh, heh – of course the only
lab work I do now-a-days is in my bar at home” and (speaking about the modest
exo/endo rate ratio for 2-norbornyl tosylate) “These are not the factors of 106 that
we like, but we have to take what we get.” A final vignette: at one of the weekly
departmental organic chemistry seminars (given by postdocs and by grad students
in their 2nd, 3rd, ... nth years) the speaker was our post-doctoral colleague Gerhard
Schröder. I don’t recall the topic, but Winstein asked a question at the end of the
talk. Schröder walked over to a portable blackboard and turned it back to front –
there was the answer in full. Said Winstein to Doering, “You set me up for this,
didn’t you?”
A personally painful moment came when Ron was giving one of those weekly
departmental seminars – his topic was “The Benzilic Acid Rearrangement,” for
which he had read every pertinent article and review. As he was presenting the
arguments of some very prominent chemists, Doering interrupted and asked,
pointedly, “Haven’t you ever heard of the Curtin-Hammett Principle?” “The
WHAT?” “The Curtin-Hammett Principle!” Silence. “Perhaps someone in the
audience could inform Mr. Magid what it says?” Dead silence. Bill’s behavior,
harsh on its face, had its designed pedagogic value, because Ron made it a point
to learn this principle and, when he became a professor, to pound it into the
heads of the students in his graduate courses. And, of course, he had the smug
satisfaction of finding occasional Curtin-Hammett errors in published work and
of putting them on exams.
We organic grad students had only limited contact with Onsager, but stories
about him were spread throughout the department. It would be a few years before
he would win the Nobel Prize, but that mattered not. Onsager and John Gamble
Kirkwood, his good friend and fellow physical chemist, are buried side-by-side
in the Grove Street Cemetery on the Yale campus. Kirkwood’s tombstone has
some twenty lines of print about his accomplishments. As the ultimate example

207
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 of “one-upsmanship,” Onsager’s has only three lines: • Gibbs Professor • Nobel
Laureate • Etc.
What else did we “know”? We knew very little about Bill’s personal life,
although rumors circulated widely and wildly. We knew that he drove a big
Mercedes, always. We knew that he would attend our weekly group seminars
if he was in town – and that when he wasn’t, the seminar would go on anyway,
often under the guidance of one of the postdocs. Those meetings were remarkable
for their range and power – and for their robustness. They did not flag when
Doering wasn’t there, as in later days he often was not. Perhaps this strength is
not too surprising, given the presence of such major intellects as Horst Prinzbach,
Wolfgang Kirmse, and Wolfgang Roth, but the staying power of these meetings
is also surely a tribute to their originator. When Bill did attend, he did not suffer
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

fools gladly (an understatement!) – if one of us was less than fully prepared to
give his/her talk, Bill made it abundantly clear that that was never to happen
again. And, of course, it didn’t. His insistence on careful preparation was a
hallmark of his research style, imparted to all of us in the clearest of terms: do a
thorough literature search; give credit to others where credit is due; make certain
that the facts are as you report them; check and check again before making any
claims; re-do the measurements without any prejudice about how you “want”
them to come out; establish the identity of a compound by an IR spectrum that is
within a pen’s width of the spectrum of the known material; and so on.
Given that he was one of the founders of physical organic chemistry, it’s
ironic that he spent much of his academic career in the study of “no-mechanism
reactions” – but his justification of this seeming “absurdity” makes all the sense
in the world: “ ‘No-mechanism’ is the designation given, half in jest, half in
desperation, to ‘thermo-reorganization’ reactions ... in which modern, mechanistic
scrutiny discloses insensitivity to catalysis, little response to changes in medium
and no involvement of common intermediates, such as carbanions, free radicals,
carbonium ions, and carbenes (4).”
Through his behavior, more than through his words, he conveyed to us that
the most important quality that a research scientist possesses is integrity. If we
lose our trust in one another, then we have lost everything. When one publishes
a paper, the expectation is that a chemist anywhere in the world will be able to
reproduce the result. When one draws conclusions from the data, they must be
sound and within the bounds of the reasonable – but if the result demands a novel
explanation or a new theory, then the experimentalist must be triply sure that the
results are as claimed. We knew of feuds that he had with two well-known chemists
concerning primacy in one instance (the synthesis of tropylium ion) and probity
in another (the validity of experiments on the stereochemistry of dihalocarbene
addition). Even worse, there were two instances of falsified results within our
laboratory: one occurred before our time at Yale (but was still whispered about) but
the other occurred while we were there; for the latter, Doering had to publish (5) an
embarrassing retraction some years later: “W.v.E.D. belatedly takes an opportunity
to accept the evidence for the irreproducibility of the alleged realization ... and
apologizes to those whose careful work has made the case.”
Doering could wield a poison pen, but he could also be “playful” when
criticizing others. Consider these statements: in 1955, “It seemed most surprising

208
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 to us (although not to Johnson or Nozoe and Sato) ... (6),” in 1990, “The values
obtained by Dewar and Wade from ‘two measurements’ ... are in spectacular
agreement – ‘wondrous strange (7).’ ” and “Humbled, WvED apologizes to
Dewar and Wade for ever having had doubts, but not without admitting to
speculation that the sage of Austin is on-line to the Delphic oracle; possesses the
ultimate computational program, which a secretive US Air Force has requested
not be revealed; and is a deserved beneficiary of Pasteur’s Postulate ‘Chance
favors the prepared mind.’ (7) ”
Let us elaborate on Bill’s seemingly innocuous barbs (7) directed at Michael
James Steuart Dewar. The rather witty retort (above) was actually a fiendish (one
of Bill’s favorite words) scheme for discrediting the work and its authors. Doering
quotes a passage from a 1987 paper by Dewar and Jie: “However, there are in
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

fact good reasons for regarding the published experimental values with suspicion.
The activation parameters ... were derived from just two measurements at two
temperatures ....” and uses it to cast doubts on experimental work published in
1972 by Dewar and Wade. Thus, Bill sets the stage for what is to follow by
letting Dewar criticize his own work. Bill concludes the paragraph with “Prudence
called for a repetition.” [No, whatever you’re thinking, Prudence was not one of
Bill’s adoring harem.] And he couldn’t resist plunging and twisting the knife a
little more by inserting, as a footnote, an exact quotation from Wade’s dissertation
with a relatively minor grammatical error highlighted by sic: “In his defense,
Wade’s comment should be heard: ‘No claim as to the accuracy of these values
is made since only two rate measurements were used to calculate them .... The
standard deviations for these rates was (sic) reasonably good.’ ” In fact, when Bill
and his coworkers measured the rate at five different temperatures, they obtained
precise values for the activation parameters, leading to this already-cited passage
“The values obtained by Dewar and Wade from ‘two measurements’ ... are in
spectacular agreement – ‘wondrous strange’.” To this is appended the footnote
“Humbled, WvED apologizes ...” that was quoted in full above.
Often he wrote with elegance and style, as in 1956: “Although the word
‘carbene’ has been defined (‘A bitumen soluble in carbon disulfide but not in
carbon tetrachloride’) ... one appears justified in assuming the innoxious risk
that the new meaning become known in the same, presumably chemical, circle
presently aware of the old. In its new meaning (collaboratively conceived by
W. von E. Doering, S. Winstein and R. B. Woodward in a nocturnal Chicago
taxi and later delivered diurnally in Boston) ‘carbene’ is synonymous with one
of the several definitions of ‘methylene’ ... (8)”; and in 2001: “Theoretical
efforts to elucidate the thermal behavior of the pristinely unsubstituted archetypal
cyclobutane in the gas phase have been extensive. At the experimental level,
intrepid purists have explored these questions with deuterium as a minimal
perturbation, while soldiering pragmatists have introduced more drastic
perturbations in the hope of achieving answers (9).”
Yale used a cume (cumulative exam) system, modeled after Harvard’s, to
assess the progress of the grad students. These exams, on unannounced topics in
synthesis, mechanism, whatever were given once a month on Saturday mornings
throughout the academic year. The requirement was to pass eight exams, which
most of us did by the end of our second years. One exam, however, sticks in our

209
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 collective memories like a rotten apple. The cume on December, 5, 1959 (the
third in our first semester of graduate work) was clearly written by Doering: no
other faculty member used this font (in those electric typewriter days) nor did
any other draw such precise structures. A polycyclic structure with a plethora of
oxygenated functional groups was shown to undergo a fairly extensive cleavage
with ethanolic KOH. Further, the keto group of the reactant could be reduced to a
pair of epimeric alcohols, each of which also reacted with ethanolic KOH in ways
entirely different from each other and from the original ketone. The challenge
was to write mechanisms for the three reactions that would also explain why they
differed so dramatically when only a small change had been made. Fair enough,
except ... The formula given for the reactant, C26H28O8, did not correspond to
the structure drawn so elegantly on the exam page. That is, if the dashed lines
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

on the structure corresponded to methyls, the carbon count was wrong; and if they
referred to hydrogens, the count was still wrong. Only one person passed the exam,
a man who was spending only his first year at Yale before going to CalTech for his
degree. Over the years, Mait and Ron would try to solve the exam – and although
we probably got close enough that our answers would have given us a passing
grade, we never were satisfied with our attempts. Then, in 1977, Ron received a
hand-scrawled note from Mait: “See, instantly, Experientia ....” Mait had found
the very article on which the question had been based! He also wrote “Note that
not only was the formula wrong, but the structure as well.” Thanks a hell of a lot,
Bill! In 2008, we made this paragraph available to him and thought, “You may be
reading this story for the first time, but had you known about it we think that you
would have been pleased with the turmoil you created.” [This cume problem also
had a life in the classroom. On a final exam in his Princeton undergraduate course,
Mait prefaced a question thus: “The first section of this question graced last year’s
final, and is part of a series of reactions that the legendary R. M. Magid and I have
been thinking about – not constantly, we are not obsessive – for decades.”]
As we’ve said, our weekly group seminars were amazing institutions. We
numbered 20-25 (grad students and post-docs) and were divided into groups
of four or five. Each week, on a rotating basis, one of the groups would read
the set of journals assigned to it and write abstracts of those articles that would
interest the rest of us; one member of the group would pose some mechanistic or
synthesis problems; and two or three members of the group would prepare talks.
It was at these sessions that we believe that we learned most of our chemistry.
[A frightening event, the Cuban missile crisis, was unfolding in October, 1962,
and became the sub-text of one evening’s seminar at which it was very difficult
to concentrate on chemistry.] Bill made it a point to attend these gatherings, but
when he couldn’t one of the post-docs or older grad students would keep things
in line. In our fourth year, however, Mait and Ron noted that the spirit of the
group was waning, that there wasn’t much enthusiasm at the seminars, and that
attendance was sporadic. So with the assistance of Phil Shevlin (a second-year
student whom we recruited to add a semblance of respectability to our nefarious
plan) we decided to prepare a totally fake seminar! Everything was wrong: the
date on the abstract sheet; the spelling of Phil’s last name; the quotation by
Christine Keeler (who was much in the news at the time and who might well have
said “There are times in a man’s life when he must subvert truth to the pursuits

210
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 of pleasure”). The abstracts reported some barely plausible results, authored
by such luminaries as a journeyman catcher for the Cincinnati Reds, various
characters from “Catch-22,” and the real Russian names of Lenin, Trotsky, and
Stalin. The concluding abstract reported on a rat sex attractant whose structure
looked remarkably like ... well, a rat. Phil gave some invented problems and
Mait reported on some nonsensical carbene work. For Ron’s talk, a letter from
his soon-to-be postdoctoral advisor, Eugene van Tamelen, ostensibly “arrived”
that very morning – “he” described some work then going on in his lab as a
prelude to what Ron would be working on. According to what “he” wrote (all
of it imaginary), his student had isolated a compound, C10H10, whose melting
point and spectral properties immediately told us it was bullvalene but which the
Stanford chemists had misidentified because Schröder’s paper had not yet been
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

published. Of course, we chose a day for our charade when Doering would be
out of town. To our horror, he announced in the morning that he would be in
attendance, so Mait and Ron (clothed in trepidation) went to his office to alert
him as to what we were planning to do. And, of course, he never showed! The
good news (or bad news, depending on one’s perspective) is that we pulled it off
without a hitch. A few days later, one of our fellow graduate students told Mait
that he was unable to find one of the references in the library, so Mait gave him a
“corrected” page number and sent him off again. A few days later, we revealed
what we had done. Many of our colleagues were quite unhappy with us.
Doering would have preferred that all of us live monastic lives, tethered to our
laboratory benches, leaving only when hunger or sleep became compelling. But
times were changing and, if memory serves correctly, more than half of us were
married and some of us even had children. So it goes.
The bullvalene story is an amazing one. Doering had just returned from a
European trip in time to come to the group seminar on October 9, 1961. It was
unusual for him to preempt someone who had planned to speak, but this time he
took chalk in hand and outlined for us the nature of his thinking. He began by
reviewing the work that he had done with Wolfgang Roth on the degenerate Cope
rearrangement of 3,4-homotropilidene. (An aside – many of us suspected that he
loved to use the phrase “degenerate Cope” because of its alternate meaning.) This
led him to speculate about what would happen if the molecule were locked in its
unfavorable cisoid conformation, the one needed for Cope reaction, by a bridge
linking the two methylene carbons, as in barbaralane (named for post-doc Barbara
Ferrier, who made it). And if that bridge were a -CH=CH- unit, then the resulting
C10H10 would have some spectacular properties – not only would all three of its
degenerate Copes lead to the same structure, but through a series of such reactions
every carbon would become like every other, every hydrogen like every other.
The prediction: if the Copes were fast enough, there would be a single peak in
the 1H and 13C NMR spectra! Doering showed how 1,2-carbon interchanges and
1,3-carbon interchanges could occur. And he confessed that the man sitting next
to him on the return flight, watching his wild scribblings, was convinced that he
must be mad. Well ...
Ron doesn’t recall (nor do his notes reveal) if Doering had already decided
that a single peak, if it occurred, would indicate that 1,209,600 identical structures
were in equilibrium, although he did ponder the question of whether a single peak

211
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 would indicate fast degenerate Copes or, just as interesting, a single averaged
structure. Lars Onsager provided a simple answer: it would have to be the former
because, “as everybody knows,” one cannot put 10 points symmetrically about the
surface of a sphere (i.e., there are only five “Platonic Solids” and the 10-vertex
structure is not one of them.) It was at this group seminar (or maybe a day or two
later – memories are cloudy on this) that either Ron or Mait blurted out the name
bullvalene for this hypothetical molecule. As related by Nickon and Silversmith,
“Which of the two students was the blurtor and which was the blurtee remains
unsettled; but their term caught on with the research group. And everyone was
surprised when Doering and Roth formally adopted it in their seminal publications
in 1963 (10).”
Fact: everyone was not just “surprised” but astonished that Doering knew the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

name we had given it. In a 1967 article, he inserted a footnote: “The precise origin
of this ignoble name, at the hands of incredulous graduate students, is lost, but their
initial response to its conception and the possibility of its existence is suggested
in E. Partridge, A Dictionary of Slang and Unconventional English p. 104 (11).”
This is what one calls a red herring – Doering knew that this was not the origin
and, as fate would have it, Mait had the opportunity to publish his own footnote in
1970: “As one of those quoted in Footnote 15 of ref 5, it is perhaps not out of order
to suggest that the source of this ‘ignoble name’ is neither lost nor contained in E.
Partridge... (12).” And there the matter remained until Nickon and Silversmith
published a full explanation: “It seems that during Professor Doering’s early days
on the faculty at Columbia University, his research students nicknamed him ‘The
Bull’ ... Perhaps it was because ‘Bill Doering’ sounds like ‘Bull Durham,’ a
popular brand of tobacco; but other explanations can be found. Dr. Doering and his
nickname subsequently transferred to Yale University, and it was there at a group
seminar on October 9, 1961 ... (10)” (The second part of the name was chosen
so that bullvalene would rhyme with fulvalene, a molecule of great interest to our
research group in those days of non-benzenoid aromatic compounds.)
Well, what about Doering’s nickname? As recently as June of 2007, Bill
professed (or pretended) not to know the origin. In an email to a German colleague,
he wrote that Fred Chang, a graduate student in Louis Fieser’s group, in the early
1940s, had nicknamed him “ ‘Ole Bool’, the meaning behind which escapes me.”
And yet ... We never knew if he knew that we called him “the bull” (which is why
we were so shocked by his adopting the moniker bullvalene), although there was an
occasion on which he cautioned us about something or other with the words Cave
Taurum (we may have butchered the declension). And, in fact, there was a pouch
of Bull Durham tobacco in the lab that had been handed down from generation to
generation. We wonder what’s become of it.
Bill told the undergrads a story from his own years as a grad student. Many of
his friends (including Fred Chang whom he nicknamed Red Fang) were in Louis
Fieser’s group. One summer, Mary and Louis Fieser decided that they wanted to
have an outdoor barbeque pit and so they requisitioned slave labor (i.e., Louis’s
grad students) to build it. Bill, who was not in Fieser’s research group, joined in the
“fun.” Throughout the hot summer days that they labored, never once were they
invited inside the great man’s house. Finally, when the construction was finished,

212
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Mary said “Louis, invite the boys inside for a glass of water.” Bill told us, “I was
so mad that I peed all over the barbeque!”
So, back to the bullvalene story. Bill recruited some grad students and
postdocs for a crash program to synthesize the compound by a rational path. Ron
once asked him why he published the idea of the molecule before he had made it –
he answered that so many people were then doing hydrocarbon chemistry that he
feared that someone would stumble across it. And that’s precisely what happened
(as is related in the article by Klärner, et al. (1)) when Gerhard Schröder, at that
time an independent researcher in Belgium, irradiated a cyclooctatetraene dimer
(of then unknown structure), which cleaved to produce benzene and bullvalene.
Ron had the good or bad fortune to have been in Bill’s office one day in May,
1963, desperately trying to conclude his research and to schedule his orals, when
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

Doering’s secretary brought in the day’s mail. “Oh, good, a letter from Gerhard,”
he said as he opened it and proceeded to ignore his anxious grad student. It would
be an exaggeration to say that steam poured from his ears and that his facial
colors went from red to purple but he finally composed himself, smiled, and said,
“Gerhard has made bullvalene. And it has one peak in the NMR.” Nevertheless,
he couldn’t resist (no he couldn’t!) publishing part of that letter as evidence of
Schröder’s “contrition” at having made this legendary compound. Here is a loose
translation from the German (italics added for emphasis): “For the last weeks and
months, I have been busy with the structure determination of fluxional molecules
C16H16 and C12H12. Thereby, it was completely unwanted and unexpected that
‘Your Bullvalene’ fell into my hands (11).”
In the 1950s, modern instrumentation was making its way into the organic
research lab. Yale had a Cary UV spectrometer and a plodding Perkin-Elmer
Model 221 IR spectrometer, to be supplemented by a speedier model, the 421.
We also had one of the first commercial NMR spectrometers: the 60 MHz
Varian V-4300 that nearly required an electrical engineering degree to operate;
maintenance of a homogenous magnetic field was controlled by a galvanometer
(called a “super-stabilizer”) that sat atop the machine and that would go haywire
on the least provocation: a truck entering the city, a bird flapping its wings, etc.
Later, we got a Varian A60, so all of us were able to run our own NMR spectra
routinely. (That the spectrometers were in the only room in the building, aside
from Bill’s office, to be air-conditioned, made their use an all-day project in the
hot summer months.) In the research laboratory, we had a Varian Aerograph A90
and other gas chromatographs (or g.l.p.c., gas-liquid partition chromatographs, as
Bill insisted we call them). There was a Perkin-Elmer 154 “Vapor Fractometer”
for preparative separations, the commercial descendent of that home-made GC at
Hickrill that had 2” wide columns! In one of the hoods was an acid bath, heated
with a Meeker burner – woe unto the person who put a piece of glassware coated
with silicone grease into that fuming cauldron. The recipe for this monster was:
fill the bath with sulfuric acid, add a few mL of nitric acid, and throw in a penny.
Heat till boiling. There was a melting point apparatus with a series of very precise
Anschütz thermometers (each covering a range of only 25 °C). And there was a
locked cabinet filled with specialized ground-glass equipment, much of it (or so
it was rumored) constructed by Bill in his guise as Master Glassblower. Money
was alleged to be short, so we all had to sign out “exotic” glassware as needed.

213
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Naturally, someone had to keep the books and act as policeman – not an enviable
task, at least for Mait! Because many of us worked with volatile materials, there
was not only an elaborate vacuum line with diffusion pump but also a series
of portable pump tables, shared by two or three students, based on a design in
Wiberg’s “Laboratory Technique in Organic Chemistry” (an essential source,
published in 1960, for all sorts of experimental insights).
Doering made very few visits to the research lab, but when he did “miracles”
would occur (some good, some bad). Ron was carrying out a five-step literature
prep and he recalls having had a terrible time trying to crystallize one of the
products; Doering wandered by, took the flask in hand, said some magic words,
and crystals rained down. He then set Ron the task of crystallizing 1 mg of benzoic
acid from diethyl ether, a task made fiendishly difficult by the small quantity, by its
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

high solubility, and by the narrow range between boiling and room temperature for
this solvent; after much practice (and using a micro filtration device that someone
showed him how to make) he succeeded. A “bad miracle” occurred when Ron’s
lab bench neighbor, Marty Pomerantz, having mistranslated a German language
procedure to read “all at once” (instead of “little by little”), added a huge slug
of concentrated sulfuric acid to a reaction mixture; he happened to have walked
away from his bench when Doering passed by the apparatus at the very moment
that acidic green noxious clouds emanated from the top of the condenser.
Doering could get “bees in his bonnet” during those visits and the results were
nearly always dire. For example, the sight of an unattended GC would set off the
“Helium is a precious natural resource!” reflex in Doering and he would demand
that we all switch to hydrogen as carrier gas. Never mind that all our previous data
would become worthless, we had to do it. Solution? Wait a while and clandestinely
switch back.
“Meerwein Ponndorf,” a mythical postdoc, figures in many of the stories
from the Yale years. One of our favorite reactions, partly because Doering
had elucidated its mechanism but mostly because of its exotic name, was the
Meerwein-Ponndorf-Verley Reduction a.k.a. the Oppenauer Oxidation a.k.a. the
MPVO Equilibration. Mait and Ron were the instigators of a prank that we played
on one of Bill’s well-meaning but terribly gullible secretaries. Whenever the
great man would leave town (which was often), Meerwein Ponndorf (for whom
Ron provided the German accent) would telephone and demand to speak to him.
When told that Doering was away, Ponndorf would get angry and promise to call
back later. This went on for some time. At one point, we decided that Meerwein
had to make a personal appearance, so we recruited a grad student from another
group (Lebanese, not German, but at least he had an accent) to appear at the office
door, announce his presence, click his heels, and storm off. We also contrived to
have a crate delivered to the office – it contained (if you’ll pardon the false Italian)
an Equilibrazioni Apparazioni. Inside was a wooden board – attached at one end
was a vial marked ketoni and at the other end another vial labeled alcoholi. In the
middle was a vial containing hydridio and, if we recall correctly, a light bulb that
would flash on/off when plugged into an outlet. (Meerwein will make another
appearance, shortly.)
One of the most dramatic and frightening events, even for a group of students
who were “accustomed” to explosions, was the violent detonation that Wolfgang

214
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Roth had one evening. He was generating gaseous diazomethane in one flask and
passing it into a second flask that contained Z-1,3,5-hexatriene and CuCl. The idea
was to determine at how low a temperature cis-1,2-divinylcyclopropane would
undergo Cope rearrangement to 1,4-cycloheptadiene. By cooling the reaction flask
to lower and lower temperatures, Roth found that the Cope was rapid at around 0
°C and even at -20 °C. It was when he cooled the flask to -55 °C that all hell broke
loose. Because the reaction flask had frosted over, he squirted a small stream of
acetone to clear a window. What he saw must have frightened him out of his wits
– the contents of the flask were bright yellow, an indication that diazomethane had
accumulated but had not decomposed. And so there was a fearsome explosion.
Fortunately, most of the force was up and down, but there was not a recognizable
piece of glassware afterwards, and the magnetic stirring motor was reduced to
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

half its height and was driven into the base of the ring stand. Doering has written,
in typical Doering style, a warning that below -50 °C, “... diazomethane is not
decomposed, accumulates in the reaction vessel, and eventually explodes with
singularly dangerous and destructive violence ... Accumulations of diazomethane
are then indicated by the turning yellow of the drops of condensate on the tip of the
reflux condenser. Experienced operators normally withdraw and continue flushing
the apparatus with nitrogen in the occasionally vain hope that the diazomethane
will thus be removed without event (13).”
Roth was hospitalized but, aside from a few pieces of glass in his face,
escaped pretty much unscathed. An article in The New Haven Register described
the incident. Alas, it contained errors: it identified Roth as a grad student and it
said that the explosion occurred in Room 203 (Doering’s office!). What it got
right was that the supervisor was Doering, that the compound was “diazomethane
(which is used as a methylating agent)” but then it concluded “Except for a test
tube, which was destroyed, there was no damage to the laboratory.” Some test
tube! Some “no damage”! The next day, a similar article appeared with this
additional information: “The experiment reportedly called for an explosion,
but the force of it was greater than anticipated.” This demanded the attention
of Meerwein Ponndorf, so “he” (Ron) called the reporter to set the record
straight: (1) Roth was a post-doc, not a grad student [true enough]; (2) he was
not working in Room 203 but in 150 [also true]; (3) he was not working with
diazomethane (a harmless, yellow solid [well, it is yellow]) but with isopropane
[this stemmed from the time when one of Marty Saunders’s students interrupted
his boss, who was conversing with some full professors in the hallway, and asked,
“Marty, how would you distinguish propane from isopropane by NMR?” and,
much to his delight, got an answer!]; (4) he was not working for Prof. William
Doering but for me, Prof. Meerwein Ponndorf; (6) he was not using a “test
tube” but rather an “equilibration apparatus”; and (7) he was in this country on a
Deuterium-Exchange Fellowship.
Even the staff at The New Haven Register had enough journalistic acumen to
be suspicious, and so the reporter called the phone in Sterling Lab 150 and asked
to speak to Meerwein Ponndorf. And who should have been making one of his
rare visits to the laboratory, but none other than Prof. William von E. Doering
who answered the phone and, without missing a beat but breaking out in a wide
grin, handed the phone to Ron and said, “I think it’s for you.” And so, German

215
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 accent intact, “Meerwein” proceeded to confirm all of the “corrections” that had
already been given. The next day, a third article appeared. Roth was still described
as a graduate student, but was now working in Room 105 (close enough) with
isopropane! The article concluded “An equilibration apparatus, which contained
the chemical, was destroyed.” Alas, neither Ponndorf’s name nor the Deuterium-
Exchange Fellowship made it into print.
Explosions were a way of life in Room 150, but generally – not quite always –
we were careful. In fact, a benign explosion was regarded as a good thing – it kept
us on our toes, and kept us putting those blast shields up. There were, however,
those whom nature seemed to favor who could get away with anything (not Roth,
apparently, nor Wolfgang Kirmse who had three diazomethane explosions in a
single day. One can only admire Kirmse’s tenacity, not to mention courage!). John
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

Munro, a postdoc from Scotland, distilled diazocyclopentadiene on the desktop in

multigram lots. Insane! – but nothing ever happened. Nonetheless, we tended
to absent ourselves from the lab when John was at the still. One of the iconic
stories of our group, from before our time, concerned the time Chuck DePuy had
an explosion that knocked him senseless; he awoke to find himself in a room filled
with dense white smoke. Doering, in re-telling this story, reported that DePuy
thought he was dead and that it took all of the boss’s powers to persuade him he was
not in heaven. On the other hand, Barbara Ferrier quipped that DePuy should have
known that he was not in heaven when he saw Doering there! Peter Leermakers,
an American post-doc, walked the halls of Sterling lab with 5-L flasks yellow with
gaseous diazomethane. Worse than insane! Doering remarked, after chewing us
all out: “I hope he was wearing a lead-lined jockstrap!”
Though most of Bill’s post-docs were German, there were a couple from Great
Britain and a notable one from the U.S., the afore-mentioned Peter Leermakers,
who had completed his Ph.D. with George Hammond at CalTech and was just a
year away from taking a faculty position at Wesleyan. He came to our lab with all
the confidence and bravado of one who is, probably, very insecure. His boastful
manner resulted in a not unexpected negative reaction from his new colleagues.
Peter was a dreadful experimentalist, but he must have had a guardian angel who
protected him from injury when he performed such terrifying acts as a closed-
system distillation or the diazomethane stroll. R. Srinivasan (from IBM) came to
Yale for a seminar and described some of his photochemical work on what would
come to be called, once Woodward and Hoffmann arrived on the scene, [2+2]
cycloadditions of alkenes and electrocyclic ring-openings of cyclobutenes. There
was one particular reaction that Doering challenged and refused to believe, but
Peter argued that it was probably correct. So Bill made a bet with him: “$50 says
that you can’t reproduce the result in less than a month.” Peter, of course, accepted
the challenge and, to everyone’s surprise but his own, won the bet, which Bill paid
off with a great flourish at an organic seminar. (Stu Staley remembers the wager
somewhat differently: “The bet was $10 at 10:1 odds – and Bill actually paid off
with a new $100 bill, after saying he was willing to accept 10 g of 95 or 97% pure
material in lieu of the specified 1 g of 99% pure material.”)
And we ourselves also did some awfully dumb things. Nighttimes would
sometimes find us (and others whom we shall not name in order to protect the
guilty) outside Sterling Lab trying to observe the blue flash that, legend had it,

216
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 occurs an instant before diazomethane explodes. Here, for those foolish enough to
repeat our experiment, are the details of one of our attempts. On a dark night, on
the lawn behind Sterling Lab, Mait was the principal instigator (others of us in the
group were his fearful unindicted co-conspirators) for what we were convinced
was a brilliantly conceived experiment. Imagine a large Erlenmeyer flask into
which was placed decalin over aqueous KOH. The idea, then, was to add solid
NMU to the flask, then cover the opening with a New York Mets trading card
onto which had been sprinkled KMnO4. Sulfuric acid was then poured onto the
permanganate. This combination of chemicals was supposed to catch fire after
about 20 seconds, giving the NMU/KOH reaction time to generate a vast quantity
of gaseous diazomethane. The result should have been an explosion that we could
all witness to see if there was, indeed, a blue flash. So the brave Mait added the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

NMU, set the card over the opening in the flask, poured acid onto it, and ran like
hell to hide behind the nearest tree with the rest of us. Yes, it was a “brilliantly
conceived experiment” except for one variable that we failed to factor in: a gust
of wind blew the card and its chemical cargo several feet from the flask – and
when it did burst into flames, it was too far from the flask to set off the yellow
gas. What to do? The smart, but cowardly, solution would have been to leave the
flask until morning – surely no other idiots would be crossing this dark expanse
of grass at night, and the diazo compound would have dissipated in a few hours.
But Mait had a sense of responsibility and would have none of it. Calling upon
his outstanding athletic skills he threw a rock at the flask ... and missed. So he
inched closer and threw another rock ... same result. And then before he could
throw the third rock, the contents of the flask exploded. He claims to still bear the
scars from this adventure. Oh, yes, although none of us observed the blue flash
Mait has written in Volume 1 of Carbenes, his monograph with Bob Moss, “the
failure of the Yale workers should not deter other investigators.”
As we related previously, much of our knowledge and understanding were
acquired in the weekly group seminars. It became our practice, when seminar
ended, to repair to a restaurant/bar called The Old Heidelberg (familiarly known
as “The OH”), where we proceeded to “get hydrolyzed.” In 2008, the Yale Alumni
Magazine published several tributes to the recently-deceased William F. Buckley.
One was by David Frum (famous right-wing zealot, although we usually hesitate
to apply such sweeping labels) in which he described an invitation that the Yale
Political Union had sent to Mr. B.: “For dinner, we students booked a restaurant
on Chapel Street called The Old Heidelberg (long-ago, and deservedly extinct).”
Alumni of the Doering seminars (Peter Gaspar, Marty Pomerantz, Stu Staley, Mait,
and Ron) reacted strongly to Frum’s ugly characterization, although in retrospect
The OH might not have been the most wholesome of places. Gaspar even recalls
that Bill attended one of these post-seminar sessions and paid for the evening’s
festivities by throwing a $50 bill on the table.
What we remember most vividly about The OH was a table populated by
graduate students in Spanish who were shouting “Viva, la guardia.” (In today’s
climate, they would have been arrested, labeled as enemy combatants, and sent
to Guantanamo.) We also recall that we conspired to have Stu Staley (he of
some 1000 sheep jokes and, also, affectionately nicknamed “mouse bladder”)
sit at the center position on the bench so that he’d have to exit under the table

217
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 for his numerous trips to the bathroom. Many of us have fond memories of two
lunchtime venues: the internationally famous Louis’ Lunch (original home of the
hamburger) and the totally ignored (but wonderful in its own way) Tony’s Apizza,
which served the universe’s best meatball sandwiches and which, according to
Mait, is now “dead as a smelt” (as Bill was wont to say).
The second-year chemistry grad students, by tradition, would host a beer party
for the department. As part of the festivities, they would put on a play about
the faculty: one (typically terribly boring and poorly written) by the physical
chemistry grad students, the other (brilliant!) by the organic chemists. In the spring
of 1961, we reserved a wonderful conference room on the top floor of Gibbs Lab.
(That some rowdies, almost certainly the physical chemistry grad students, threw
beer kegs off the roof guaranteed that future classes would never be allowed to use
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

that venue again.)

Mait and Ron wrote the organic chemists’ play, based on the horrific events
then unfolding in Belgian Congo. Set in the capital city of Sterlingville, the
principal characters, modeled after Doering and Saunders, were Patrice von
Eggers Lubullba and Martin Tschaunbe, roles played with distinction, pathos,
and libelous intent by Mait and Ron, respectively. We conspired to skewer
every member of the organic faculty in our not so subtle satire – hell, we even
killed off Lubullba in a people’s riot, led by Jungle Jim! Other characters were
Harold Cassidivubu, Richard Gebairdu, Gary W. Griffutu, Equus Wassenga, and
Count Lwalter Lwowski; these were the noms de scènes, respectively, for Jim
English, Harold Cassidy, Dick Baird, Gary Griffin, Harry Wasserman, and Walter
Lwowski). Most faculty, wisely, stayed away, but one who didn’t was visiting
professor Melvin Calvin (who had not yet won the Nobel Prize, although he was
sure that he deserved it). Tom Mattingly (from the Griffin research group) dressed
up in a bald wig and suspenders, to mimic Calvin’s appearance, and played the
role of “Malvin Kelvin.” Every word uttered by “Kelvin” in the play came from
the lecture series that Calvin was giving. Examples: “Having published most of
the excellent work in this field myself” and “It is a real tribute to your Department
Chairman that I am here today.” Saul Winstein, who had been a visiting professor
the previous semester, was portrayed as Saul Hammerstjein from the U.N. – this
character, also, spouted authentic Winsteinisms, such as attaching “-wise” to
words, as in “reaction mechanismwise.”
We mentioned the course that Winstein taught in Fall, 1960. It was toward the
end of the semester that Winstein discovered, to his great dismay, that a student
had mistakenly registered to take the course for credit. That student, Stu Staley, to
his even greater dismay, was required to meet with Winstein for an oral exam. As
it turned out, the exam was easy and Stu received an A, even though he claimed
to have answered none of the questions correctly.
The next semester, Calvin lectured twice a week in the late afternoon. It was
his habit to look up at the clock, note that it was close to the 5:30 ending time, and
say, “I have a few more minutes, so let me continue” (which he did for more than
a “few more minutes”). At his final lecture, we noticed (but Calvin didn’t) that the
wall clock had stopped at 5:25. Calvin glanced at it, noted the time, and said “I
have a few more minutes, so let me continue.” And when he looked up again and
saw the clock at 5:25, he repeated that sentence. We don’t remember how long

218
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 this went on, but he eventually recognized his error and let us go. A few weeks
later, Doering said, “Wasn’t it great what the assistant professors did to Calvin?”
Huh? We didn’t know, until that very moment, that some of the younger faculty
had conspired with a custodian to stop the clock at 5:25. (It was hinted that a bottle
of booze was the price that they had to pay.)
Speaking of Stu Staley, as we did earlier when discussing The Old Heidelberg,
he achieved great fame as a recipient of the “coveted” Emil Fischer award. The
award was given for notable chemical indiscretions (i.e., monumental screw-ups)
and was, for some time, the permanent property of various hacks in the Saunders
research group. Stu received the honor for having isolated a compound by gc,
taking its IR spectrum within salt plates(!), discovering that it was water, and
leaving the spectrum on his desk for all to see. The award, itself, consisted of
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

three layers of liquid in a bottle: CCl4 on the bottom, water in the middle, and
decalin on top. A blue compound (azulene?) was dissolved in the lowest layer
and a red compound in the top. When Ron decided that such an award was
needed during his first teaching position at Rice University, he named it after
Arthur Rudolph Hantzsch, an early 20th century German chemist, who was wont
to propose brilliant ideas but was unable to realize them experimentally. Sad to
say, one day a custodian saw the three-layered-liquid, was intrigued, shook the
bottle, and destroyed the beautiful red-white-blue pattern. Mait also stole the
idea. At Princeton, the “Lead Star” was awarded for many years for “conspicuous
acts contributing to the decline of chemistry.”
We return, again, to the weekly group seminars where we learned to write
concise abstracts of literature articles, to present cogent talks, and to hone
our skills for solving problems in mechanism and synthesis. As we began to
recognize patterns in these problems, Ron developed a facetious set of guidelines
that Mait nicknamed “Magid’s Rules.” Not only did he teach these rules to his
undergraduates, but he immortalized them by including them in his popular [“Not
popular enough!” according to Mait] textbook. First Rule: “All neutral products
are lactones” applied to road-map problems in which, typically, a carboxylic acid
(whose molecular formula, but not its structure, was given) was transformed into
a “neutral product” – more often than not, such a compound turned out to be a
lactone. Second Rule: “Always try the Michael first” referred to “fun-in-base”
mechanism questions that might involve sequences of aldol condensations and
Michael additions (the Robinson annulation being a typical example). Third Rule:
“When all else fails and desperation is setting in, look for the hydride shift” often
proved to be the key element in a mechanism problem that might have involved
an oxidation/reduction at some stage.
We will close with short descriptions of two other chapters in Doering’s life:
his Chinese connection and his work with the Council for a Livable World.
In the early 1980s, Doering was the American designer of the Chinese
Graduate Program (CGP) through which the best young Chinese graduate
students in chemistry were to be educated at a consortium of Western universities.
Special examinations were held, unusual language training applied, and, at least
at the start, precious hard currency exchanged. Eventually 240 extraordinary
young people came to this country and others. The program was a spectacular
success on this side, as anyone close to one or more of these students will attest.

219
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 However politics and fate intervened to make it a failure from the Chinese side.
Returning students were not met with open arms. Far from it: they were greeted
with suspicion and the anticipated junior-faculty positions were clearly not
about to materialize. Word got back to those still abroad who swiftly went into
postdoctoral parking orbits as a result. After Tiananmen in 1989, it became clear
that there were few who would go back.
Mait attended a reunion of CGP students in Los Angeles in 1994. About 50
attended and nearly 25 gave talks on what they were doing. They came from the
U.S. and Canada, from Hong Kong and other places. They were in academia and
in industry. Still others had moved quite far from chemistry and were in the world
of commerce. He remembers the day as “one of the most inspiring I have ever
had in chemistry. The energy and excitement were palpable ... It was a glorious
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

day. I think that Doering thought that the CGP program would contribute to the
rebirth of Chinese chemistry. That it did not do, or at least has not done yet. But
I am certain that when some future Ph.D. student in sociology writes her thesis
on this subject, she will conclude that Western chemistry profited immensely ...
Doering has much to be proud of in his professional life, certainly in the number
of scientific ‘children’ he has produced. This group of CGP ‘kids’ may prove to
be the most remarkable in the long run.”
In 1962, nuclear physicist Leo Szilard founded the Council for a Livable
World, a non-profit group dedicated to the goal of reducing the spread of nuclear
weapons. Doering chaired the Board of Directors from 1962 to 1973; and from
1973 to 1978 he served as president of the association. Mait and Ron’s tenure as
graduate students coincided, in part, with Doering’s increasing involvement with
this organization. When he was out of town, it seemed that it was as often for
Council meetings as to give a research seminar at a U.S. or European university.
The Council is still active in lobbying Congress, endorsing electoral candidates,
and holding symposia on how to achieve nuclear non-proliferation.
In the preceding pages, we’ve described what life was like as students in
Doering’s lab. We’ve related serious incidents as well many that produced only
amusement. What was the point of all this japery and foolishness – and of now
writing about it? We were a dedicated lot in Sterling Chemistry Laboratory’s Room
150 (and we knew it), but at the same time, we were having an enormous amount
of fun. We suspect that the atmosphere in contemporaneous research labs was
similar. Some of those pranks and silly behavior reflected the fact that we just
loved what we were doing. We wonder if the same is true today – it seems awfully
serious and intense out there.
We were also very goal-directed – we wanted to make those molecules and
see what they were like. But there was more to it. The atmosphere in the Doering
research lab was deeply intellectual, despite practical issues and all the tomfoolery.
That attitude came right from the top, as we have tried to make clear. In a sense we
were being instructed, by words and deeds, to emulate Rainer Maria Rilke, who
wrote (in “Letters to a Young Poet”) “... be patient towards all that is unsolved in
your heart, and try to love the questions themselves.”

220
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 References
1. Klärner, F.-G.; Jones, M., Jr.; Magid, R. M. Acc. Chem. Res. 2009, 42,
169–181.
2. Wu, G.-X.; Jones, M., Jr.; Doering, W. von E.; Knox, L. H. Tetrahedron 1997,
53, 9913–9920.
3. Woodward, R. B. In Aromaticity: An International Symposium; The
Chemical Society: London, 1967; p 246.
4. Doering, W. von E.; Roth, W. R. Tetrahedron 1962, 18, 67–74.
5. Doering, W. von E.; Birladeanu, L.; Sarma, K.; Teles, J. H.; Klärner, F.-G.;
Gehrke, J.-S. J. Am. Chem. Soc. 1994, 116, 4289–4297.
6. Doering, W. von E.; Denny, D. B. J. Am. Chem. Soc. 1955, 77, 4619–4622.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch009

7. Roth, W. R.; Lennartz, H.-W.; Doering, W. von E.; Birladeanu, L.; Guyton, C.
A.; Kitagawa, T. A J. Am. Chem. Soc. 1990, 112, 1722–1732.
8. Doering, W. von E.; Knox, L. H. J. Am. Chem. Soc. 1956, 78, 4947–4950.
9. Doering, W. von E.; Ekmanis, J. L.; Belfield, K. D.; Klärner, F.-G.;
Krawczyk, B. J. Am. Chem. Soc. 2001, 123, 5532–5541.
10. Nickon, A.; Silversmith, E. F. Organic Chemistry: The Name Game;
Pergamon: New York, 1972; p 131.
11. Doering, W. von E.; Ferrier, B. M.; Fossel, E. T.; Hartenstein, J. H.;
Jones, M., Jr.; Klumpp, G.; Rubin, R. M.; Saunders, M. Tetrahedron 1967,
23, 3943–3963.
12. Jones, M., Jr.; Reich, S. D.; Scott, L. T. J. Am. Chem. Soc. 1970, 92,
3118–3126.
13. Doering, W. von E.; Roth, W. R. Tetrahedron 1963, 19, 715–737.

221
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Chapter 10

60 Years of Research on Free Radical Physical

Organic Chemistry
K. U. Ingold*
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

National Research Council, Ottawa, ON K1A 0R6, Canada

*E-mail: [email protected]

In this Chapter the reader is conducted on a gentle cruise through

the green and pleasant land of mechanistic radical chemistry.
My love of chemistry arrived with a bang (potassium metal
+ water) and I was already enamored by “how fast?”. These
two loves combined at the NRC as I helped to convert radical
chemistry from a tarry, black art, into clean, practical, useful,
predictable and comprehensible chemistry.

My father, Christopher Kelk Ingold, Professor and Head of the Chemistry

Department at University College London (UCL, 1930-61), was the first recipient
of the James Flack Norris Award for Physical Organic Chemistry (1965). This was
a great honor and one that he very much appreciated. (To both of us, this was a
very personal example of Americans’ generosity of spirit, particularly since it was
a very distinguished American, Louis Hammett, who coined the words “Physical
Organic Chemistry”. Hammett’s seminal role in this subject was honored by his
receipt of the second J. F. Norris Award in 1966).
I went to the ACS meeting in Atlantic City to see the Award presented and
then drove my father, and two chemists from UC Santa Barbara, Clifford Bunton
(ex-UCL) and Tom Bruice (Norris Award winner in 1996), back to my home in
Ottawa. I live on the Rideau River which is at the bottom of my garden (in Summer,
in Spring the garden is at the bottom of the river!) and had a ski boat (woefully
underpowered relative to later boats). My father surprised all of us by expressing
a wish to water ski though he’d never done it before. With many misgivings I
agreed and even drove the boat. To our joint surprise, my father got up the first
time he tried, but then he had always kept fit. He used to do, and encouraged me

Published 2015 by the American Chemical Society

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 to do, a lot of walking, rock climbing (Wales), and mountaineering (Alps), sports
he continued into his late 50s (Figures 1-4).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

Figure 1. Getting ready for the great water-ski caper: (right to left) John Ingold
(age 5), Bruice, Bunton, CKI, and KUI.

Figure 2. CKI skiing for first time at age 72

224
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

Figure 3. KUI skiing at 85. One of the advantages of living with a river!

Figure 4. Other than the first two pictures in this Chapter, this is the only known
snap of myself and my father together. We are sitting with Lionel Jones, husband
of my sister, Dilys, in their garden in Sussex.

225
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 C. K. Ingold was chosen to be the 1st recipient of the Norris Award largely for
his pioneering mechanistic and kinetic studies of heterolytic chemical reactions.
The statement: “I should tell you that in this Department, homolysis, even between
consenting adults, is grounds for instant dismissal.” was not made by CKI but by
Peter de la Mare as comment on Alwyn Davies’ wish to start work on peroxide
homolysis (1).
That I would also become a research chemist, like my father (and mother) was
unsurprising, as was my very conscious decision to steer well clear of heterolytic
chemistry. Emigration to Canada (with a freshly minted D.Phil from Oxford) in
1951 at the age of 22 may have been one of my wiser decisions since my father
and I had, perhaps, too much in common to get on as well as we actually did if
we’d lived closer to one another, cf., pictures of the two of us through the years
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

(Figures 5 and 6).

I believe my love of chemistry was sparked by an event in 1941-2. World
War II disrupted everyone’s life. My father and mother were evacuated with
half of UCL’s Chemistry Department to Aberystwyth (Aber), Wales. Eventually,
(summer 1940) they were joined by me and my two sisters. I had attended at least
3 schools after evacuation and before my arrival in Aber in June, 1940. While
living in Aber I attended 3 more schools (and went to yet another upon our return
to London in 1944) before my entry as a chemistry undergraduate into UCL. My
frequent school changes were not due to (sometimes well-deserved) expulsion,
they were just a consequence of Hitler’s disruptions. In Aber there was little to do
on a weekend. In a forlorn attempt to keep me out of trouble, my father arranged
for the Department’s glass blower to teach me some of his skills on Saturday
mornings (which proved very useful later). However, time hung heavy (until I
discovered the joys of rock climbing). One Sunday my father took me into the
labs and, after finishing whatever work had dragged him there, he decided to show
me some interesting chemistry. Interesting was an understatement! He wanted to
show me what happened when a small piece of sodium was dropped into water.
Fortunately for me, he didn’t know where the sodium was kept. However, he
found the potassium and, being a bit rattled by his failure to find the sodium in
“his” new-to-him lab, cut off several grams of potassium and threw it straight into
a sink prefilled with water. Spectacular! There was an enormous bang, far louder
than any I later heard from V2 rockets landing in London. The sink was instantly
dehydrated, small puddles by the hundred appeared on the floor, the bench-tops,
everywhere, each one with little bits of purple-burning potassium speeding around
on its surface. We were not wearing safety glasses, but survived unharmed. Even
though we had to clean the entire lab (to keep all others in the Department ignorant
of the Professor’s sins) I was instantly and totally hooked on chemistry.
Hooked on chemistry, yes, but I’d been hooked on kinetics almost a decade
earlier. Well before I was a teenager, I overheard some adults describing a lady
who lived near us as a “fast woman”. I’d seen the lady in question and she didn’t
look fast to me, though I did consider the possibility that high-heels gave speed on
the running track. Since this seemed a mite unreasonable, I decided that when I
grew up I would adopt an experimental approach and find out what made a woman
fast. The results of these experiments were so rewarding that I’ve remained an

226
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 experimentalist who measures how fast things go, though today it is only chemical
reactions, unfortunately.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

Figure 5. CKI and KUI as ca. 12 year olds.

Figure 6. CKI and KUI in their 70s.

227
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Four years (and 2 Post-Docs) after I arrived in Canada, I abandoned gas
phase chemistry to accept a position in the Division of Applied Chemistry at the
National Research Council in Ottawa (despite snide comments from many of my
‘pure’ chemistry friends). I wanted to work in the liquid phase because there is
so much more interesting chemistry there than there is in the gas phase (witness,
the entire field of biochemistry). I was hired to investigate the mechanism of
oxidative degradation of automobile engine oils and to try to find ways to retard
such degradation. A few years of glacially slow progress as a “real” applied
chemist working on “real” engine oils convinced me to go back to basics. To this
end, during 1960 I wrote my first Chemical Review, Inhibition of the Autoxidation
of Organic Substrates in the Liquid Phase, in order to acquaint myself fully with
the current state of knowledge (2). My career is now “book-ended” by my second
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

Chemical Review (2014): Advances in Radical-Trapping Antioxidant Chemistry

in the 21st Century. A Kinetic and Mechanistic Perspective (3). Although
antioxidants remain my “first love”, the present Norris Chapter will show that my
research has extended into a few other areas of physical organic chemistry.

Mechanism of Inhibition of Hydrocarbon Oxidation by Phenols

In the early 1960s I started work on this topic. Literature results were very
confusing. Controlled oxidation of a hydrocarbon, RH, is achieved by thermal
decomposition of an azo-initiator, iN=Ni, and occurs by a two-step chain reaction:

Such oxidations are retarded by phenols, ArOH, but not by anisole, PhOMe.
This implied that the phenolic OH group was required for inhibition. The obvious
inhibition step is an H-atom transfer by which a chain carrying peroxyl radical is
converted into a non-chain carrying phenoxyl radical:

However, if this reaction was responsible for the inhibiting effect of ArOH,
it would be expected to show a deuterium kinetic isotope effect, DKIE, but
none had been found. The DKIE experiments involved measuring the rates
of azo-initiated oxygen uptake by some readily oxidizable hydrocarbon (e. g.,
cumene or Tetralin) under an atmosphere of O2. To the hydrocarbon was added,
in mM concentrations, a phenol, ArOH, and in a matched experiment, the same
concentration of the corresponding pre-formed O-deuterated phenol, ArOD. In
such experiments the reduction in the rate of oxygen uptake induced by the ArOH
228
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 and ArOD were identical, indicating that there was no DKIE for inhibition.
This negative result was confirmed by many authors (myself included) and it
led to some extraordinary mechanistic suggestions in the literature. During the
writing of my first Chemical Review, I realized one evening that these inhibition
experiments actually had no chance whatever of uncovering a DKIE (if there
really was one). Rates of oxidation were followed in sealed systems under pure
O2 by measuring the consumption of O2. What I belatedly realized was that to
measure such a rate required the absorption of an appreciable amount of O2. In the
equipment of the day, this measurement required the uptake into the hydrocarbon
of a higher concentration of O2 than the concentration of ArOH or ArOD that
had been added. This meant that in order to measure the rate of O2 uptake it
was necessary to make a higher concentration of the substrate’s hydroperoxide,
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

ROOH, than the concentration of added ArOD. I had recently completed some
IR studies on phenols, incidentally demonstrating (contrary to then accepted
wisdom) that the O-H bond in 2,6-di-tert-butyl phenols lay in the aromatic
ring plane and that 2-tert-butylphenols existed as an equilibrium mixture of syn
and anti structures. This work had taught me that OH-containing compounds
underwent proton exchange with ArOD extremely rapidly. I concluded, therefore,
that in all the experiments searching for a DKIE in reaction 5, essentially all
ArOD had been converted to ArOH before the rate could be properly measured.

These thoughts led me to suggest to my first Post-Doc, J. A. (Tony) Howard,

that he add a drop of D2O to a hydrocarbon oxidation retarded by 2,6-di-tert-
butyl-4-methylphenol (BHT, the major phenolic antioxidant used commercially)
so as to make sure that the BHT remained fully deuterated during the reaction. I
also suggested using styrene as the oxidizable substrate because it was known to
yield a polyperoxide (with no exchangeable H-atoms) rather than a hydroperoxide.
Tony’s results (4) were a game changer in antioxidant research. He found a DKIE
of 10.6 at 65 °C, thus demonstrating that phenols are antioxidants because they
donate their phenolic H-atom to an attacking peroxyl radical.

The length of time the oxidation is retarded by the phenol, known as the
induction period, τ, is given by:

Work on phenolic antioxidants was continued by an examination of the

effect of ring substituents on the abilities of phenols to retard the autoxidation of
styrene. Electron donating (ED) para and meta substituents improved antioxidant
activity while electron withdrawing (EW) substituents reduced activity. In these
experiments, the rates we were measuring only allowed us to determine the rate
constant ratio, k2/k5. A Hammett plot of log(k2/k5) against Brown and Okamoto’s
σ+ values for the substituents yielded a straight line. This was completely
unexpected because σ+ substituent constants had been derived from the relative
229
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 rates of a heterolytic reaction, the ionization (solvolyses) of m- and p-substituted
cumyl chlorides in aqueous acetone.

The reason for the linear free energy relationship between log(k2/k5) and σ+
was only uncovered many years later. It arises because the XC6H4O-H bond
dissociation enthalpies (BDEs) also correlate with σ+. This last correlation was
recognized (by an outstanding Post-Doc, Gino DiLabio) to arise because the O•
moiety is, like the +CMe2 group, an extremely powerful EW substituent. As a
consequence of the EW effect of O•, ED substituents weaken and EW substituents
strengthen the ArO-H BDE, making the antioxidant reaction 5 thermodynamically
more and less favorable and, hence, faster and slower, respectively. In fact, the O-
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

H BDEs in substituted phenols and the N-H BDEs in substituted aromatic amines
trend with σ+ due to the electron-poor nature of O• and N• as discussed in a 2004
Account: Bond Strengths of Toluenes, Anilines, and Phenols. To Hammett or Not
(5). The mechanism of retardation of hydrocarbon oxidation by diarylamines,
Ar2NH, was also contentious until we demonstrated an NH/ND antioxidant DKIE.
This class of industrially important antioxidants therefore owed their activity to
donation of their amino H-atom to ROO• radicals. Thus both phenols and aromatic
amines are Radical Trapping Antioxidants, RTAs.
If further progress in antioxidant chemistry was to be made it was necessary
to know the magnitude of k5 rather than k2/k5 rate constant ratios. After all, my
ultimate goal was to see whether a phenolic antioxidant could be designed that
reacted with the first peroxyl radical it encountered! This Holy Grail was achieved
several decades later by Derek Pratt, who had worked with me in 1998-99 as a
NRC summer student, and there caught “grail fever”, see reference (3) for the
full story. Such an antioxidant could not be bettered since reaction 5 would be
diffusion-controlled (with k5 ~ 109 M-1s-1). It had already been shown that absolute
rate constants could be determined for a number of free radical polymerization
chain reactions. The main focus of research in my lab therefore shifted to the
measurement of absolute rate constants for radical reactions, a subject that looked
boundless and a subject to which I have (mainly) stayed faithful.

Absolute Rate Constants for Hydrocarbon Oxidation

Some polymer chemists had been using photolysis to initiate their reactions
and wanted to know how the rate of polymerization varied with the light intensity.
Since no instrument was available to measure light intensities they inserted a metal
disc, from which a 90° sector had been removed, between the light source and the
reaction vessel. Rotation of this disc would block the light for 75% of the time
and so the rate was expected to be 25% of the rate with full illumination. This was
the case when the disc was rotated slowly but as the rotation rate was increased
the rate of polymerization also increased up to a limit of 50% of the full-light rate.
This odd behavior arose because chain termination in these polymerizations was a
second-order process, Pn• + Pm• → Pn+m. Provided the rate of chain initiation, Ri,
was known, the rate constant for chain termination could be calculated from the
230
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 duration of the flash at rotation speeds where the rate was >25% but <50% of the
full-light rate.
The transfer of this rotating sector technology to hydrocarbon oxidation was
surprisingly easy. The experimental temperature was decreased as far as possible
(30 °C) to minimize any thermal formation of radicals. UV photolysis of an
azo compound provided a steady rate of initiation, Ri, and the magnitude of Ri
was determined from the length of the induction period induced by addition of
a known concentration of BHT. Product studies by George Hammond (Norris
Award, 1968) had shown that each molecule of BHT traps two ROO• radicals
(hence the 2k5 term in equation II). With Ri known, the measured uninhibited
rate yields the rate constant ratio, k2/(2k4)½, the rotating sector yields 2k4 and
hence k2, and finally the inhibited rate yields k2/k5 and hence k5. This may seem
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

a long road, but in the 1960s it was the only road available for obtaining the
desired absolute rate constants. Measurements on some 40 alkyl-aromatic and
olefin hydrocarbons (6) laid the foundation for all subsequent quantitative work
on autoxidation and inhibition by radical trapping antioxidants. It was found that
k2 was highly dependent on the RH structure, mainly for enthalpic reasons (R-H
BDEs), for example, k2 / M-1s-1 = 0.18 for isopropylbenzene (cumene), 60 for
methyl linoleate (MeLin), and 1500 for 1,4-cyclohexadiene (CHD) for which the
chain carrying peroxyl is hydroperoxyl, HOO•. {Note that work from Alwyn
Davies’ lab has shown that there is an analogous propagation step in the oxidation
of organoboranes, i.e., ROO• + R3B → ROOBR2 + R•, and that this step is many
orders of magnitude faster than any ROO• + RH reaction.} The values of 2k4 varied
just as dramatically, with values for tertiary alkylperoxyls being very much lower
than for secondary and primary alkylperoxyls, e.g., 10-6 x 2k4 / M-1s-1 = 0.015
for cumene, 9 for MeLin, and 1300 for CHD. For this uninhibited termination
reaction, tertiary and secondary alkylperoxyls almost certainly react by different
mechanisms but, despite the efforts of many physical organic chemists (notably,
Paul Bartlett, Norris Award, 1969, and Glen Russell, Norris Award, 1983), only
the mechanism for tert-alkylperoxyls can be said to be fully understood. Chain
propagation for styrene autoxidation involves the addition:

for which k8 was determined to be 41 M-1s-1 at 30 °C (6). This allowed the values
of k5 for BHT and diphenylamine to be calculated. Both of these radical trapping
antioxidants had k5 ~ 2 x 104 M-1s-1, some 4 to 5 orders of magnitude below the
diffusion limit, leaving much room for later (huge) improvements (3).

Absolute Rate Constants for Reduction of Alkyl Halides by

Tin Hydrides
In the mid-60s I had the good fortune to hear Cheves Walling (Norris Award,
1971) give a wonderful lecture on the radical chain reduction of organo-halides
by tri-n-butyltin hydride, reactions 9 – 14. This was brand new chemistry to
me. Walling described the reduction of 5-hexenyl bromide because he was
investigating cyclization of the 5-hexenyl radical (chemistry later studied in detail
231
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 by Athel Beckwith). The main products were 1-hexene and methylcyclopentane
which are formed in a ratio dependent on [Bu3SnH] and on the rate constant ratio,
k15/k16. I decided, while Walling was speaking, to investigate the kinetics and
measure the absolute rate constants for these reactions. Walling was supportive
of my decision and a friendship developed that survived his becoming Editor of
J. Am. Chem. Soc. and quite a few arguments over the suitability for JACS of
some of my later submissions! Indeed friendly relations were maintained with
subsequent JACS Editors, all the way to the present day (Peter Stang, Norris
Award, 1998).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

An easy decision to make, but how to monitor these, necessarily

photo-initiated, reactions? Another excellent Post-Doc, Dave Carlsson, took
up the challenge. We decided to measure the heat evolved in these exothermic
reactions using a Pt / Pt-10%Rh thermocouple mounted at the center of a Pyrex
cell that was, in its first incarnation, about 1.5 cm long and 3 cm in diameter.
The design of the system evolved, the cell shrunk and the problem of direct
photo-heating of the thermocouple was evaded by using twin cells side-by-side
in the light beam with the full slate of reactants in cyclohexane present in
only one of these cells, the other contained the reference thermocouple and
cyclohexane. Conditions in the reactant cell remained adiabatic for ca 10-15 s,
i.e., the temperature rise was linear, and the rate of temperature increase could
be converted to a rate of reaction after some separate (essentially classical)
calorimetric experiments. After each rate measurement, the system was allowed
to return to thermal equilibrium (10-15 min).
The kinetics were fascinating. For RCl the rates were proportional to
[chloride] and independent of [Bu3SnH], but for RBr and RI the rates were
independent of [halide] and proportional to [Bu3SnH] (7). Thus, with chlorides,
232
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 chlorine atom abstraction was rate controlling and termination was by the
bimolecular self-reaction of the tin-centered radical, reactions 11 and 14,
respectively. However with bromides and iodides, hydrogen atom abstraction was
rate controlling and termination was by the bimolecular self-reaction of the alkyl
radicals, reactions 10 and 12, respectively. Simple competition experiments with
chlorides gave rate constants for bromine and iodine abstraction by Bu3Sn•. As
we wrote in 1968 (7): The evaluation of the rate constants for both propagation
reactions and two out of the three possible termination reactions means that these
processes are now among the best understood two-step chain reactions.
The advantage of synthesis via a chain reaction with a long chain length is
that by-products (e.g., from initiation and termination) become no more than very
minor impurities. Although tin hydride reductions are of no interest industrially
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

(because of the toxicity of tin), these reactions are of immense value to academics
exploring free radical-based syntheses. With all these rate constants now available
to synthetic chemists some, most notably Dennis Curran, made use of them to plan
complex syntheses in which the reagent concentrations and other conditions could
be pre-selected for success.
From the physical organic chemist’s viewpoint, there were two especially
interesting results in this work (7).

[i] Since the rate constant for cyclization of the 5-hexenyl radical was
now available, this rearrangement could be employed to time, i.e.,
to determine the rate constants, for other reactions involving primary
alkyl radicals using kinetic competition methods and product analyses.
This was the birth of the Free Radical Clocks concept, as described in
an Account (8) with this title, and later exploited with great elegance
by Marty Newcomb (Norris Award, 2000). This was another step on
our way to a full quantitative understanding of free radical chemistry.
However, it is the peroxyl radical clocks, developed in the lab of Ned
Porter (Norris Award, 2013), that have proved the most helpful to the
physical organic chemistry of highly reactive RTAs (3). Porter had
demonstrated that the complex mixtures of isomeric hydroperoxides
formed during the autoxidation of polyunsaturated fatty acids, arose
because O2 addition to pentadienyl radicals is reversible, see Scheme 1
(reaction 18). The initial pentadienyl radical adds O2 to give a cis-trans
peroxyl radical which may be trapped by an H-atom donor to give a
cis-trans hydroperoxide or may undergo a bond rotation followed by a
β-scission with loss of O2 to form a new trans-trans pentadienyl radical.
Addition of O2 to this new pentadienyl yields the thermodynamically
favored trans-trans peroxyl and hence the corresponding trans-trans
hydroperoxide. The rate constants for β-scission were determined by
competition with H-atom donation from 1,4-cyclohexadiene and hence
can be traced back to much earlier measurements of the absolute rate
constants for 1,4-cyclohexadiene autoxidation (6). A variety of peroxyl
radical clocks have been calibrated and have proved invaluable for
determining rate constants for the near diffusion-controlled radical
trapping antioxidants developed in recent years (3).

233
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Scheme 1

[ii] The rate constants for chain termination via bimolecular self-reactions of
the radicals: n-hexyl, c-hexyl, t-butyl, Bu3Sn• (and various other R3Sn•),
were all ca. 2 x 109 M-1s-1. That these reactions must all be diffusion-
controlled was demonstrated by showing that 2k12 for one of the radicals,
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

t-Bu•, was larger in the less viscous solvent n-pentane, and smaller in
the more viscous n-tridecane, than in cyclohexane. {Amusingly, some
gas-phase chemists claimed years later that the t-Bu• + t-Bu• rate constant
was only ca. 106 M-1s-1, although by that time we had confirmed the
109 M-1s-1 value using EPR spectroscopy, a technique later brought to
perfection by Hanns Fischer. Complex mixtures of reagents were used
to generate t-Bu• in our EP(aramagnetic)R (e.g., t-BuOOBu-t, Me3SiH,
t-BuCl, in c-C3H6) which led to the following (Paris bar) exchange with
the adamant 106 M-1s-1 proposer, as recorded in a footnote to a 1974 JACS
paper: Sceptic, “Getting reliable radical decay rate constants out of that
mixture mustbe like trying to find a needle in compost heap.” Author, “It
would be if we did not use a magnet.”}

The tin hydride work induced me to purchase an Electron Paramagnetic

Resonance (EPR/ESR) spectrometer in the late 1960’s. I expected this would
become an important instrument for a lab devoted to measuring the kinetics and
absolute rate constants for radical reactions, particularly since Jay Kochi (Norris
Award, 1981) had just developed several novel ways to photo-generate radicals in
an EPR. My expectations were fully realized. The new EPR turned out to be ideal
for studying the kinetics of radical decay whether this occurred by a bimolecular
process (combination and/or disproportionation, cf., reactions 4, 12, and 14), or
by some unimolecular process such as isomerization into a different radical (of
the same molecular weight) by an intramolecular rearrangement, e. g., reaction
16, or by a bond-scission to form an unsaturated molecule and a smaller radical.
Measurements of the decay kinetics of a very wide structural range of radicals
having the unpaired electron (formally) centered on C, O, N, Si, etc. led to the
crystallization of some ideas I’d long been mulling over.

Persistent Carbon-Centered Radicals

Is the title of an Account (9) in which these ideas were presented. The
problem addressed was centered around the meaning of “stable radical”. These
words were being used throughout the literature to refer two very different radical
properties. Consider the benzyl and ethyl radicals, the former is obviously “more
234
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 stable” than the latter, but both radicals decay by bimolecular self-reactions
with the same rate constant, viz. 25% (the doublet-doublet spin correction)
of the diffusion-controlled limit. Indeed, nearly a decade of careful kinetic
work by several Post-Docs, notably David Griller, demonstrated that sterically
non-congested and EPR-visible radicals of all types (note that RO• and RS• have
degenerate SOMOs and are invisible by EPR.) underwent their self-reactions with
this rate constant provided they could combine to give thermodynamically stable
dimers or disproportionate, e.g., 2 R2CHX• → R2C=X + R2CHXH.
The need to separate thermodynamic stabilities from kinetic stabilities was
met in this Account (9). Thermodynamic stabilities are normally defined by R-H
BDEs. Thus, for C-centered radicals, thermodynamic stabilities are normally
referenced to the [CH3-H] BDE or, with a nod to obvious structural effects, to
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

the [CH3CH2-H] BDE for primary alkyls, the [(CH3)2CH-H] BDE for secondary
alkyls, etc. The Account dealt with thermodynamic stability in the conventional
manner (9). However, to describe kinetic stabilities, or lack thereof, this Account
formally introduced the words Transient to describe radicals that decayed at
(1/4) of the diffusion-controlled rate, and Persistent to describe radicals that
decayed more slowly. Thus, benzyl, PhCH2•, is a stabilized transient radical, but
the sterically crowded benzyl, PhC(Me3C)2•, is a stabilized persistent radical.
Similarly, Ph• is a destabilized transient radical, but the sterically encumbered
phenyl, 2,4,6-tri-t-butylphenyl, is a destabilized persistent radical. The words
stable radical were restricted to radicals that could be “put in a bottle” such
as 2,2-diphenyl-1-picrylhydrazyl (dpph•), di-t-alkyl nitroxides (e.g., TEMPO),
and di-1-adamantylketiminoxyl. Di-sec-alkyl nitroxides and all alkylperoxyls
do undergo bimolecular self-reactions but with rate constants well below the
diffusion limit, they are therefore stabilized, but mildly persistent radicals.
Interestingly, even the secondary alkyl radical, (t-Bu)2CH•, is persistent, but it
slowly dimerizes to form (t-Bu)2CHCH(t-Bu)2. This compound was the most
sterically crowded molecule made to that time. A fact that reminded me of a
fictional molecule, thiotimoline, that was so crowded that it pushed a part of
itself into the future and another part into the past! In 1977, I celebrated steric
crowding in JACS (thanks to Editor Walling, who ruled in my favor over a
mirthless reviewer) with two footnotes:
21Despite severe crowding ((t-Bu)2CHCH(t-Bu)2) does not appear to show

the unusual properties reported22 for the “super” sterically hindered molecule,
thiotimoline.
22I. Asimov, J. Astound. Sci. Fict. 50, 120, (1948).

Today, the persistent / transient nomenclature is widely employed in both

radical and non-radical chemistry (but thiotimoline has still to be synthesized).

Electron Paramagnetic Resonance and the Art of Physical

Organic Chemistry
Is the title of an Account (10) which summarizes highlights from a decade
or so of lovely work by several Post-Docs, notably David Griller. I acquired my
first EPR spectrometer in the late 1960’s because it was obviously an important
235
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 instrument to have in a lab devoted to measuring the kinetics and rate constants
for radical reactions. A rich research vein was soon tapped, some 42 papers being
published between 1971 and 1983 in J. Am. Chem. Soc. under the title: Kinetic
Applications of Electron Paramagnetic Resonance Spectroscopy and even more
EPR-based research appeared that was not a part of this series.

Quantum Mechanical Tunneling

A kinetic EPR study showed that the persistent 2,4,6-tri-tert-butylphenyl

Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

radical (generated by UV photolysis from the corresponding bromide and a

source of tin- or silicon-centered radicals) decayed with first-order kinetics.
Decay was shown to involve a 5-center intramolecular H-atom transfer from one
of the ortho tert-butyl groups to the vacant position on the ring so as to form
a transient primary alkyl (neophyl) radical, reaction 19. Both of these radicals
could be observed and, on shuttering the UV light, the phenyl decayed with
first-order kinetics and the neophyl with second order kinetics. The temperature
dependence for phenyl radical decay (183–243 K) showed an unexpectedly small
Arrhenius pre-exponential factor (A19 = 2 x 105 s-1). I commented to my two
coworkers (Griller and Ross Barclay) that this H-atom transfer must involve
Quantum Mechanical Tunneling (QMT)! Both laughed heartily and said some
less than complimentary things about my intelligence. Quite understandable, the
year was 1974 and R. P. Bell had spent the past couple of decades looking for an
unequivocal example of QMT in proton transfer reactions with little or no success.
The subject of QMT was virtually “off limits” for serious chemists. However, in
my mid-teens I had read the physicist, George Gamow’s wonderful “Mr. Tomkins”
books. I had been particularly taken by the short story “Quantum Billiards” and
the accompanying cartoons, particularly that of Mr. Tomkins’ car tunneling out of
his closed garage. In this story one reads: “It is impossible to hold anything inside
an enclosure provided there is enough energy for running away after crossing the
wall. Sooner or later the object will just ‘leak through’ and go away.” A bit of
a stand-off ensued but eventually I got my way and my two ‘doubting Thomas’
colleagues synthesized a bromide with three per-deuterio-tert-butyl groups, the
phenyl radical from which also decayed with first order kinetics but much more
slowly than the non-deuterated phenyl, k19H/k19D = 50 at 243 K, a result that
quieted my two doubters.

Work on reaction 19 (and on its deuterated analogue) was later extended. QMT
was firmly established by demonstrating that at temperatures below ca. 100 K, k19H
236
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 (and k19D) became independent of the temperature, with k19H/k19D ~ 100,000! The
decay of phenyl radicals having related rigid and crowded structures also occurred
by intramolecular 1,5-H-atom isomerizations in which QMT is dominant.
A number of other H-atom transfers were also demonstrated to occur
exclusively by QMT at low temperatures. In all cases, these reactions involved
rigid systems. Some of these reactions were exothermic, e.g., reaction 20 at
very low temperatures in rigid methanol glasses where k20H becoming constant at
temperatures below ~45 K and exhibits a very large DKIE. At temperatures above
45 K but below the glass’ softening temperature, this reaction was particularly
interesting because CH3• decay followed a “stretched exponential”, i.e., each
half-life was longer than the preceding half-life. The reason for this behavior
is that the CH3• radicals occupy an array of sites within the methanol glass,
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

each with its characteristic first-order rate constant. Rates of QMT are highly
dependent on the “jump” distance with the smaller jumps occurring far more
rapidly than the longer jumps. A detailed treatment of the kinetic data was utilized
to derive jump distances (no variables and no assumptions) and gave a structure
for the rigid methanol glass that was fully consistent with that deduced by other
(non-kinetic) methods.

QMT was also explored in a number of thermoneutral “reactions”, e.g.,

inversion of the oxyranyl radical, reaction 21, for which the occurrence of
QMT had been denied. However, the rate constant for inversion, k21H, became
temperature independent below 140 K and again there was a huge DKIE.

Serious attempts to obtain evidence by EPR for QMT in exothermic

intermolecular H-atom transfers in solution were made in my lab, but all
were inconclusive (low A-factors, but ‘normal’ DKIEs and no rate constants
independent of temperature). All of this QMT work has been summarized (11).
My conclusion is that while QMT must always play a role in H-atom (and proton)
transfers, it only becomes dominant, and hence obvious, in reactions where
the H-atom source and sink are physically prevented from achieving a close
(optimum) separation. The large DKIEs reported by J. P. Klinman and others for
certain enzyme-catalyzed reactions are consistent with this idea. Shamefully, I
had overlooked a lovely example of QMT in what is, formally, an intermolecular
H-atom transfer. This example had been right under my nose for decades! I’m
referring to a recent report from the Porter group that reaction 2, with RH =
tetralin, has a primary DKIE at 65 °C that is at least twice the classical maximum
(~6-7). These results indicate, to me, that this apparently ‘simple’ H-atom transfer
involves QMT in a pre-reaction ROO•/RH complex.
237
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The Return to My First Love, Peroxyl Radical Trapping
Antioxidants

After a separation of many years, I was enticed back to antioxidant research

in 1980 by the ending of a long running controversy in the biochemical literature,
a controversy that, in my opinion, the “wrong” side had won, pretty much by
default. The question at issue had been: Does Vitamin E owe its bioactivity
to its ability to trap radicals? I “knew” the answer should have been “Yes”
simply because “Vitamin E” is any one of, or is any mixture of, four phenols,
the tocopherols (see Chart 1). Since Howard and I had shown that every one
of the 50-odd phenols we had examined were RTAs, Vitamin E just had to
be one. However, numerous reports in bio-oriented journals claimed that the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

tocopherols were not RTAs and some quite extraordinary explanations for this
dichotomy were published. Furthermore, there were also reports that γ-tocopherol
was a better antioxidant than α-tocopherol. Since it was well established that
α-tocopherol had a stronger Vitamin E activity than γ-tocopherol, Vitamin E
could not, therefore, owe its activity to radical trapping (or so the argument went).
When I finally forced myself to read these papers I was amazed to discover that
the oxidizable substrate was most commonly the homogenized organ from a rat,
generally its liver. Apparently, this made the experiments biologically relevant,
though I know of no animal that can survive if its liver is homogenized! In my
view, both then and today, the oxidation of the lipids in a homogenized organ
would be a metal ion-catalyzed (mainly FeII/III) oxidation, with an unknown,
uncontrolled, high, and highly variable, Ri. The stronger antioxidant action of
γ-tocopherol relative to α-tocopherol in these systems was, in my view, irrelevant
to the question of whether or not Vitamin E owed its activity to peroxyl radical
trapping. γ-Tocopherol would obviously be oxidized to an ortho-quinone and, of
course, ortho-quinones are metal ion chelators. Such an oxidation product would
not be formed from α-tocopherol since it lacks a non-methylated ortho-position
on its aromatic ring. Thus, γ-tocopherol would not only be able to trap peroxyl
radicals, its (main) oxidation product would chelate some of the catalytic metal
ions present in the liver homogenate and so reduce Ri. No wonder γ-tocopherol
was a “better” antioxidant than α-tocopherol in tissue homogenates! However,
living animals keep a very tight control of catalytic ions via storage (e. g., ferritin)
and transport (e. g., transferrin) proteins, so the observed “better” antioxidant
activity of γ-tocopherol compared with α-tocopherol reveals nothing of chemical
significance and does not translate to the living.

Chart 1

238
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The problem remained as to whether or not the tocopherols were RTAs.
Another excellent Post-Doc, Graham Burton, agreed to explore this matter,
starting with the building of a much more sensitive apparatus for measuring
oxygen uptake than our original. This work, and much more, is reported in the
Account: Vitamin E. Application of the Principles of Physical Organic Chemistry
to the Exploration of its Structure and Function (12). Graham discovered that
all four tocopherols were excellent ROO•-trapping antioxidants, with activities
increasing in the order: δ<γ~β<α, an order that is the same as the order of their
bio-activities (as judged by their abilities to cure animals of Vitamin E deficiency
symptoms). Furthermore, α-tocopherol was far and away the best trap for peroxyl
radicals yet reported, e. g., k5 = 320 x 104 M-1s-1 for α-tocopherol vs. 1.4 x 104
M-1s-1 for BHT. Extensive experiments (applying Physical Organic Chemistry
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

principles) demonstrated that the high activity of α-tocopherol was related to

the extent of overlap of the (rehybridized) 2p-type lone pair on the para-oxygen
with the π-electrons on the aromatic ring. In α-tocopherol the heterocyclic ring
ensures that orbital overlap is high. However, it is not perfect (the dihedral angle,
Θ, between the orientation of this lone pair and the ring plane is ca. 73° while
the optimum angle would be 90°). Indeed, flattening the heterocyclic ring by
reducing the number of its carbon atoms from 3 to 2 increased Θ to 84° and, as
expected, also increased antioxidant activity (cf. Scheme 2). This phenol had k5 =
570 x 104 M-1s-1. Although k5 is still two or three orders of magnitude below the
diffusion limit, for many years I thought it represented the best phenolic RTA that
could ever be made. I was wrong, see reference (3). Perhaps more interesting than
my own short sightedness, was that when this compound was “decorated” with
the C16 phytyl “tail” of the tocopherols, the resultant compound showed roughly
twice the biopotency of Vitamin E in one animal model (the rat curative myopathy
assay). This last result was one of the two final blows to the powerful “Vitamin
E is not an antioxidant” movement. The other was our demolition of the highly
popular Diplock-Lucy theory that Vitamin E stabilized bio-membranes by some
(unprecedented) physical interaction of the methyl groups on its phytyl ‘tail’ with
cis-ene ‘pockets’ in the polyunsaturated fatty acid ‘tails’ of the phosphoplipids.
However, when we replaced the phytyl group in α-tocopherol by an unbranched
C11 or C13 paraffinic tail the resulting compounds showed (rat curative myopathy
test) the same bioactivity as α-tocopherol!

Scheme 2

239
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Having established that the tocopherols were superb peroxyl radical trapping
antioxidants, the obvious question became: Is Vitamin E the sole lipid-soluble
RTA in humans? To answer this question my lab had to become bio-oriented. We
learned how to draw blood from one another. (No worries about AIDS in those
days.) We learned how to separate red blood cells and lipoproteins, and how to
remove the interior content of the red cells to leave behind “ghost” membranes
(to get rid of hemoglobin and so prevent iron catalysis of the planned oxidations).
We also invested in a very early liquid chromatograph – mass spectrometer
(LC/MS) instrument to enable us to quantify the tocopherols (mainly α) in our
samples. Then we exploited a procedure, pioneered by Lee Mahoney and Stefan
Korcek at the Ford Motor Company, that “titrates” for all RTAs in an (oxidizable)
sample by providing a steady supply of ROO• at a known rate, Ri, from the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

thermal decomposition of an azo-initiator (the reason we had to eliminate the

hemoglobin). The length of the induction period, τ (see Equation III), then yields
the concentration of the unknown RTA via:

When we finally felt confident that our protocols covered all foreseeable
eventualities, we took samples of blood plasma and ghost membranes, divided
each sample into two portions, one portion for the measurement of [tocopherols]
by LC/MS, the other for [unknown RTAs] by oxidation (equation IV, adding a
readily oxidized hydrocarbon when necessary). Within our limits of accuracy,
[tocopherols] = [unknown RTAs] for all the chemists in my lab. This was also
the case for an individual with very severe Vitamin E deficiency. So Vitamin E
must certainly be the major lipid-soluble RTA in humans. (Later work by others
showed that the hydroquinones, ubiquinol-9 and -10, can also function as RTAs
in low density lipoproteins. In general, hydroquinones are poor RTAs because
they yield semiquinone radicals and these can react with O2 to form the chain
carrier, HOO•.)
“Antioxidants” were as popular 30 years ago as they are today. Then, as now,
the word was frequently employed indiscriminately to describe (bio)-molecules
having structures that precluded any ability to trap ROO• radicals or to reduce Ri
by interfering with redox reactions that can lead to free radicals (e.g., FeII + H2O2
→ FeIII + HO- + HO•) and hence to an increase in lipid peroxidation. All kinds of
‘natural’ products were (and are) promoted because they contained reducing agents
that have been mislabeled as ‘antioxidants’. I have generally ignored even the
wildest of such claims because they generally harm only the wallets of unscientific,
but health conscious, consumers. However, on half-a-dozen occasions I’ve been
sufficiently annoyed to react. Two of my ‘reactions’ to Vitamin E nonsense are
described below.
α-Tocopherol is sold to consumers as Vitamin E (d,l-α-tocopheryl acetate,
or all-rac-α-tocopheryl acetate) and (a bit more expensively) as Vitamin E “from
natural sources” (d-α-tocopheryl acetate). The former is synthetic and is correctly
2RS, 4′RS, 8′RS-α-tocopheryl acetate. The latter is made from a mixture of natural
tocopherols (mainly γ-tocopherol) distilled from soy beans, then fully methylated
on its aromatic rings, and finally acetylated, it has the 2R,4′R,8′R configuration.
(Note: animals, including man, have a tocopherol transport protein, TTP, which
240
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 discriminates strongly in favor of α- over γ-tocopherol. This protein is responsible
for the inter- and intra-cellular distribution of the water-insoluble tocopherols
and protection from cytochrome P450-mediated oxidation.) The actetate is the
preferred vehicle for human and animal consumption because (ex-vivo) the
free phenol is fairly rapidly oxidized in air. The then accepted answers to two
questions posed below “got my goat”.

1) What is the ‘cost’ in Vitamin E uptake when taking the acetate rather
than the free phenol? Remember that only the free phenol can capture
ROO• radicals. Furthermore, the acetate is not absorbed from the gut, it
must first be hydrolyzed to the free phenol and it is the free phenol that is
absorbed. Accepted answer: The acetate has twice the bioactivity of the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

phenol!!
2) What is the activity of ‘natural’ Vitamin E (as acetate) vs. synthetic α-
tocopheryl acetate (the Vitamin E standard)? Accepted answer: 1.36 /
1.00!

To answer these (and a host of other physiological) questions correctly,

and in the hope of obtaining information more acceptable to rational, unbiased
observers, Graham and I applied the competitive kinetic methods of Physical
Organic Chemistry. Building on commercial practices, pure γ- and δ-tocopherol
were methylated with CD3 groups to give 2R,4′R,8′R-α-tocopherol with masses
3 and 6 Daltons, respectively, greater than the unlabelled material. Traditional
bioassays for Vitamin E involve dissolving the material to be assayed in a
large excess of tocopherol-stripped corn oil and administering this oil directly
(intubation) into the stomachs of several Vitamin E deficient rats (several rats
because bio-scientists love statistics, while most chemists are content with
Avagadro’s 6 x 1023). We followed this procedure using tocopherol-stripped corn
oil to which was added equimolar concentrations d3-α-tocopheryl acetate and
d6-α-tocopherol. Some rats were E-deficient (traditional bioassay) and some that
were not E-deficient. Later, the rats were “sacrificed” (killed, in English), divided
into bits (dissected), and the ratios of α-tocopherol from phenol to α-tocopherol
from acetate (d6-/d3-α-tocopherol) were measured in about ten different tissues.
In the blood and tissues of all the rats, the d6/d3 ratio was about 0.5, supporting
earlier conclusions that the acetate was a better source of Vitamin E than Vitamin
E! Since this was a senseless result, one evening I took my courage in both hands,
used another hand to grab and then swallow a capsule containing equimolar
d6-α-tocopherol and d3-α-tocopheryl acetate, and yet another hand to pick up a
large glass of red wine as a ‘chaser’. Supper continued very pleasantly, in part
because I didn’t tell my wife that what I’d been doing might make it my last.
The next morning I was bled at work and the LC/MS analysis of α-tocopherol
gave d6/d3 ratio of 1.0. This proof that “Man is not a rat” was confirmed by other
‘volunteers’ in my lab, but was disputed by my secretary: “All men are rats”.
Upon reflection, I decided she might be partly correct! However, I was not willing
to swallow a huge dose of corn oil on an empty stomach, so the rats were again
called into play. This time the rats were given the two d-labeled Vitamin Es with
their food (no wine). All these new results gave a d6/d3 ratio of 1.0, proving that

241
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 my secretary understood the world better than I. Apparently, when following the
long-approved protocol, about 50% of the free α-tocopherol given to the rats is
oxidatively destroyed in their stomachs, whereas the acetate passes safely through
the stomach into the small intestine where it is hydrolyzed and absorbed. Clearly,
the long-accepted test for Vitamin E activity had been sacrificing rats on the altar
of stupidity, not science.
Far more contentious (because of its financial implications for the
manufacturers of synthetic Vitamin E) was a ratio Graham and I obtained by
application of the same d-labeling/competitive kinetic technique to the second
question. We found that the 2R stereoisomer of α-tocopherol was absorbed and
retained in the human body twice as effectively as the racemic material, meaning
that Vitamin E “from natural sources” is twice as good a source of this Vitamin
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

for humans as is plain, i.e., synthetic, “Vitamin E”! This result was contested
(futilely) but is fully consistent with TTP only binding (accepting) the “natural”
stereoisomer, as was later proven by others using recombinant human TTP. If you
are into ‘antioxidants’ just remember to read the label and the mnemonic: “To
hell with the l”, when purchasing Vitamin E.
Our pioneering work on the “biokinetics” of α-tocopherol uptake and loss in
animals and in humans (both healthy and with certain genetic diseases such as
an inability to synthesize TTP), was interesting and exciting. Uptake (and loss)
rates differ dramatically between tissues with the brain and nerves being the most
determined to hang on to their supply when ‘new’ α-tocopherol from food was
not available. Under such conditions, TTP is up-regulated, and lipoproteins in
the blood shuffle α-tocopherol from less important tissues, such as the liver, to
the brain. The kinetics of these processes follow “stretched exponentials”, like
the [CH3• + CH3OH]glass reaction described above. In this case, the stretched
exponentials are due to the presence of many organelles within each tissue, each
of which will have its characteristic first-order uptake/loss kinetics. This is an
interesting area, but one I thought unlikely to be scientifically, or intellectually,
rewarding (except in terms of publication numbers!).

Vitamin E as a Pro-Oxidant!
About the time that I finally gave up Vitamin E biokinetics I was seduced
into studying the oxidation of Low Density Lipoproteins (LDL). There was
some evidence that (per)oxidized LDL was a causative factor in the development
of atherosclerosis. Most work was qualitative and of little value. However,
eventually Australian researchers (Roland Stocker and Vince Bowry, an ex-PDF
of mine) undertook quantitative studies that used (mainly) water-soluble
azo-initiators, iN=Ni. Their studies allowed the number of ROO• radicals
generated / LDL particle to be calculated and led to the amazing discovery
that α-tocopherol was a pro-oxidant in LDL particles dispersed in saline at
physiological pH. I was able to explain at least a part of this astonishing result to
the Australians and a very fruitful collaboration ensued. In brief, there are two
reasons that α-tocopherol is a pro-oxidant in LDL:

242
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 (i) The water soluble iOO• radicals can only pass their radical character into
an LDL particle by abstracting the phenolic H-atom from an α-tocopherol
molecule which, we had demonstrated by NMR methods, has its OH
group at the surface of LDL-mimicing phospholipid liposomes dispersed
in water.
(ii) The α-tocopheroxyl radical formed in this initiation event, actually can
abstract a bis-allylic H-atom from a polyunsaturated fatty acid moiety
in the LDL. (This reaction is very slow, but the tocopheroxyl radical
is “stuck” in the LDL by its insolubility in water and the pentadienyl
radical it produces is very rapidly trapped by O2.) Moreover, two radicals
cannot coexist in such small particles which means that LDL particles
contain either no radical centers (and are not oxidizing), or a single radical
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

center (and are oxidizing). The kinetics are virtually identical to those
for emulsion polymerization, as described by Walling (vide supra) in
his book Free Radicals in Solution (Wiley, 1957). This similarity was
celebrated by a joint 1993 PNAS publication with the two Australians and
Walling, as described in the Account: The Unexpected Role of Vitamin E
(α-Tocopherol) in the Peroxidation of Human Low Density Lipoprotein
(13).

Invention of the First Superoxide Thermal Source, SOTS-1

During my lab’s bio-era, we were naturally interested in DNA, particularly in
quantitative studies on peroxyl radical-induced DNA damage. The only existing
work of this kind involved a commercially available, water-soluble, positively
charged, azo-initiator, +iN=Ni+. Plasmid DNA was cleaved, but the number of
double strand breaks per +iOO• radical generated was not reported. Moreover,
this result was simply assumed to be true for all peroxyl radicals. I was unwilling
to accept this assumption because DNA bears a negative surface charge that would
attract and hold +iOO• and because the electron-withdrawing effect of the positive
charge would be expected to make these +iOO• radicals far more reactive than
neutral peroxyls, cf., the reactivities of Cl3COO• vs. Me3COO•. Physical organic
chemists must be able to synthesize the molecules they wish to study. We did
exactly that and found that neither water-soluble neutral iOO•, nor negatively
charged -iOO•, cleaved DNA to any significant extent. These results made me
doubt reports that superoxide (O2•-) readily cleaved DNA. Superoxide is the major
radical formed in vivo. It generally functions as a reducing agent (despite its name),
but its conjugate acid, HOO•, pKa 4.6, is present at ca. 1% of the [superoxide] at
physiological pH (7.4). (The peroxyl radical, HOO•, is slightly more reactive than
alkylperoxyls.) So we had to synthesize a superoxide thermal source that would
produce superoxide at a known and useful rate in water at 37 °C and pH 7.4. This
was achieved by means of the chemistry shown in Scheme 3. Plasmid DNA was
not cleaved by superoxide!

243
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Scheme 3

Invention of a Nitric Oxide Cheleotropic Trap, NOCT

Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

A second small inorganic radical, nitric oxide (•NO), is also of large

biological importance. It is synthesized in vivo both as an antibacterial agent in
macrophages and as an important signaling molecule (e.g., for blood pressure
control). When •NO is generated together with O2•- in the vacuole of a white
blood cell that has engulfed a bacterium, these two rather unreactive radicals
combine to form peroxynitrite, ONOO-. We demonstrated (contrary claims
notwithstanding) that peroxynitrite’s conjugate acid undergoes a fast homolysis
under physiological conditions to produce nitrogen dioxide and the very reactive
hydroxyl radical, ONOOH → NO2• + HO•. Under physiological conditions, most
peroxynitrite will react with CO2 to give ONOOCO2-. This ion (with the lovely
empirical formula CNO5) also undergoes a fast homolysis to NO2• + CO3•-, both
of which are also powerful oxidizing radicals. The engulfed bacterium does not
have a chance! Nor did early workers in this field who were unaware of: (i) the
reaction of ONOO- with CO2, and (ii) the reaction of •NO2 with O2•- to form
peroxynitrate, O2NOO- which, confusingly, absorbs at 302 nm, the wavelength
normally employed to monitor peroxynitite decay! My lab had lots of fun helping
to sort out some of this ‘messy’ chemistry.
Since nitric oxide was so ‘hot’ in 1990, I thought it would be nice to have a
method for monitoring its production in time and space in vivo. It seemed possible
that •NO might add to a suitably constrained diene to form a stable nitroxide
radical that would be easy to monitor by EPR spectroscopy. However, •NO is
very unreactive so addition to the diene would require a strong driving force.
This driving force might be provided by aromatization of a 1,3-cyclohexadiene
moiety. All of which suggested that 1,2-bis(exo-isopropylidene)cyclohexa-3,5-
diene (Scheme 4) might be suitable. This (somewhat transient) compound can
be synthesized by the photo-decarbonylation of 1,1,3,3-tetramethyl-2-indanone
and, if the diene trapped •NO as hoped, it would yield a known stable nitroxide
radical, see Scheme 4, Since I was going on a Humboldt Fellowship to Germany, I
collaborated with Hans Korth, an EPR expert at U. Essen (and several other organic
and medicinal chemists), to explore this reaction. To my surprise, the chemistry
worked like a charm, first time! However, this and related •NO trapping reactions
were later shown by my German collaborators to be too slow to be of much use
for monitoring •NO in vivo.

244
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Scheme 4

Nanosecond Laser Flash Photolysis (LFP) and Solvent Effects

Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

The wonderful LFP technique proved ideal for measuring the rate constants
for reactions involving reactive free radicals. It was brought to my lab by another
great Post-Doc, J. C. (Tito) Scaiano, and quickly replaced EPR spectroscopy as
our main research tool. This technique was more versatile than EPR and, most
importantly, LFP made it simple to study the kinetics of alkoxyl radical, RO•,
reactions, particularly H-atom abstractions. Such studies took on increased vigor
when we discovered that benzyloxyl and related radicals, ArCRR′O•, had a fairly
strong absorption in the visible (unlike Me3CO•). This meant that the reactions
of ArCRR′O• radicals could be monitored directly (rather than indirectly via the
“probe” method).
The ease and speed with which LFP measurements could be made encouraged
me to investigate the effect of solvents on the rate “constants” of radical reactions.
The prevailing wisdom among synthetic organic chemists was that radical
reactions did not exhibit any significant solvent effects. Physical organic chemists
knew this was not true for two good reasons:
First, in the 1950’s Glen Russell (vide supra) had discovered that the
free radical chain chlorination of 2,3-dimethylbutane using Cl2 gave a higher
tertiary/primary chloride product ratio, [Me2CHCMe2Cl]/[Me2CHCHMeCH2Cl],
in benzene than in CCl4. Russell attributed this result to formation of a chlorine
atom-benzene π-complex that was less reactive and more selective in H-atom
abstractions than a ‘free’ Cl• atom in CCl4. Russell’s interpretation was challenged
some 30 years later and his two H-atom abstractors, Cl• and π-(ClC6H6)•, were
increased to three by the addition of a third abstractor, a Cl• atom-benzene
σ-complex. Our LFP and spectroscopic studies soon demonstrated that all the
facts could be fully and most simply (Occam’s razor) explained with Russell’s
two original abstractors. This work is summarized in an Account: The Unusual
and the Unexpected in an Old Reaction. The Photochlorination of Alkanes with
Molecular Chlorine in Solution (14).
Second, and of more relevance to organic chemists, roughly a decade later
Cheves Walling (vide supra) had reported dramatic solvent effects on tert-butanol
/ acetone ratios in competitions between H-atom abstraction from cyclohexane by
tert-butoxyl and its β-scission, reactions 22 and 23.

245
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 For example, with 0.1 M cyclohexane at 40 °C the [tert-butanol] / [actone]
ratios were 7.5, 3.5, 1.6, and 1.4 in CFCl2CF2Cl, C6H6, MeCN, and MeCOOH,
respectively. Walling pointed out that “solvent effects on competitive radical
reactions must reflect different degrees of solvent interaction with the transition
states rather than with the radicals” and that “Solvent interactions with the
transition state for β-scission presented no difficulties, but in the transition state
involving an alkoxy(l) radical and a substrate such as cyclohexane, solvent
molecule should be sterically excluded from close vicinity to the alkoxy(l)
radical.” Solvent effects on the [Me3COH] / [Me2CO] ratio were therefore
ascribed “chiefly to solvation of the transition state for the β-scission process”.
Roughly 30 years later, our LFP measurements proved that this was correct. That
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

is, kPhCMe2O•/C6H12 was 1.2 x 106 M-1s-1 in six solvents of widely different polarity,
whereas the rate constants for β-scission of PhCMe2O• (to acetophenone and
Me•) varied substantially. When we reported these results, we naturally added the
footnote: “Hats off to Cheves Walling”.
Two years later, in 1995, we discovered (thanks to LFP) some very much
more dramatic solvent effects on the rate “constant” for H-atom abstractions by the
cumyloxyl radical, PhCMe2O•. The substrates were not hydrocarbons, but were
phenol and tert-butyl hydroperoxide. It was clear from our earlier studies that
these kinetic solvent effects (KSEs) could not be due to effects on the cumyloxyl
radical but must, instead, be due to effects on these hydroxylic substrates. A
simple reaction Scheme was proposed for H-atom abstraction from a hydrogen
bond donor (HBD), XH, by radical Y• in a hydrogen bond acceptor (HBA) solvent,
S:

Scheme 5

From this Scheme, the measured rate constant in solvent S is given by,

where k0XH/Y• represents the rate constant in a non-HBA solvent (i. e., an alkane).
Scheme 5 was based on three assumptions:

(1) Each XH can act as an HBD to only one single HBA molecule, S, at any
one time.
246
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 (2) The magnitude of the equilibrium constant, KSXH/S, for formation of the
HB complex, XH---S, was independent of the nature of the surrounding
medium (e. g., its dielectric constant).
(3) Cumyloxyl radicals cannot abstract the hydroxylic H-atom from the XH-
--S complex (for steric reasons).

The magnitude of the KSE for these cumyloxyl radical reactions was greater
for phenol than for tert-butyl hydroperoxide, e. g., kCCl4/kMe3COH was 240 for
phenol but only 37 for the hydroperoxide. It took another six years and a lot
of LFP measurements before I could provide a quantitative explanation for our
observations. However, in the interim, two predictions that come directly from
Scheme 5 were quickly put to the test.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

(i) It was predicted that equilibrium constants for HB formation could be

determined by kinetic measurements. This was readily demonstrated
by comparison of KSPhOH/S calculated for phenol from kSPhOH/RO• values
measured for several S at different phenol concentrations in CCl4, with
PhOH/S equilibrium constants determined by infrared spectroscopy in
CCl4.
(ii) It was predicted that KSEs should be independent of the attacking
radical (since they depend only on the strength of the HB between
the substrate and a solvent molecule). This was verified by a visting
chemical dynamo, Luca Valgimigli (U. Bologna), using phenol (8
solvents) and α-tocopherol (13 solvents) and two H-atom abstracting
radicals having very different reactivities, tert-alkoxyl (RO•) and
2,2-diphenylpicrylhydrazyl (dpph•). Rate constants in alkane solvents
were about two orders of magnitude faster than those in the ‘slowest’
solvents used. Plots of log kSRO• vs log kSdpph• gave excellent straight
lines with the predicted slope of 1.0 despite the RO• rate constant in any
particular solvent being faster than that for dpph• by factors of more
than 106 for α-tocopherol and more than 1010 for phenol! A new and
most important principle for physical organic chemistry was presented:
“Provided rate constants have been measured for the reaction of one
radical with a substrate in a range of solvents, the measurement of the
rate constant for reaction of the same substrate with some different
radical need be made in only one of these solvents for values in all the
other solvents to be predicted accurately.”

Each of these early KSE experiments produced one similar anomaly in the
plots of log kSRO• vs log kSdpph• for phenol and α-tocopherol. For both substrates,
the values of kSRO* were “normal” for S = tert-butanol, but the values of kSdpph•
in this solvent were “too large”. The reason for these abnormally fast “hydrogen
atom transfers” from phenols to the dpph• radical in tert-butanol were eventually
traced, thanks to some great detective work by Grzegorz Litwinienko (U. Warsaw),
to a solvent-dependent difference in mechanism. The mechanism is a (slow) H-
atom transfer from the phenol to dpph• in most solvents, but a (fast) electron
transfer from even traces of the phenoxide anion to the (very electron-deficient)

247
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 dpph• radical in solvents that could support at least some ionization of the phenol,
i.e., alcohols and water, Scheme 6. This (new) mechanism, christened Sequential
Proton Loss Electron Transfer (in part because of its lovely acronym, SPLET),
is described together with the development of the standard H-bond quantitative
model of KSEs that is based on Scheme 5 (vide infra), in the Account: “Solvent
Effects on the Rates and Mechanisms of Reaction of of Phenols with Free Radicals”
(15).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

Scheme 6

During our long struggles to accumulate enough data to quantify KSEs for
H-atom transfers, Mike Abraham (University College London) was struggling
to develop two reliable, quantitative, thermodynamic scales relating to 1:1
HBD/HBA complex formation. One scale gave the HBD activities of many
molecules, symbol αH2 (range 0.00 for alkanes, to ca. 1.0 for carboxylic acids)
and the other a parallel thermodynamic scale of HBA activities of an equally
huge number of molecules, symbol βH2 (range 0.00 for alkanes to 1.00 for
hexamethylphosphorustriamide, HMPA, the strongest organic HBA). Both scales
rely on 1:1 HBA/HBD equilibrium constants in CCl4 that can be easily and
precisely determined by infrared spectroscopy. To cut a very long story short, I
found that the KSEs for all H-atom abstractions examined (that had no significant
SPLET chemistry) gave linear plots when log kSY•/XH values (in roughly a dozen
solvents) were plotted against the βH2 value for the solvent. Furthermore, the
slopes of these linear free energy plots (which were, of course, independent
of Y•) varied from 0.00 for cyclohexane (which has no KSE), through -3.0
for α-tocopherol, -3.5 for tert-butyl hydroperoxide, -5.1 for phenol, to -6.0
for 3,5-dichlorophenol. These (and many other) slopes in neat solvents were
proportional to Abraham’s αH2 values for the substrates. Not only did these results
confirm our earlier assumption #2 (vide supra), but also, and more importantly,
they led to a simple equation for the quantitative description, and prediction,
of kinetic solvent effects on any H-atom transfer in solution, Equation VI. This
equation implies that for an XH having αH2 = 1.0, kSXH/Y* would decrease by
just over eight orders of magnitude on changing S from an alkane to HMPA!
When I’d started work on KSEs I’d hoped that we might eventually learn how to
describe their cause in some quantitative and predictive manner. When I finally
discovered the answer, I was flabbergasted by its simplicity. Equation VI stands,
in my mind, as a fitting monument to the rationalizing power of Physical Organic
248
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Chemistry, to the hard work and dedication of the many kineticists who have
worked in my lab, to the totally separate efforts of Mike Abraham at UCL to
provide quantitative hydrogen bonding scales, and (as always) to serendipity.

Over some 60 years I have been privileged to work with a great many brilliant
chemists: students, Post-Docs, NRC colleagues, and Professors from Academia.
Unfortunately, only a few of these individuals could be named in this Chapter, but
I recognize that everything I know about Physical Organic Chemistry, I owe to
them. Thank you. I also express my love and gratitude to my late wife, Cairine,
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

who not only put up with my enthusiasm for chemistry, but also “mothered” each
and every one of my coworkers (Figure 7).

Figure 7. Cairine and Keith Ingold as I receive the 1998 NSERC medal. I "hold
the gold", but it was Cairine who earned it.

References
1. Davies, A. G.; Garratt, P. J. UCL Chemistry Department 1828-1974; Science
Reviews 2000, Ltd.: 2013; pp 144−145.
2. Ingold, K. U. Chem. Rev. 1961, 61, 563–589.
249
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 3. Ingold, K. U.; Pratt, D. A. Chem. Rev. 2014, 114, 9022–9046.
4. Howard, J. A.; Ingold, K. U. Can. J. Chem. 1962, 40, 1851–1864.
5. Pratt, D. A.; DiLabio, G. A.; Mulder, P.; Ingold, K. U. Acc. Chem. Res.
2004, 37, 334–340.
6. Howard, J. A.; Ingold, K. U. Can. J. Chem. 1967, 45, 793–802.
7. Carlsson, D. J.; Ingold, K. U. J. Am. Chem. Soc. 1968, 90, 7047–7055.
8. Griller, D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 317–323.
9. Griller, D.; Ingold, K. U. Acc. Chem. Res. 1976, 9, 13–19.
10. Griller, D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 193–200.
11. Ingold, K. U. In Hydrogen-Transfer Reactions; Hynes, J. T., Klinman, J. P.,
Limbach, H.-H., Schowen, R. L., Eds.; Wiley: New York, 2007; Vol. 2, pp
875−893.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch010

12. Burton, G. W.; Ingold, K. U. Acc. Chem. Res. 1986, 19, 194–201.
13. Ingold, K. U.; Bowry, V. W. Acc. Chem. Res. 1999, 32, 27–34.
14. Ingold, K. U.; Lusztyk, J.; Raner, K. D. Acc. Chem. Res. 1990, 23, 219–225.
15. Liwinienko, G.; Ingold, K. U. Acc. Chem. Res. 2007, 40, 222–230.

250
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Chapter 11

Diradicals – A Fifty Year Fascination

Weston Thatcher Borden*

Department of Chemistry and the Center for Advanced, Scientific

Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Computing and Modeling, University of North Texas, 1155 Union Circle,

#305070, Denton, Texas 76203-5070
*E-mail: [email protected]

Diradicals are molecules in which two electrons occupy two

molecular orbitals that have the same or nearly the same
energies. This chapter describes the author’s half century of
fascination with this class of molecules and the theoretical and
experimental research on diradicals that he has done during this
period.

Introduction
Diradicals/Biradicals

Diradicals are molecules in which two electrons occupy two molecular

orbitals (MOs) that have the same or nearly the same energies (1, 2). Some
authors call such molecules “biradicals.” However, molecules with two positive
charges are called “dications” (not “bications”); molecules with two negative
charges “dianions” (not “bianions”); and molecules with two unpaired electrons
are called “diyls” (not “biyls”). It follows, therefore, that the correct name for
molecules in which two electrons occupy two MOs that are degenerate or nearly
degenerate in energy should be “diradicals”, not “biradicals” (3). “Diradicals”
will be used exclusively throughout this chapter.

The Contents of This Chapter

In 1982 a book entitled Diradicals, which was edited by me and to which I

contributed a chapter, was published (2). A huge amount of new research about
diradicals has been published during the past 33 years; and, were I to edit a
similar book now, it would have to be a multivolume series. Therefore, within

© 2015 American Chemical Society

In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 the confines of this book chapter, I cannot possibly provide anything approaching
an encyclopedic coverage of diradicals.
Three years ago Professior Carl Lineberger and I coauthored a review
entitled, “The Synergy Between Qualitative Theory, Quantitative Calculations,
and Precise Experiments in Understanding, Calculating, and Measuring the
Energy Differences Between the Lowest Singlet and Triplet States of Organic
Diradicals” (4). I knew that the chapter I wrote for this book would wind up
discussing some of the diradicals that were covered in that review. However, I
certainly wanted to avoid duplicating in this chapter the review that I published
with Carl.
Tom Strom, the Co-Editor of this book, had requested that I include some
biographical information in this chapter. As I began to think about how I would
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

write this biographical section, I realized that one way of describing some of what
is now known about diradicals would be to describe the history of how I learned it.
Recounting the evolution of the author’s understanding of an area of science
is not the traditional way to write a book chapter about that area. Nevertheless,
I thought that my taking this opportunity to give an account of how my own
understanding of diradicals has developed over the past 50 years might be a useful
contribution to the history of this area of chemistry. In addition, I thought that
reading about the slow pace of the evolution of my understanding of diradicals
may be encouraging to those readers, who, like me, take a long time to figure
things out (5).
Having decided that the first part of this chapter would be a description of
the electronic structures of diradicals, organized around the history of how my
own understanding of this area of theoretical chemistry developed, it seemed
appropriate that the second major part of this chapter should recount how some
of my theoretical predictions about diradicals have been tested experimentally.
I believe very strongly that providing experimentalists with explanations is
only part of a theoretical chemist’s job. Even more important is using one’s
understanding to make clear predictions about the outcomes of experiments that
have not yet been performed.
Making experimentally testable predictions is usually a very good way
to motivate experimentalists to perform the experiments that are necessary, in
order to test those predictions. Occasionally, I have even designed the necessary
experiments myself and then collaborated with the chemists who performed the
experiments on interpreting the results..
The second part of the chapter begins with a description of the experimental
techniques for measuring the singlet-triplet energy differences (ΔEST), in
diradicals. I have focused on negative ion photoelectron spectroscopy (NIPES),
because it can, in principle, provide very accurate values of ΔEST, no matter how
large or small ΔEST is.
This introductory section on experimental methodology is followed by
descriptions of the results of measurements by NIPES of the values of ΔEST in
several different types of diradicals. The experimental results on the effects on
ΔEST of the replacement of methylene groups in trimethylenemethane (TMM)
and in meta-benzoquinodimethane (MBQDM) by oxygens are described. The
predictions, published prior to any experiments, were that this substitution would

252
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 have a huge effect on ΔEST in TMM → oxallyl (OXA) (6), but very little effect
in meta-benzoquinodimethane (MBQDM) → meta-benzoquinone (MBQ) (7).
My group’s collaborations with Professor Carl Lineberger on the experimental
verification of the first prediction (8, 9) and with Dr. Xuebin Wang on the
experimental verification of the second (10, 11) are described.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

When one thinks singlet-triplet energy differences in diradicals, one thinks of

Hund’s rule and its prediction that the triplet should be the ground state of any
diradical in which the two partially occupied MOs have exactly the same energies
(12, 13). One of the aspects of diradicals that has interested me most over the
past half century is the possibility that some diradicals might be found to violate
Hund’s rule. Therefore, I have included in this chapter a discussion of predicted
Hund’s rule violations and the experimental testing of these predictions in several
diradicals (14).
One of these experiments involved obtaining the NIPE spectrum of the radical
anion of cyclooctatetraene (COT•–) (15). This experiment, which was done in a
collaboration with Paul Wenthold and Carl Lineberger, resulted in the observation
of not only the D4h transition structure (TS) for ring inversion in D2d COT but also
the D8h TS for π–bond shifting in singlet COT (Figure 1). The NIPE spectrum of
COT•– showed that, as had been predicted (16, 17), the D8h TS for bond shifting
in singlet COT is lower in energy than the D8h equilibrium geometry of the lowest
triplet state, thus providing a clear violation of the strictest version of Hund’s rule
(15).

Figure 1. The D4h transition structure (TS) for ring inversion of D2d
cyclooctatetraene (COT), and the D8h TS for bond shifting.

The next-to-last section of this chapter describes my group’s very recent

research on the (CO)n series of molecules. The discussion focuses on the very
surprising prediction that (CO)4 has a triplet ground state (18), and on the
experimental confirmation of this prediction by NIPES experiments, carried out
253
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 by Dr. Xuebin Wang (19, 20). The chapter concludes with a brief description of
some of the future collaborations that Xuebin Wang and I have planned.

A Brief History of Diradicals and of My Interest in Them

A Brief History of the First Fifty Years of Diradicals
The history of diradicals extends back over a century to the preparation of
Chichibabin’s (21), Thiele’s (22), and Schlenck’s (23) hydrocarbons, respectively
1-3 in Figure 2. Although resonance structures with two unpaired electrons can be
written for 1 and 2, the dominant resonance structure for each of these derivatives
of para-benzoquinodimethane is that shown in Figure 2, with all of the π electrons
paired and occupying bonding MOs. Neither 1 nor 2 is actually a diradical (24).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Figure 2. Chichibabin’s (1) (21), Thiele’s (2) (22), and Schlenck’s (3) (23)
hydrocarbons.
In contrast, 3 is a derivative of meta-benzoquinodimethane (MBQDM), a
hydrocarbon for which it is impossible to write any Kekulé structures in which
all of the electrons are paired. Longuet-Higgins showed that in such non-Kekulé
hydrocarbons two electrons occupy a pair of non-bonding MOs that have the same
Hückel energy, α (25). Therefore, 3 is a true diradical, and 2015 marks the 100th
anniversary of its preparation (23).
In 1966 two landmark publications ushered in the modern era of diradical
chemistry. First, Rowland Pettit reported the generation of cyclobutadiene (CBD)
by oxidation of its Fe(CO)3 complex. Pettit did not isolate CBD; instead, as shown
in Figure 3, he trapped it by Diels-Alder reactions (26). The stereospecific Diels-
Alder reaction of CBD with fumarate and maleate diesters was taken as evidence
that the reactive state of CBD is a singlet, rather than a triplet.

Figure 3. The generation and stereospecific, Diels-Alder trapping of CBD (26).

254
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Second, Paul Dowd reported the preparation of the simplest non-Kekulé
hydrocarbon, trimethylemethane (TMM), by photolysis of 4-methylenepyrazoline
(Figure 4) (27–29). The triplet state of TMM was characterized by its EPR
spectrum, and Dowd subsequently showed the triplet to be the ground state of
TMM (30).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Figure 4. The generation of triplet TMM by photolysis of 4-methylenepyrazoline

(27).

CBD and TMM each contain four π electrons. In each molecule one pair
of electrons occupies a bonding π MO, leaving the remaining two π electrons to
be distributed between two MOs. In both square (D4h) CBD and in trigonal (D3h)
TMM these two MOs are nonbonding and are degenerate by symmetry. Therefore,
square CBD and trigonal TMM are both diradicals; and according to Hund’s rule,
at these high-symmetry geometries, both diradicals should have triplet ground
states (12, 13). Dowd’s EPR experiments did, in fact show that TMM does have
a triplet ground state (30), but Pettit’s trapping experiments (26) at least raise the
question of whether CBD might actually have a singlet ground state.

How I Became Interested in Diradicals

In 1966 I was a graduate student at Harvard; and I had several late-night talks
about TMM with Paul Dowd, who was a Junior Faculty member in the Chemistry
Department at the time. However, my interest in diradicals actually began four
years earlier, when I was an undergraduate at Harvard.
In the autumn of 1962 I borrowed my roommate, Bob Joffe’s, copy of Andy
Streitwieser’s book, Molecular Orbital Theory for Organic Chemists (31). At the
end of that academic year, Bob decided to give up chemistry, in order to become
a lawyer. In fact, he attended Harvard Law School and went on to become the
Managing Partner of Cravath, Swain, and Moore, a very prestigious New York
City law firm. Bob and I remained friends until he died, much too young, of
pancreatic cancer in 2010 at age 66.
I kept Bob Joffe’s copy of Streitwieser’s book, and I still occasionally consult
it. It was, in fact, this book which first sparked my interest in CBD and in TMM;
and my fascination with diradicals has now persisted for over half a century (32).
During the autumn semester of 1962 Bob Joffe and I were both enrolled in
an undergraduate organic chemistry course at Harvard, which was taught by E.J.
Corey. On night, at the end of the semester, I went to E.J.’s office to pick up my
final exam. I had done well enough on the exam that E.J. asked me about my future
career plans.
255
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 I explained to E.J. that my first chemistry Professor at Harvard had been Bill
Lipscomb, who convinced me that all of chemistry could be understood through
chemical theory. My second chemistry Professor had been Frank Westheimer,
who emphasized the importance of experimental research in chemistry. In fact,
although Professor Westheimer was arguably the father of molecular mechanics
calculations, he at least pretended to be very skeptical that theoretical chemistry
could predict anything worth measuring. I told E.J. that, having been influenced by
both Professors Lipscomb and Westheimer, I planned to become a physical-organic
chemist and to combine doing theory with performing experiments, in order to test
my own predictions.
On hearing this Professor Corey pulled down his copy of Streitwieser’s book
from his bookshelf. I expected that he was going to tell me that, if I wanted to do
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

research in theoretical chemistry, I would have to read Streitwieser’s book. Having

just finished working my way through Bob Joffe’s copy of this book, I could barely
prevent myself from visibly glowing with pride. However, to my utter dismay, E.
J. said, “Well if you are going to do theoretical chemistry, you will have to do much
better than the Hückel theory in this book.”
Crestfallen, I asked, “So what should I do?” E.J. thought for a moment and
then said. “If I were you, I would go to study with H. C. Longuet-Higgins in
Cambridge, England.”
And that is exactly what I did do, immediately after I received my B.A. degree.
I applied for and won a Fulbright Fellowship; and in the early fall of 1964, together
with all of the other Fulbright Fellows who were going to England that year I
boarded a ship that was bound for Southampton. (It is hard to believe that, only
half a century ago, people frequently crossed the Atlantic by ship, rather than by
airplane).
In a biography of Professor H. C. Longuet-Higgins that I wrote for another
ACS book, which was co-edited by Tom Strom (33), I have described some aspects
of my year in Cambridge. I recounted how, at least once during each of my
weekly tutorials with Christopher Longuet-Higgins, he would exclaim, “Weston,
stop talking nonsense!”
Despite his impatience with me, I learned a great deal from Christopher that
year about how to solve scientific problems. Mostly I learned by watching how
he worked out the answers to the questions that I asked him. Perhaps the most
important thing I saw in witnessing Christopher solve problems was that he had
complete confidence that he would eventually find the correct answer to any
problem, even if he did not know the answer when he began. It took me decades
to develop that type of confidence (5).
During my year in Cambridge, I learned a great deal about many things
from Christopher; however, I did not learn anything about diradicals from him.
Nevertheless, Christopher did encourage me to invest some of my time in working
through Robert Parr’s book on Pariser-Parr-Pople (PPP) calculations (34). I was
anxious to learn how to do PPP calculations, because they explicitly include in
the Hamiltonian operator the Coulombic repulsion between electrons. Therefore,
PPP calculations take an important step beyond Hückel theory. As I would learn,
understanding the effects of electron-electron repulsion in diradicals is the key to
understanding diradicals.

256
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 When I was in Cambridge in 1964-65, I do not think that Christopher Longuet-
Higgins knew how to write a computer program. Nor did I. Therefore, I did all of
my PPP calculations with pencil and paper. In retrospect, this was very fortunate,
because I had the opportunity to work through and really understand the nuts and
bolts of doing PPP calculations on diradicals.

The Results of PPP Calculations on Diradicals

What I Learned about Diradicals from Doing PPP Calculations
In Cambridge, England in 1964-65, by doing PPP calculations, I started
learning about the wave functions for diradicals and how to calculate their
energies. For example, I came to understand that
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

is not a proper wave function for an open-shell, singlet diradical, in which one
electron occupies MO ψi and another occupies MO ψj. I learned that the wave
function in eqn. 1 is actually a 50:50 mixture of the wave functions for an open-
shell singlet state and the Sz = 0 component of the triplet state with the same orbital
occupancy. The correct wave functions for an open-shell singlet and the Sz = 0
component of a triplet state are, in fact,

where the positive sign is for the singlet wave function and the negative sign is for
the triplet.
The symbol |.…> in eqns. 1 and 2 is called a “ket”. It denotes that the many-
electron wave function inside the ket has been properly antisymmetrized, so that
the wave function changes sign on exchanging the labels on any pair of electrons.
Expanding the kets in eqn. 2 gives, for the singlet,

and for the triplet,

It is easy to see that the wave functions in eqns. 3 and 4 are both
antisymmetrized. The negative sign in each wave function causes it to change
sign on switching the labels on the electrons. However, the singlet wave function
in eqn. 3 and the triplet wave function in eqn. 4 differ in an important way. The
singlet wave function has a symmetrical spatial part but an antisymmetrical spin
part; whereas, the triplet wave function has an antisymmetrical spatial part but a
symmetrical spin part.
Because the spatial wave function for the triplet is antisymmetrized, it is
easy to see that it gives zero for ψi = ψj. Therefore, two electrons with the
same spin – α(1)α(2), β(1)β(2), or [α(1)β(2) + β(1)α(2)]/√2 – cannot occupy the
same orbital, because, if they do, the triplet wave function vanishes. Thus, the
257
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 antisymmetrization of the spatial part of a triplet wave function results in its
obeying the Pauli exclusion principle.
Antisymmetrization of the spatial part of a triplet wave function does more
than just keep electrons of the same spin from occupying the same MO. Eqn.
4 shows that two electrons of the same spin in different MOs cannot appear
simultaneously in the same region of space (for instance in the same AO); or
the triplet wave function again vanishes. However, the spatial part of the singlet
wave function in eqn. 3 is not antisymmetrized; so there is some probability that
the electrons of opposite spin in a singlet state will simultaneously appear in the
same region of space. Consequently, triplet wave functions usually have smaller
Coulombic repulsion energies than their singlet counterparts.
This gives rise to Hund’s rule: If ψi and ψj have the same energy, then the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

electronic state of lowest energy is the triplet, in which the two electrons minimize
their Coulombic repulsion energy by occupying different orbitals with the same
spin (12, 13).
In order to obtain the probability distributions for the electrons in the singlet
and triplet wave functions in eqns 3 and 4, the wave functions must be squared.
The lower Coulombic repulsion energy of the electrons in the triplet than in the
singlet arises mathematically from the signs of the cross terms in the spatial parts
of ΨS and ΨT, when ΨS and ΨT are squared. The different signs of the cross terms
make the electrostatic repulsion energy between an electron in ψi and another in
ψj Jij + Kij in the singlet and Jij - Kij in the triplet, where

and

The Coulomb integral, Jij, in eqn. 5 gives the electrostatic repulsion energy
between a spinless electron in ψi and another in ψj. As shown in eqn. 5, this
Coulombic repulsion energy is calculated from ψi(1)2 and ψj(2)2, the probability
distributions for these two, spinless electrons.
Of course, electrons do have spin. For all three components of a triplet state,
the exchange integral provides a correction, -Kij, to the Coulomb integral, Jij. The
physical reason why Jij is corrected by -Kij for a triplet state is that, as is evident
from eqn. 4, two electrons of the same spin are prevented by the Pauli exclusion
principle from simultaneously appearing in the regions of space that ψi and ψj have
in common. Therefore, for electrons of the same spin, Jij for spinless electrons
must be corrected by -Kij.
In contrast, doing the math for the Coulombic repulsion between the electrons
in ψi and ψj in the open-shell singlet state shows that Jij is corrected by +Kij. The
physical reason for this correction is that, in an open-shell singlet state, the motions
of the electrons in ψi and ψj are, actually anti-correlated. What this means is that,
in an open-shell singlet, the pair of electrons in the singly-occupied MOs tend to
simultaneously appear in the same region of space with a higher probability than
that predicted by Jij for spinless electrons.
258
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Since the Coulomb repulsion energy between the two electrons is Jij + Kij in
the open-shell singlet state and Jij - Kij in the triplet state, the energy difference
between these two states is

However, it should be noted that eqn. 7 assumes that the optimal MOs for
an open-shell singlet state and the corresponding triplet state are the same. Many
years after I left Cambridge, England and returned to Cambridge, MA, I would
learn that in some diradicals, including TMM, the optimal sets of MOs for a singlet
and a triplet state, and even for two singlet states, are not the same.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

PPP Calculations Show that TMM and CBD Are Very Different Types of
Diradicals

Having learned the rudiments of how to compute the energies of wave

functions, such as those in eqns 3 and 4, for molecules with open shells of
electrons, I moved on to performing PPP calculations on two diradicals, D3h
TMM and D4h CBD. As already noted, each of these diradicals has four π
electrons; and in each of these diradicals two of these electrons occupy a bonding
π MO, ψ1, and the two remaining electrons are distributed between a pair of
non-bonding π MOs, ψ2 and ψ3. Despite these similarities, the PPP calculations
I performed revealed that there are some major differences between TMM and
CBD.
In the lowest singlet state of D3h TMM, group theory predicts that there are
two wave functions, Ψx and Ψy, that form the two components of a degenerate,
singlet (1E′) electronic state. These two wave functions are:

and

The manner in which the bonding π MO, ψ1, and the degenerate pair of
nonbonding π MOs, ψ2 and ψ3, are occupied by electrons in Ψx and Ψy is depicted
graphically in Figure 5. As shown symbolically in eqns 8 and 9 and graphically
in Figure 5, both singlet wave functions consist of two different configurations.
Since Ψx and Ψy are the two components of the degenerate, singlet (1E′)
electronic state in D3h TMM, Ψx and Ψy must, of course, have exactly the same
energies. However, as shown graphically in Figure 5, the MOs ψ2 and ψ3 are
occupied very differently in Ψx and Ψy. Therefore, I set out to convince myself
that Ψx and Ψy really do have the same energies.
259
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 5. Schematic depiction of the two components, Ψx and Ψy, of the 1E′ state
of TMM. Each component consists of two configurations.

When I worked through the math, it appeared that there actually is a difference
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

between the energies of Ψx and Ψy. This energy difference is equal to

where J23 is the Coulombic repulsion between one electron in ψ2 and one in ψ3;
J22 = J33 is the Coulombic repulsion between a pair of electrons in the same,
degenerate, non-bonding MO in TMM, and K23 is the exchange integral between
ψ2 and ψ3.
In order to use PPP theory to compute the Coulomb and exchange integrals
in eqn. 10, I wrote out the expressions for the non-bonding MOs, ψ2 and ψ3, in
TMM as the e″ combinations of the 2p-π AOs on the three peripheral carbons, ɸ1
- ɸ3.

and

I then used these MOs to compute the Coulomb and exchange integrals in eqns.
5 and 6 in terms of the Coulomb repulsion energies between two electrons in the
same 2p-π AO (γaa) and in 2p-π AOs on two different CH2 groups (γac). I found
that in TMM

Therefore, in eqn. 10

260
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 proving that E (Ψx) = E (Ψy) in TMM. Group theory told me that this had to be
true, but proving that it actually is true gave me the satisfaction of knowing that I
really did understand how to do PPP calculations correctly.
According to eqn. 7, the energy difference between the lowest triplet state
(3A2′) and the open-shell singlet state (Ψx) of TMM is 2K23. Since Ψx and Ψy,
have the same energy in D3h TMM, 2 K23 is the energy difference between 3A2′
and both components of the 1E′ state. Because 1E′ is the lowest singlet state of D3h
TMM, from eqn. 15 the triplet is predicted to be the ground state of TMM by (γaa
– γac)/3. Consequently, TMM is predicted to follow Hund’s rule (12, 13), which
is, of course, what Paul Dowd found experimentally (30).
I also did PPP calculations on CBD. I discovered that, due to the difference
between the three-fold symmetry of the nonbonding MOs in D3h TMM and the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

four-fold symmetry of these MOs in D4h CBD, these two diradicals differ in several
ways.
In the rectangular geometry of CBD in which C1-C2 and C3-C4 are shorter
bonds than C1-C4 and C2-C3, the MO,

is lower in energy than the MO,

Consequently, at such a rectangular geometry, the wave function, ΨS, for the lowest
singlet state can be written as

However, at the rectangular geometry where C1-C4 and C2-C3 are shorter C-
C bonds than C1-C2 and C3-C4, then ψ3 is lower in energy than ψ2. Consequently,
at such a geometry, the wave function for the lowest singlet state can be written as

At a square geometry, which is the transition structure for the interconversion

of the two rectangular geometries of CBD, ψ2 and ψ3 have the same energies;
and the lowest singlet wave function, ΨS is a linear combination of the two one-
configuration wave functions in equations 19 and 20.

The lowest singlet wave function for square CBD, ΨS in eqn. 21, has the
same form as Ψy in eqn. 9 for one of the two degenerate singlet wave functions
in TMM. However, group theory predicts that the lowest singlet state of square
CBD is non-degenerate. In fact, it is easy to use PPP theory to show that in square
CBD, the wave function

which is of the same form as Ψx in eqn. 8, is higher in energy than ΨS.

261
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 As in TMM, where the energy difference between Ψx and Ψy is given by eqn.
10, in CBD

However, using the CBD MOs in eqns 17 and 18, rather than the TMM MOs in
eqns 11 and 12 to evaluate J22, J33, J23, and K23, PPP theory finds that in square
CBD
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Therefore, according to PPP theory, in CBD

so that eqn. 23 becomes

As predicted by group theory, eqn. 27 shows that the energies of ΨS′ and ΨS are
not the same in square CBD.
In square CBD (as in TMM) the wave functions for the open-shell singlet,
ΨS′, and the triplet state ΨT are given by eqn. 2; and, according to eqn. 7, their
energies differ by 2K23,

Subtracting eqn. 28 from eqn. 27 shows that in CBD

Thus, in contrast to the case in TMM, where the triplet is calculated to be

2K23 lower in energy than both components of the lowest singlet state, in square
CBD the lowest singlet state and the triplet are predicted by PPP calculations to
have the same energies (35). This difference between TMM and CBD is shown
in Figure 6, in which the symmetries of the low-lying electronic states in each
of these two diradicals are given and the relative PPP energies of these states are
shown schematically.
In 2015 I am amazed that in 1965 I was apparently not interested in delving
into the physical reason why PPP calculations predict a large singlet-triplet energy
separation in TMM, but not in CBD. However, attempting to get some physical
insight into computational results is something that I only began to do around 1968,
inspired by the papers that Roald Hoffmann started publishing after he had joined
the Faculty at Cornell.
262
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Figure 6. Schematic depiction of the relative PPP energies of the low-lying

electronic states of TMM and CBD.

Roald had been a lab instructor in the first chemistry course that I had taken
in 1961 as a freshman at Harvard; and my roommate, Bob Joffe, had done
undergraduate research with both Roald and Lionel Salem. In 1965, I learned
from H. C. Longuet-Higgins that Roald was working with Bob Woodward on
what would eventually become the Woodward-Hoffmann rules (36). Therefore,
when Roald started publishing independent papers on organic chemistry in 1968,
I paid close attention to them; and they have had a huge influence on my approach
to theoretical chemistry throughout the remainder of my career.

Planar Allene and Its Similarity to CBD

As my time in Cambridge, UK was drawing to a close, I decided that,
before I returned to Cambridge, MA, I would like to use what I had learned
about performing PPP calculations, in order to make a prediction that might turn
out to be publishable. I thought it was possible that the lowest triplet state of
allene would prefer a planar equilibrium geometry to the non-planar equilibrium
geometry of the singlet; so I did PPP calculations to see if this was, in fact, the
case. The calculations predicted that triplet allene should, indeed, have a planar
equilibrium geometry.
What I had not realized when I began these calculations was that the planar
→ non-planar allene transformation is very similar to the transformation of
rectangular → square CBD. In both cases a geometry that has two bonding π
MOs is transformed into a geometry that has one bonding and two non-bonding
orbitals (NBMOs). Therefore, what I learned from doing PPP calculations on
planar allene, I was subsequently able to apply to understanding the PPP results
for CBD.
As shown in Figure 7, in planar allene the singly occupied MOs are a σ AO that
is confined to the central carbon and an allyl non-bonding π MO that is confined
to the two terminal carbons. Although in 1965 I was not thinking much about
the shapes of MOs, it was clear to me even then that the overlap between these
two NBMOs in planar allene must be very close to zero. Therefore, the exchange
263
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 integral between electrons in these two NBMOs must also be close to zero, so that
the energy difference between the lowest singlet and triplet states, in which each
of these MOs is singly occupied, must also be close to zero. This is what my PPP
calculations found for planar allene.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Figure 7. The two nonbonding (NB)MOs of planar allene are a σ AO on the

central carbon and an allyl π NBMO on the two terminal σcarbon atoms.

When I returned to Harvard as a first year graduate student in the fall of 1965,
I decided to work for E.J. Corey. He seemed quite interested in the results of my
PPP calculations on allene. In fact, he said, “We should do an experimental test of
your prediction, that triplet allene has a planar equilibrium geometry.”
However, before beginning to work in the lab on the photoracemization of an
optically active allene, I decided to write a manuscript about my computational
prediction, that triplet allene has a planar equilibrium geometry. When I had
completed a first draft, I sent it to Professor Longuet-Higgins and asked him if
he would like to have his name on the manuscript that I planned to submit for
publication. He declined. At the time I chose to believe that he was being gracious;
but now I think that he probably did not want his name on a paper that fell below
his very high standards.

Like Square CBD, Planar Allene Is Calculated To Violate Hund’s Rule

My paper on planar allene was my first publication on electronic structure
theory (37, 38). Probably the most interesting finding in it appeared in a footnote,
which read, “Carrying out complete configuration interaction to resolve the
degeneracy [between the lowest singlet and triplet states], one finds that the singlet
lies 0.05 eV below the triplet” (37, 39). In other words, my PPP configuration
interaction calculation predicted that Hund’s rule (12, 13) is violated in planar
allene.
Unfortunately, I did not believe my own computational result. In fact, I
devoted the rest of the footnote to explaining that, “Quite possibly, the triplet
lies lowest” (37). Nevertheless, as I subsequently discovered, my finding that in
planar allene the singlet state is calculated to be lower in energy than the triplet
state was not the first predicted violation of Hund’s rule in a diradical.
The first prediction of a violation of Hund’s rule in a diradical seems have
been made by Craig in 1950 (40). Interestingly, he found that, with inclusion of
264
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 configuration interaction (CI), square CBD has a singlet ground state. In 1968
Buenker and Peyerimhoff made the same prediction about square CBD, based on
the results of their ab initio CI calculations (41).
Thus, square CBD and planar allene have more in common than just a lowest
singlet and triplet state that have the same energies in PPP theory. In addition,
with inclusion of CI, the singlet is predicted to fall below the triplet state in both
diradicals, in violation of Hund’s rule.
It took me nearly ten years, but eventually I realized that the reason for the
predicted violation of Hund’s rule in square CBD is essentially the same as that in
planar allene. However, rather than asking the reader to wait for a description of
the next ten years of my career in chemistry, let me skip ahead from 1965 to 1975,
with a brief stop in 1969.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Why the Lowest Singlet and Triplet States Are Calculated To Have the Same
PPP Energies Not Only in Planar Allene but Also in Square CBD
It was not until 1969 that I first realized that the lowest singlet and triplet
states of CBD have the same energies in PPP theory for the same reason that the
lowest singlet and triplet states of planar allene do. In CBD, as in planar allene,
the degenerate MOs can be chosen so that they have no atoms in common (35).
Therefore, since the exchange integral, Kij in eqn. 6 involves the overlap of the
two degenerate MOs, ψi and ψj, if these MOs are confined to different regions of
space, Kij ≈ 0. Then, according to eqn. 7, if Kij ≈ 0, ΔEST ≈ 0.
As already noted, the nonbonding (NB)MOs in planar allene – the σ NBMO
on the central carbon and the allyl π NBMO, which is confined to the two terminal
carbons – are disjoint (i.e., they have no atoms in common). On the other hand,
the CBD NBMOs in eqns. 17 and 18 are clearly not disjoint. In fact, they span
exactly the same set of 2p-π AOs; and the exchange integral, K23, between them in
eqn. 25 is certainly not zero. Consequently, the energy separation of 2K23 between
the open-shell singlet, ΨS′, and the triplet ΨT is quite large in CBD.
This was confusing, until I realized that it is not the open-shell singlet, ΨS′
in eqn. 22, but the two-configuration singlet, ΨS in eqn. 21, that is calculated to
have the same PPP energy as the triplet state in square CBD (eqn. 29). Eqn. 21
shows that ΨS consists of two configurations, in each of which one of the NBMOs
is doubly occupied. One configuration places both of the nonbonding electrons in
ψ2; the other places both nonbonding electrons in ψ3.
Such a two-configuration wave function is hard to interpret physically.
However, it can be factored into

The sum and difference MOs on the right-hand side of eqn. 30 are called
Generalized Valence Bond (GVB) MOs (42). GVB MOs provide a useful physical
interpretation of two-configuration wave functions, because the wave function on
the right hand side of eqn. 30 is that for an open-shell singlet, in which each GVB
MO is occupied by one electron.
265
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Using eqns 17 and 18 for the non-bonding MOs of CBD, the GVB orbitals
for singlet CBD are

and

It should be noted that ψ2GVB for CBD in eqn. 31 has no 2p-π AOs in common
with ψ3GVB in eqn. 32. Therefore, like the two non-bonding MOs in planar allene,
ψ2GVB and ψ3GVB in CBD are disjoint. It is the disjoint nature of these singly
occupied MOs of planar allene and of square CBD that result in the calculated
PPP energies of the lowest singlet and triplet states to be the same in both of these
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

diradicals.
In 1969 I published a communication on why the disjoint nature of the
NBMOs of square CBD make this diradical different from cyclopentadienyl
cations and benzene dianions, which have singly-occupied MOs that are not
disjoint (35). In reading this communication I am struck by the fact that I opted
for mathematical elegance, rather than physical clarity (35). I used MOs with
complex coefficients in my proof that the PPP energies of the lowest singlet and
triplet of CBD are the same; and nowhere in this communication is there a single
drawing of an MO. It is interesting and sometimes a little embarrassing to read
the papers that one published early in one’s career and to see how much the way
one writes manuscripts has evolved with the passage of time.

Why Do CI Calculations Predict that Hund’s Rule Is Violated in Planar

Allene and in Square CBD?
I did not find the answer to this question until I started writing Chapter 9 of
my textbook, Modern Molecular Orbital Theory for Organic Chemists (43) (the
same title that Professor Andrew Streitwieser used for his book (31), except for the
addition of the word “Modern” to my title). Chapter 9 discussed calculations on
open-shell π systems in general and on the allyl radical in particular. I described
how a small configuration interaction (CI) calculation, involving the three possible
configurations that can be generated by exciting an electron from the bonding π
MO in allyl radical into the antibonding π MO, generates the negative spin density
at the central carbon of allyl that is seen experimentally (44).
Of course, planar allene has an allylic π system. The reader will also recall
that it is only with inclusion of CI, involving the same configurations that produce
the negative spin density in allyl radical, that planar allene is calculated to have
a singlet ground state (37, 39). Therefore, it is the negative spin density at the
central carbon of the allylic π system of planar allene that produces the predicted
violation of Hund’s rule in this diradical.
Including the negative spin density at the central carbon of the allylic π system,
the singlet and triplet states of planar allene can be depicted schematically as in
Figure 8. The component of the triplet state, in which the electrons in the σ and
π NBMOs both have α spin is shown on the left. It can be seen that the negative
266
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 spin density in the 2p-π AO at the central carbon in the triplet is β, opposite to the
α spin in the σ NBMO that is localized at this carbon.

Figure 8. Schematic depiction of one component of the triplet state (left) and of
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

the two configurations that comprise the open-shell singlet state of planar allene
(right). The negative spin density in the 2p-π AO of the central carbon is shown
in both electronic states. In the triplet state the negative spin in the 2p-π AO of
the central carbon is antiparallel to the spin in the σ AO on this carbon, However,
in the singlet these two spins in the AOs on the central carbon are parallel, so
that these two electrons cannot simultaneously appear in the regions of space
that these two AOs have in common.

Shown at the right of Figure 8 is the open-shell singlet state of planar allene.
Figure 8 shows that in both configurations of the open-shell singlet state the spin
of the electron in the σ NBMO is the same as that of the electron that gives rise to
the negative spin density in the 2p-π AO at the central carbon. Since electrons of
the same spin are prevented from appearing in the regions of space that the σ and
π AOs at the central carbon have in common, the Coulombic repulsion between
these two electrons is smaller in the open-shell singlet state than in the triplet state,
where the spins of these two electrons are antiparallel. Therefore, the origin of the
predicted violation of Hund’s rule in planar allene is easy to understand as an effect
that arises from the presence of negative spin density in the 2p-π AO at the central
carbon.
The predicted violation of Hund’s rule in square CBD has a similar origin
(45). When a CI calculation is performed, the triplet and lowest energy singlet
state each have two nonbonding π electrons that can affect the distribution of the
two electrons of opposite spin in the lowest π MO. Using the GVB orbitals for
square CBD in eqns. 31 and 32, each of these nonbonding π electrons is confined
to two carbons of the four membered ring.
As shown schematically in Figure 9, in the singlet state the asymmetric
distribution of spin in the NBMOs, ψ2 and ψ3, allows the α spin electron in the
lowest π MO, ψ, to become partially localized at the two carbons at which an α
spin electron is localized in one of the NBMOs. Similarly, the β spin electron in
the lowest MO can become partially localized at the two carbons at which a β
spin electron is localized in the other NBMO. Even partially confining electrons
of the same spin to two carbons and electrons of opposite spin to the other two
carbons reduces the Coulombic repulsion energy between the electrons in ψ and
those in ψ2 and ψ3 in the lowest singlet state of square CBD.

267
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Figure 9. Schematic depiction of dynamic spin polarization in the lowest singlet

state of square CBD. The spins of the electrons in the nonbonding π GVB MOs,
ψ2GVB and ψ3GVB, favor partial localization of the electrons in the bonding π
MO, ψ1, so that electrons of opposite spin in these three π MOs are found
predominantly at different sets of carbons. This partial localization makes the
Coulombic repulsion between the electrons in ψ2 and ψ3 and those in ψ1 lower in
the singlet state than in the triplet, which leads to the violation of Hund’s rule
that is predicted to occur in square CBD (40, 41, 45).

It should be noted once again that the wave function for an open-shell singlet
state consists of two configurations, which differ from each other by reversal of
the spins of the electrons in the singly occupied MOs. Consequently, the spin
polarization depicted in the left- and right-hand sides of Figure 9 is not static, but
dynamic. The name “dynamic spin polarization” has, in fact, been given to the
type of spin polarization shown in Figure 9 (46), and this name also describes the
type of spin polarization shown in Figure 8.
In the triplet state of square CBD electrons of the same spin occupy the
nonbonding MOs, ψ2 and ψ3. Consequently, there is a uniform distribution of the
same type of spin (e.g. α) at all four carbons. As a result, no polarization of the
spins of the electrons in the lowest π MO is possible; and this is the reason that
the triplet state of CBD is calculated to be higher in energy than the singlet state,
not only at rectangular geometries but also at the square geometry.

The Singlet and Triplet States of Trimethylenemethane

In 1966 I was in my first year of graduate school at Harvard when Paul

Dowd, who was a junior faculty member in the Chemistry Department, succeeded
in generating TMM and obtaining the EPR spectrum of its triplet state (27,
29). I was very interested in Paul’s experiments, and he took the time to talk to
me about them. We discussed why photolysis of 4-methylenepyrazoline leads
268
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 to the formation of triplet trimethylenemethane, which then dimerizes to form
1,4-dimethylene-cyclohexane; whereas, pyrolysis of 4-methylenepyrazoline leads
to the quantitative formation of methylenecyclopropane (28).
I conjectured that, although both of these two different modes of deazatizing
of 4-methylenepyrazoline lead to formation of TMM, the thermal reaction
produces the lowest singlet state. The singlet closes to methylenecyclopropane
faster than it undergoes intersystem crossing to the triplet ground state. With
Paul’s encouragement, I wrote a communication that contained this conjecture;
and I was very pleased when it was accepted for publication in Tetrahedron
Letters (47).
Personally, I think that the only reason that such a lightweight manuscript
was accepted was that it contained some correlation diagrams for conrotatory and
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

disrotatory deazatization of 4-methylenepyrazoline. 1966 was the heyday of the

Woodward-Hoffmann rules (36); and Longuet-Higgins (48), as well as Woodward
and Hoffmann (49) had used correlation diagrams to understand, respectively,
electrocyclic and cycloaddition reactions.

The Geometry of the Lowest Singlet State of TMM

By 1970, there were several lines of experimental evidence which indicated

that, although triplet TMM is planar (27, 29), singlet TMM prefers a non-planar
geometry with one CH2 group twisted out of conjugation (50–52). Indeed, in
1971 Dewar and Wasson had published the results of MINDO/2 calculations which
confirmed that this interpretation of the experimental results is correct (53).
I decided to perform PPP calculations of my own, to test this conclusion. My
calculations found that the MOs of D3h TMM give a Coulombic repulsion energy
for the planar singlet state that is higher by 3(γaa - γac)/8 than the Coulombic
repulsion energy for twisted TMM, which consists of an allyl radical plus an
electron localized in a non-conjugated 2p AO on the third CH2 group. Although
the Hückel energy of TMM is lower than that of allyl plus 2p by 2(√3 - √2)β =
0.6β, any reasonable values of γaa, γac, and β confirm the MINDO 2 conclusion,
that allyl plus 2p has a lower energy than planar singlet TMM.
In fact, eqns. 14 and 15 can be used to show that the Coulombic repulsion
energy between the two nonbonding electrons in the open-shell singlet state of
planar TMM is J23 + K23 = (γaa + 2γac)/3; whereas the Coulombic repulsion energy
between the two nonbonding electrons in twisted TMM (i.e., allyl + 2p) is just γac.
The difference of (γaa - γac)/3 is almost 90% of the total difference of 3(γaa - γac)/8
between the total Coulombic repulsion energies of planar singlet TMM and allyl
plus 2p.
The reason for the lower energy of the allyl plus 2p NBMOs is evident in
Figure 10. It shows that the NBMOs of allyl plus 2p in Figure 10b are disjoint;
whereas those of planar TMM in Figure 10a are not. The disjoint NBMOs of allyl
+ 2p in Figure 10b make the singlet wave function for twisted TMM much lower
in energy than the singlet wave function that employs the non-disjoint NBMOs for
planar TMM, which are shown graphically in Figure 10a.

269
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 10. NBMOs in (a) planar TMM and (b) in TMM with one CH2 group
twisted out of conjugation, giving the MOs for allyl radical and a 2p AO.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Singlet TMM versus Singlet CBD

Having published a paper on TMM in 1966 (47), of which I was less than
proud, I had not planned to publish the results of my PPP calculations on TMM;
and I had, instead, made them the subject of Problem 7.18 in my book on MO
theory (43). However, Lionel Salem changed my mind; and the manuscript on
TMM that I had not wanted to submit for publication wound up appearing as a
Communication in JACS (54).
As I have recounted elsewhere (32), in 1973 Lionel was a visiting Professor
at Harvard, where I was in the terminal year of being an Assistant Professor.
Lionel was one of my heroes at the time. He had not only been one of Christopher
Longuet-Higgins’ most successful graduate students; but also, in 1962, I had
attended, together with my roommate Bob Joffe, a most extraordinary party at
Lionel’s apartment in Cambridge.
In 1973 Lionel was interested in modeling the effects of electron repulsion on
the different possible geometries for sigmatropic reactions, a subject about which
he had just published a paper with Jerry Berson (55). Therefore, Lionel wanted
to compare planar singlet TMM with allyl + 2p as two models for the transition
structure in methylenecyclopropane rearrangement. In addition, he also wanted to
compare singlet CBD with allyl + 2p, as models for the transition structures, in,
respectively, a forbidden but concerted [1,3]sigmatropic shift (36) and a stepwise
reaction.
My PPP calculations showed that electron repulsion disfavors planar singlet
TMM, relative to allyl + 2p, by 3(γaa - γac)/8; but electron repulsion disfavors
singlet CBD, relative to allyl + 2p by only (γaa - γac)/8 – one third as much as
singlet TMM is disfavored, relative to allyl + 2p. As Lionel and I pointed out in
the conclusion to our JACS Communication (54), the difference between singlet
CBD and planar singlet TMM is that in singlet CBD the NBMOs can be chosen
so that they have no atoms in common (see Figure 9); whereas in planar singlet
TMM, the NBMOs do have atoms in common (see Figure 10a).
270
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Four years later, in 1977, I would use the words “disjoint” to describe the
NBMOs in CBD and “non-disjoint” to describe the NBMOs in planar TMM.
This difference between the NBMOs in these two diradicals would subsequently
provide a paradigm for making general predictions about the ground states of
diradicals (56).

The Wave Functions for Planar Singlet TMM

The rules for predicting the ground states of diradicals came about in a
collaboration with Professor Ernest R. Davidson. We began to collaborate
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

after I moved from Harvard to the University of Washington in 1973, and my

collaborations with Ernest on multiple projects began with our mutual interest in
the wave functions for the lowest singlet state of planar TMM.
Based on the Communication that I had published with Lionel Salem,
it seemed that their must be a very large energy difference between a planar
geometry and a twisted allyl + 2p geometry for singlet TMM. However, Bill
Doering, with whom I talked frequently at Harvard, told me that, based on the
results of his experiments (50), this could not be correct. His experiments showed
that, for singlet TMM, a twisted allyl + 2p geometry is preferred to a planar
geometry but only by a small amount of energy.
A paper published in 1974 by Yarkony and Schaefer (57) convinced me that
Bill Doering was right. They found that the energy of the wave function for planar
singlet TMM is much lower in energy if only C2v, rather than D3h symmetry,
is imposed upon its MOs. With the imposition of D3h symmetry the NBMOs
for planar TMM are those given in eqs. 11 and 12 and depicted in Figure 10a.
However, if only C2v symmetry is imposed on the MOs, then, even at a planar
geometry, the NBMOs for singlet TMM can be those for an allyl radical plus an
electron localized in a 2p AO. Fritz Schaefer was kind enough to send me the MOs
for the planar singlet TMM in C2v symmetry; and, as I had guessed, the MOs he
sent me were those for an allyl radical plus an electron localized in a 2p AO.
Until I saw Fritz’s C2v MOs for planar singlet TMM, I had believed that MOs
belonged to molecules. For example, I had thought that, at a D3h geometry, the
MOs for singlet and triplet TMM are the same. Fritz’s result taught me that the
MOs for different electronic states of a molecule, even at the same geometry, can
be very different (58).
This is true not only in TMM but also in other diradicals in which the singly
occupied MOs in the triplet have atoms in common (i.e., are non-disjoint). The
wave functions for the lowest singlet states of such diradicals are comprised of
MOs that are more disjoint than their triplet counterparts. In fact, a hallmark
of diradicals with NBMOs that are non-disjoint is that the MOs for different
electronic states are different (59).
However, there is a problem with singlet TMM having C2v MOs at a D3h
geometry, because at a D3h geometry, the MOs for any state must have D3h
symmetry. Nevertheless, I was able to show that, by mixing different D3h
electronic configurations that contain e″ MOs, 1E′ wave functions that have
271
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 only C2v symmetry can be obtained. I showed that this is the case by carrying
out both PPP CI calculations (62) and ab initio CI calculations (63) on planar
singlet TMM. The latter calculations were performed in collaboration with Ernest
Davidson, and that was the first of many projects on which we collaborated.
The CI calculations with Ernest on TMM eventually spawned a review on
real and artifactual symmetry breaking that Ernest and I coauthored (64). We
also carried out ab initio calculations on the potential surfaces for singlet TMM
(65) and for singlet CBD (66). However, the most important paper that Ernest
and I coauthored was probably a paper on the prediction of the ground states of
diradicals, according to whether they do or do not have disjoint NBMOs (56).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

A Simple Way To Predict the Ground State of a Diradical

As already discussed, eqn. 4 shows that in an antisymmetrized wave function
the probability of two electrons of the same spin simultaneously appearing in the
same region of space is zero. Thus, if the two singly occupied MOs in a diradical
have atoms in common, the triplet state will have a lower Coulombic repulsion
energy than the open-shell singlet state that has the same orbital occupancy.
However, this does not necessarily mean that the triplet is the ground state
of such a diradical. In addition to the open-shell singlet state, which has the
same orbital occupancy as the triplet, there is another low-energy singlet state that
consists of two configurations. As shown in eqn. 30, this two-configuration wave
function can be factored into a wave function in which orbitals, that are the sum
and difference of the original MOs, are each occupied by one electron. As already
noted, the sum and difference orbitals are called the GVB orbitals.
The original set of MOs may have atoms in common, as do the CBD MOs in
eqns 17 and 18; but the GVB MOs formed from them may be disjoint. Indeed, the
GVB MOs for CBD in eqns 31 and 32 have no atoms in common. Therefore,
before deciding whether the degenerate MOs for a diradical can or cannot be
confined to different sets of atoms, one has to check not only the orginal pair of
degenerate MOs but also the GVB orbitals derived from them.
If in a diradical there is no combination of the degenerate MOs that is disjoint,
then the ground state of the diradical is the triplet. One example of such a diradical
is TMM; but there are many other diradicals that also have triplet ground states,
because all of the possible choices of the degenerate MOs have atoms in common.
Another of these diradicals, meta-benzoquinodimethane (MBQDM) is discussed
later in this chapter.
However, if in a diradical, a pair of degenerate MOs can be found that are
disjoint (i.e., have no atoms in common), then electrons that singly occupy these
two MOs will not appear in the same region of space, no matter whether the spins
of the two electrons are parallel or anti-parallel. Consequently, in such a diradical,
to a first approximation, the singlet and triplet states will have the same energies.
Under these circumstances, it is possible for the effect of the non-uniform
distribution of spin in the singlet state to make the singlet the ground state. As
discussed above, dynamic spin polarization (46) makes a singlet the ground state
of both planar allene (37, 39) and square CBD (40, 41, 45). Since Hund’s rule
272
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 predicts that the triplet state of a diradical is always the ground state (12, 13),
planar allene and square CBD are both predicted to violate Hunds rule.

Determining Whether the Degenerate Nonbonding MOs of a Diradical Are

Disjoint or Non-Disjoint

In order to determine whether the singly occupied MOs of a diradical are

disjoint or non-disjoint, it might seem that one would have to obtain these MOs
by performing DFT or ab initio electronic structure calculations. However, there
is one class of diradicals for which it is trivial to obtain the singly-occupied MOs,
by performing a back-of-the-envelope calculation (25). Members of this class are
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

called alternant hydrocarbon (AH) diradicals.

In an AH it is possible to separate the carbon atoms into two sets – traditionally
called the starred set and the unstarred set, such that no two atoms of the same set
are nearest neighbors (43, 67). Figures 11 and 12 show that the diradicals – TMM,
meta-benzoquinodimethane (MBQDM), CBD, and tetramethyleneethane (TME)
are all AHs.

Figure 11. NBMOs of TMM and of meta-benzoquinodimethane (MBQDM),

derived from use of the zero-sum rule (25). In both of these diradicals the number
of starred atoms (n*) minus the number of unstarred atoms (n) is equal to two.
Consequently, the NBMOs are both confined to just the starred set of carbon
atoms and have carbons in common. Therefore a triplet ground state is predicted
for both of these diradicals (56).

The nonbonding (NB)MOs of an AH diradical can be found from Longuet-

Higgins’ zero sum rule (25). It states that in both NBMOs the coefficients of the
starred atoms must sum to zero about each unstarred atom; and the coefficients of
the unstarred atoms are either zero or sum to zero about each starred atom.
273
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 11 shows how the NBMOs of TMM and of MBQDM can be found
from the zero-sum rule. In order for the coefficients of the unstarred atoms to sum
to zero about each starred atom in these two AH diradicals, the coefficients of the
unstarred atoms must all be zero. This is generally the case when the number of
starred exceeds the number of unstarred atom by 2 (56).
To find the coefficients of the starred atoms, one starts by assigning a
coefficient of a to one starred atom and then determines how the zero-sum rule
can be satisfied by assigning ±a, multiples thereof, or zero to the other starred
atoms. The magnitude of a can be determined from the requirement that the
NBMO be normalized. Therefore a must be equal to 1, divided by the square root
of the sum of the squares of all of the coefficients of a in the MO.
If, unlike the case in TMM and MBQDM, the number of starred and unstarred
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

carbons is the same in an AH diradical, then one NBMO can be chosen to be

confined to the starred set of carbon atoms, while the other NMBO can be chosen
to be confined to the unstarred set of carbons (56). This is illustrated in Figure 12
for CBD and TME.

Figure 12. NBMOs of square CBD and tetramethyleneethane (TME), derived

from use of the zero-sum rule (25). In both of these diradicals the number of
starred atoms (n*) and unstarred atoms (n) is the same; so one NBMO is confined
to the starred set of atoms and the other to the unstarred set (56). Therefore the
NBMOs are disjoint; and, with inclusion of dynamic spin ploarization (45, 46), a
singlet ground state is predicted for both of these diradicals (56).

The NBMOs in Fig. 12 do not have the full molecular symmetry. For
example, the CBD NBMOs in Fig. 12 are the GVB MOs in eqns. 31 and 32,
which are linear combinations of more delocalized sets of CBD NBMOs in eqns
17 and 18. Taking the sum and difference of the MOs in Figure 12 gives the
more delocalized set of NBMOs for CBD and for TME, which do have the full
molecular symmetry.
Although using MOs that have the full molecular symmetry is advantageous
for actually doing calculations, the orbitals in Figure 12 have the virtue of showing
274
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 that the NBMOs of CBD and TME can be chosen to be disjoint. Therefore, based
on the disjoint topology of their NBMOs, it can be predicted that dynamic spin
polarization is likely to lead to a singlet ground state for CBD and TME.
Although it is easy to use the zero-sum rule to find the singly-occupied,
nonbonding (NB)MOs in AH diradicals, Ernest Davidson proved that it is
possible to discover whether or not these NBMOs are disjoint (i.e. have no atoms
in common) or non-disjoint, just by counting the numbers of starred and unstarred
atoms (56). Ernest proved that, if the number of starred carbons (n*) exceeds the
number of unstarred carbons (n) by two in an AH diradical, the two NBMOs are
both confined to the starred set of atoms. Hence, they are, in general, non-disjoint.
Consequently, n* - n = 2 in an AH diradical predicts non-disjoint NBMOs and,
hence, a triplet ground state.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Ernest also showed that if n* = n, then one NBMO is confined to the starred set
of atoms, and the other NBMO is confined to the unstarred set of atoms. Therefore,
the NBMOs are disjoint. Consequently, n* - n = 0 predicts a singlet ground state
for a diradical.

Ovchinnikov’s Equation

Another qualitative method for predicting the ground state of a diradical is due
to Ovchinnikov (68). He used valence-bond (VB) theory to derive the formula in
equation 33 for the spin quantum number, S, of the ground state of any AH.

This method is also based on the difference, n* - n, between the number of

starred and unstarred carbons in an AH diradical. However, instead of using n* - n
to determine whether the NBMOs are disjoint or non-disjoint, in the Ovchinnikov
method n* - n provides information about π bonding in a molecule, using VB
theory. If S = 1 in eqn. 33, a triplet ground state is predicted; but, if S = 0, the
ground state is predicted to be a singlet.
For example, in TMM, |n* - n| = 2, so eqn. 33 gives S = 1; and the ground
state is predicted to be a triplet. In contrast, in CBD n* = n; so the ground state is
predicted to have S = 0 and to be a singlet. It can be seen that the predictions made
from Ovchinnikov’s formula, which is based on VB theory, are the same as those
made on the basis of the disjoint/non-disjoint topologies of the singly-occupied π
MOs in a diradical.
Why do these two different approaches lead to the same set of predictions?
The MO approach to predicting ground state of a diradical starts by finding the
orbitals that maximize bonding and then determining whether or not the singly
occupied MOs in the diradical are disjoint or non-disjoint. In contrast, VB theory
starts by assuming that an AH has one electron in the 2p-π MO of each carbon and
then determining how many of these π electrons are involved in forming bonds. It
is gratifying that either starting point leads to the same prediction about the ground
state of an AH diradical.
However, the classification of the MOs in diradicals as being disjoint or non-
disjoint does have the advantage over Ovchinnikov’s equation of being applicable
275
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 to antiaromatic ions, containing 4n electrons in rings of 4n ± 1 or 4n ± 2 carbons
(35). In addition, although, neither method, by itself, provides any information
about the size of ΔEST in a diradical, using the zero-sum rule to actually find the
Hückel NBMOs and then inspecting the degree to which they overlap can provide
some qualitative information.
Inspecting the Hückel NBMOs of a non-disjoint diradical has the additional
advantage of providing a good indication of the ways in which these delocalized
NBMOs for the triplet state are likely to become localized in the lowest singlet
states, in order to reduce the Coulombic repulsion between the nonbonding
electrons. For example, from inspection of the NBMOs of triplet TMM in Figure
10a, it is easy to deduce that the Coulombic repulsion in the corresponding
singlet state will be reduced by localization of the ey″ MO on the unique carbon
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

atom, where the ex″ MO has a node. This localization should make twisting
of this carbon out of conjugation, as shown in Figure 10b, very likely in the
lowest singlet state of TMM; and, as already discussed, this non-planar geometry
actually is favored by the lowest singlet state of TMM (50–54).

Computing Not Just the Sign but Also the Size of ΔEST
The previous section discussed how the ground state of an AH diradical can
be predicted, just by counting the number of starred and unstarred carbons. Of
course, one wants to do more than just predict the sign of the singlet-triplet energy
difference (ΔEST > 0 for a triplet, and ΔEST < 0 for a singlet); one also wants to
predict the size of ΔEST. For this purpose, the software to perform high-quality
electronic structure calculations is now readily available.
However, calculations, based on one-configuration wave functions, will
usually prove inadequate for calculations on diradicals. As already noted in the
discussion of eqns. 1 and 2, a one-configuration wave function for an open-shell
“singlet” diradical furnishes MOs and energies that are 1:1 averages of those for
the actual open-shell singlet state with S2 = 0 and the triplet state with S2 = S(S +
1) = 2. Such a one-configuration “singlet” wave function has S2 ≈ 1.
From the energy and calculated S2 value of a one-configuration, open-shell,
“singlet” state it is possible to estimate the energy of the actual two-configuration
(S2 = 0) singlet state, using the energy of the triplet state, calculated at the same
geometry (69). However, the MOs on which the energy of a one-configuration,
“singlet” is based will, unfortunately, be an average of the MOs for the triplet and
for the two-configuration S2 = 0 open-shell singlet state. Therefore, the MOs for
the one-configuration, S2 ≈ 1, “singlet” state will, in general, not be the optimal
MOs for the two-configuration, (S2 = 0) singlet state.
This can be a very serious problem when the triplet NBMOs are non-disjoint,
as is the case in TMM. As discussed above, the allyl + 2p MOs, which are optimal
for the open-shell (S2 = 0) singlet state of planar TMM are very different from the
MOs for the triplet. Therefore, an average of these two sets of MOs will not be
very useful for accurately computing the energy of the S2 = 0 singlet state.
Coupled-cluster calculations with single and double excitations and a
perturbative correction for triplet excitations [CCDSD(T)] are currently the
276
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 “gold standard” of computational chemistry (70). However, garden variety
CCSD(T) calculations are based on one-configuration wave functions for singlet
states. Therefore, although CCSD(T) calculations may provide a very accurate
value for ΔEST, when the singlet can be well-represented by just one reference
configuration, the CCSD(T) value of ΔEST will, in general, not be reliable if
the singlet is open-shell and thus requires a two-configuration, reference wave
function.
When I began doing ab initio calculations on diradiucals, I started by
performing configuration interaction (CI) calculations. With small basis sets,
full π CI calculations were possible, and it was even possible to include some
single σ excitations. However, as faster computers made calculations with larger
basis sets practical, I began doing complete active space (CAS)SCF calculations.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

In a CASSCF calculation, not only are the weights of the most important
configurations in a limited CI wave function determined, but the MOs in those
configurations are also optimized.
However, CASSCF calculations ignore the effects of dynamic correlation
– how the electrons outside of the active space respond dynamically to all of
the electrons that are included in the active space. I learned through experience
that the effects of dynamic electron correlation can be quite important (71).
Therefore, about 20 years ago my group started doing CASPT2 calculations
(72). CASPT2 uses second-order perturbation theory (PT2) to provide dynamic
electron correlation for a CASSSCF wave function; and for the past 20 years, my
research group has mostly used the MOLCAS suite of programs (most recently
MOLCAS 7) (73) to perform CASPT2 calculations on diradicals.
My own interest in calculations has always been largely directed toward
using them to make new, experimentally testable predictions. Therefore, I try to
publish the results of my group’s calculations, before the experiments to test them
have been performed. Consequently, in the subsequent sections of this chapter on
the values of ΔEST in different diradicals, I have mostly chosen to reference the
calculations that were published by my research group, before the experimental
values of ΔEST were measured.
I owe an apology to those computational chemists who have published the
results of their calculations of ΔEST in different diradicals, after the experimental
values of ΔEST were measured, for not discussing their calculations in this chapter.
Comparing calculated values with known experimental values is important
for benchmarking different computational methods; but predicting, rather than
benchmarking, has almost always been the goal of the calculations carried out by
my research group.

Measurements of ΔEST in Diradicals

Measuring the Sign and Size of ΔEST with Curie Plots
If the triplet is the ground state of a diradical, a Curie plot of its EPR signal
intensity versus 1/T should give a straight line (74). It was through the linearity of
Curie plots that a triplet ground state was established for Schlenck’s hydrocarbon
(75, 76), the parent MBQDM (77), and TMM (30).
277
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 There are two different circumstances under which a curved Curie plot
will be observed. If a plot of signal intensity curves upward with decreasing
temperature, it means that more molecules in the triplet state are being populated
as the temperature is reduced. This implies that the triplet is the ground state, but
that there is a thermally accessible singlet state which is being depopulated as the
temperature is reduced.
On the other hand, if the Curie plot curves downward, it means that the triplet
is being depopulated as the temperature is lowered, The triplet is then a thermally
accessible excited state, and the ground state must be a singlet.
A linear Curie plot will be obtained if a triplet ground state is well below the
lowest singlet excited state in energy; but if the triplet state and a singlet state have
exactly the same enthalpy, then a linear Curie plot will also be obtained. In other
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

words, a Curie plot cannot distinguish between ΔEST = 0 and ΔEST >> 0. This
ambiguity actually makes it incorrect to claim, unequivocally, that a linear Curie
plot establishes that a molecule has a triplet ground state.
A curved Curie plot has the virtue of establishing whether the triplet is slightly
above or slightly below the lowest singlet state. However, as emphasized by Platz
and by Berson, curvature will only be detectable if the triplet and lowest singlet
state are very close in energy (74). Particularly if the triplet is the ground state, the
presence of a low-lying singlet state will be very hard to detect.
Making predictions is a less than satisfying activity, unless the predictions
can, at least in principle, be tested experimentally. The severe limitations of Curie
plots in providing accurate values of ΔEST in diradicals sometimes made me feel
that I was wasting my time in calculating ΔEST values, since it seemed that none
of them would ever be measured.

Measuring the Sign and Size of ΔEST by Negative Ion Photoelectron

Spectroscopy (NIPES)

This feeling of futility vanished entirely with the publication of the negative
ion photoelectron (NIPE) spectrum of the radical anion of TMM by Wenthold,
Hu, Squires, and Lineberger in 1996 (78). It then became clear that, at least in
principle, NIPES could provide a very accurate experimental value of ΔEST in any
diradical on which I performed calculations (4).
As shown schematically in Figure 13, in a NIPES experiment a beam of
radical anion molecules (D•-) is crossed by a laser, whose energy is sufficient to
ionize any of the electrons in the highest energy MOs of the radical anion. The
kinetic energies of the photoelectrons, ejected by absorption of the laser light by
the molecules of the radical anion, are measured. From conservation of energy,
the energy of each electronic state of the neutral diradical, minus the energy of the
radical anion, must be equal to the energy of the laser, minus the kinetic energy of
the electron that is ejected.

278
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Figure 13. Schematic depiction of a NIPES experiment in which a laser detaches

a photoelectron from a radical anion, D•–, forming three different electronic
configurations of the neutral diradical. Each configuration is accompanied by
a photoelectron with a different kinetic energy. Configuration C is part of both
the Sz = 0 component of the triplet and the open-shell singlet state. Formation
of the triplet state produces a photoelectron with the same kinetic energy (KEA)
as the photoelectron produced by formation of state A, the Sz = 1 component
of the triplet. However, formation of the open-shell singlet state produces a
photoelectron with a different kinetic energy (KEC).

Of course the energy of the laser is the same for each of the electronic states
that is formed in a NIPES experiment. Therefore, the relative energies of the
electronic states of a neutral diradical are given by the differences between the
kinetic energies of the electrons that are ejected in forming each of these states
from the radical anion. As discussed below, it is usually possible to identify each
of the electronic states of a diradical, whose formation accompanies the ejection
of a photoelectron with a particular kinetic energy.

279
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 As shown schematically in Figure 13, a radical anion can give rise to three,
low-lying, electronic configurations of the neutral diradical, formed from the
radical anion. Configuration A is the Sz = 1 component of the triplet state.
Configuration B is a singlet state in which both electrons occupy the same MO.
In configuration C, electrons with opposite spins are found, one each, in each of
the two MOs.
As discussed in connection with eqns.1 and 2, configuration C is not a proper
spin state. It is, instead a 1:1 mixture of the Sz = 0 component of the triplet
state and the open-shell singlet state. Formation of the Sz = 0 component of the
triplet produces a photoelectron with the same kinetic energy as formation of the
Sz = 1 component of the triplet; whereas formation of the open-shell singlet state
produces a photoelectron with a different kinetic energy.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

In the absence of a strong magnetic field, radical anions in which the unpaired
electron has spin up, (i.e., radical anions with Sz = ½), should be accompanied by
an equal number of radical anions in which the unpaired electron has spin down
(i.e., radical anions with Sz = -½). Loss of an electron from an Sz = -½ radical
anion again gives rise to three configurations. In this case they are: the Sz = -1
component of the triplet state, closed-shell configuration B again, and a variant of
configuration C, with the spins of the electrons in the two singly occupied MOs
reversed
Starting with a radical anion in which the unpaired electron is found in only
one MO, there is one way to form each of the three components of the triplet state,
two ways to form the singlet state in which one MO is doubly occupied, and one
way to form the open-shell singlet state. Thus, on a purely statistical basis, the
peak in a NIPE spectrum for formation of the triplet should be 50% more intense
than the peak for formation of the closed-shell singlet state and three times more
intense than the peak for the formation of the open-shell singlet state.
Differences in Franck-Condon factors (79) (vide infra) can cause the relative
intensities of the peaks in a NIPE spectrum to deviate from the expectations based
on the above statistics for formation of the triplet and the two low-lying singlet
states of a diradical. However, it generally seems to be the case that the peaks for
formation of the triplet are the most intense peaks in a NIPE spectrum (4).
On this basis, in the NIPE spectrum of (CO)5•– in Figure 14 (80), it is possible
to assign the group of peaks marked A to the triplet state and the group of lines
marked X to the lowest singlet state. The onsets of these two groups of peaks
are at 3.830 ± 0.005 eV (X), and 4.680 ± 0.010 eV (A), giving the singlet-triplet
energy difference in (CO)5 as ΔEST = -0.85 eV = -19.6 kcal/mol. The negative
sign indicates that the singlet is lower in energy than the triplet.

Assignment of the Peaks in a NIPE Spectrum by Vibrational Progressions

and by Calculations of Franck-Condon Factors

The NIPE spectrum in Figure 14 obviously does not consist of one peak for
formation of the singlet state and one peak for formation of the triplet state, but,
instead, of groups of peaks for formation of each state. The multiple peaks are
280
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 the result of vibrational progressions in each state. These vibrational progressions
are seen because the geometries of both the singlet and triplet states of the neutral
molecule are different from the geometry of the radical anion, from which the
singlet and triplet states of the neutral are formed.
Intensities of electronic transitions depend on the squares of the overlaps of
the vibrational wave functions for the ground and excited states. The squares of
these overlaps are called the Franck-Condon factors (79). If there is no difference
between the geometry of the ground and excited state, then, as shown in Figure
15a, a single peak is observed, because the lowest vibrational wave function of the
ground state has a non-zero overlap with only the lowest vibrational wave function
of the excited state.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Figure 14. NIPE spectrum of (CO)5•- (80). The greater intensity of band A
than of band X indicates that band A should be assigned to formation of the
triplet state of (CO)5. Reprinted with permission from Bao, X.; Hrovat, D. A.;
Borden, W. T.; Wang, X.-B. J. Am. Chem. Soc. 2013, 135, 4291. Copyright 2013
American Chemical Society.

281
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Figure 15. Vibrational wave functions with non-zero overlaps when (a) the
geometries of the ground excited states are the same and (b) the geometries are
different in a particular coordinate. In (a) only the (0,0) peak is seen in the
electronic excitation spectrum; but, in (b) several peaks – for example (0,0),
(0,1), and (0,2) – may be observed. The frequency of the vibrational progression
seen in (b) is the frequency of the vibration in the excited state that affects the
internal coordinates along which the ground and excited state geometries differ.

However, as shown in Figure 15b, if there is a geometry change upon

excitation, the lowest vibrational wave function of the ground state may have
non-zero overlaps with the wave functions for several vibrational levels of the
excited state. As a result, several vibrational levels of the excited state may be
populated; and a vibrational progression will then be seen. The peaks in this
progression are separated by the frequency of the vibration in the excited state
that affects the internal coordinates along which the geometries of the ground and
the excited state differ.
In NIPES the ground state is the radical anion and the excited states are
the electronic states of the neutral molecule that is formed by ejection of a
photoelectron from the radical anion. Therefore, in a NIPE spectrum vibrational
progessions are seen when the geometries of the radical anion and the neutral
282
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 species formed from it differ. The frequencies in these progressions are those in
the neutral species for vibrations that affect the internal coordinates along which
the geometries of the radical anion and the neutral differ.
Consequently, from calculations of the optimized geometries of a radical
anion and those of different low-lying electronic states of the neutral molecule
formed from it, one can qualitatively predict what types of vibrations are likely to
appear in the NIPE spectrum of the radical anion. From vibrational analyses on the
electronic states of the neutral, one can also predict what the frequencies of these
vibrations are likely to be. Finally, one can use a program, such as ezSpectrum
(81), to compute the Franck-Condon factors for formation of different electronic
states of the neutral and thus provide a quantitatively reliable simulation of the
NIPE spectrum. From comparison of the observed and simulated NIPE spectra,
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

all of the peaks in the experimental spectrum can usually be assigned. Some
examples will be provided in the sections that follow.

The NIPE Spectra of Some Radical Anions

The NIPE Spectrum of TMM•–

TMM is the simplest and the best-studied of the Non-Kekulé, AH diradicals

(82). As shown in Figure 11, n* - n = 2 in TMM; so the NBMOs of TMM are
non-disjoint. As can be seen in both Figures 10 and 11, the D3h NBMOs of TMM
do, in fact, have two carbon atoms in common. Consequently, TMM should have
a triplet ground state.
The linear Curie plot obtained for TMM (30) is, of course consistent with this
prediction. Unfortunately, the Curie plot provides no information about the size
of ΔEST in TMM. However, this information is available from the NIPE spectrum
of the radical anion of TMM (TMM•–) (78).
TMM•– was generated by treatment of 1,3-bis(trimethylsilyl)-2-
methylenepropane with a mixture of F2 and F– in the gas-phase (the Squires’
reaction) (83). As expected, the lowest energy peak in the NIPE spectrum of
TMM•– is for formation of the 3A2′ state (78). This peak is followed by a second
peak for formation of a singlet state of TMM. From the energy separation
between these peaks, ΔEST = 16.1 ± 0.2 kcal mol-1 in TMM.
The lowest energy singlet state of TMM has one CH2 group twisted out of
conjugation, as shown in Figure 10b. However, the Franck-Condon factors for
formation of this twisted geometry of neutral TMM from the planar geometry
of the TMM•– radical anion are zero. Therefore, the singlet state that is formed
by photo-detachment of an electron from TMM•– is not the lowest singlet
state of TMM, but the lowest energy singlet state that can be accessed by
photo-detachment of an electron from the planar radical anion.
The planar singlet state of lowest energy in TMM is 1A1, which is depicted in
Figure 16 (65). The value of ΔEST = 16.1 ± 0.2 kcal/mol from the NIPE spectrum
of TMM•– presumably corresponds to the energy difference between the 1A1 and
3A2′ states of planar TMM (78).

283
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 16. Three electronic states of planar TMM. 3A2′ is the ground state, with
1A1 predicted to be slightly lower in energy than 1B2. The three equivalent 1A1

structures are the minima on the Jahn-Teller potential surface for pseudorotation
of the lowest singlet state of planar TMM around the 1E′ cusp that exists at
D3h geometries, and the three equivalent 1B2 states are the transition structures
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

that connect the minima (65).

An early ab initio calculation on TMM predicted ΔEST = 18.2 kcal/mol as

the energy difference between the 1A1 and 3A2′ states of planar TMM (84). This
calculated value is about 13% higher than the experimental value of ΔEST = 16.1 ±
0.2 kcal/mol (78). More recent calculations, performed at higher levels of theory
(85, 86), are in slightly better agreement with the experimental value. However,
the results of these calculations were published after the experimental value of
ΔEST = 16.1 ± 0.2 kcal/mol had been measured. In contrast, the calculated value
of ΔEST = 18.2 kcal/mol (84) was truly a prediction, since it was published 14 years
before the NIPE spectrum of TMM•– had been obtained (78).

The NIPE Spectrum of MBQDM•–

Like TMM, MBQDM is a non-Kekulé AH diradical (82). The Hückel

NBMOs of MBQDM, 3b1 and 2a2, are shown in Figure 17. As demonstrated in
Figure 11, they can easily be found using the zero-sum rule (25). However, since
n* - n = 2 in MBQDM, the conclusion that the NBMOs are non-disjoint can be
made without actually finding them (56).

Figure 17. Schematic depiction of the Hückel NBMOs of MBQDM.

284
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Because the NBMOs of MBQDM are non-disjoint, it is easy to predict that
the ground state of MBQDM is a triplet. Indeed, as already noted, the triplet EPR
signal of MBQDM gives a linear Curie plot, strongly suggesting that MBQDM
does, in fact, have a triplet ground state (77). This conclusion is confirmed by the
NIPE spectrum (87), which not only provides the value of ΔEST = 9.6 kcal/mol,
but also provides the relative energies of both of the low-lying singlet states of
MBQDM.
The radical anion of MBQDM was generated by allowing m-xylene to react
with O•– in the gas-phase. The NIPE spectrum shown in Figure 18 was obtained
(87). As expected, the lowest energy peak is for formation of the 3B2 state of
neutral MBQDM, followed by a peak that is assigned to the 1A1 state and then by
another that is assigned to formation of the open-shell 1B2 state. What is the nature
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

of these two singlet states, and what determines their relative energies?

Figure 18. The NIPE spectrum of MBQDM•-. Reprinted with permission from
Wenthold, P. G.; Kim, J. B. Lineberger, W. C. J. Am. Chem. Soc., 1997, 119,
1354. Copyright 1997 American Chemical Society.

The non-disjoint nature of the Hückel NBMOs of MBQDM in Figure 17

means that although these NBMOs may be optimal for the triplet, they will not
be optimal for either singlet state. Nevertheless, from the NBMOs for the triplet,
it is possible to guess what the NBMOs will be like in both low-lying singlet states.
Hence, the nature of the bonding in each singlet state can be qualitatively predicted.
For example, in the open-shell 1B2 state, the electrons of opposite spin each
singly occupy the 3b1 and 2a2 π MOs. These opposite-spin electrons minimize
their Coulombic repulsion by tending to localize at different carbon atoms. This
can be accomplished by localization of the 2a2 NBMO at just two ring carbons and
partial localization of the 3b1 NBMO at a third ring carbon. The bonding in the
1B2 state of MBQDM is depicted schematically in Figure 19.

285
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 19. Schematic depiction of the bonding in the three lowest electronic
states of MBQDM (X = CH2) and in MBQ (X = O). In order to avoid the high
Coulombic repulsion that is associated with two electrons of opposite spin
appearing in the same region of space, the nonbonding electrons are more
localized in the singlet states, 1B2 and 1A1, than in the triplet state, 3B2, of
MBQDM and of MBQ.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

The way that the MBQDM NBMOs in Figure 17 localize in the 1A1 state is
a little harder to understand, because the 1A1 state has a two-configuration wave
function of the form,

The two GVB orbitals derived from it are, therefore, the sum and difference of the
3b1 and 2a2 NBMOs in Figure 17.
It seems paradoxical; but, in order for the GVB orbitals to be as disjoint as
possible, the two NBMOs in Figure 17 need to both localize at the same set of
carbons, specifically, at the carbons of the exocyclic methylene groups. Then the
two GVB MOs, which are the sum and difference of the 3b1 and 2a2 NBMOs,
will each be localized at a different CH2 group, thus minimizing the Coulombic
repulsion between this pair of electrons. The bonding in the 1A1 state of MBQDM
is depicted in Figure 19.
The depictions in Figure 19 of the bonding in the 1B2 and 1A1 states of
MBQDM, as well as the optimized bond lengths in these two states (7, 88),
indicate that 1A1 preserves the aromaticity of the benzene ring in MBQDM,
whereas the 1B2 state destroys it. It is for this reason that, in the NIPE spectrum
of MBQDM•– in Figure 18 (87), the 1A1 state is found to lie about 12.5 kcal/mol
below the 1B2 state, with the 3B2 state 9.6 kcal/mol below 1A1. Calculations
performed both before (7) and after (88) the NIPE spectrum of MBQDM•- was
published are in good agreement with these relative energies of the 3B2, 1A1, and
1B2 states of MBQDM.

The Effect of the Substitution of O for CH2 on ΔEST in Going from TMM
to OXA

Substitution of O for one CH2 group in TMM gives the oxallyl diradical
(OXA). The results of calculations on ΔEST in OXA were published (6) nearly
286
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 20 years before the NIPE spectrum of the OXA•– radical anion was obtained (8,
9). These calculations predicted ΔEST ≈ 0 (6).
Two decades later, more sophisticated calculations found the singlet actually
to be the ground state of OXA. One set of calculations found ΔEST = -1.3 kcal/mol
(8, 9), and the other ΔEST = -1.5 kcal/mol (89). Perhaps fortuitously, ΔEST = -1.3
kcal/mol is the same as the value obtained experimentally from Lineberger’s NIPE
spectrum of OXA•– (8, 9).
Based on Lineberger’s experimental values of ΔEST = 16.2 kcal/mol in TMM
(74) and -1.3 kcal/mol in OXA (8, 9), the substitution of the O in OXA for one
CH2 group in TMM stabilizes the 1A1 state, relative to the 3B2 state, by 17.5
kcal/mol. At least part of this large, selective, stabilization of the singlet can be
ascribed to the greater electronegativity of oxygen, relative to carbon. The greater
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

electronegativity of oxygen lowers the energy of the 2b1 π NBMO of OXA (ey″ in
TMM in Figure 10a), which has a large coefficient on oxygen, relative to the a2 π
MO (ex″ in TMM in Figure 10a), which has a node at the oxygen atom.
Since the 2b1 and a2 MOs do not have the same energies in OXA, the form of
the two-configuration wave function in eqn. 30 must be modified, so that the two
configurations are allowed to have different weights. The form of a more generally
useful two-configuration wave function is

The ratio, c1/c2, as well as the MOs, ψ1 and ψ2, can be optimized computationally;
and the magnitudes of c1 and c2 in eqn. 35 can be chosen, so that the wave function
is normalized.
In OXA use of a two-configuration wave function, like that in eqn. 35, with
ψ1 = ψ(2b1) and ψ2 = ψ(a2), allows the occupation of the lower energy 2b1 MO to
be significantly greater than that of the a2 MO in the 1A1 state. In contrast, in both
the triplet (3B2) and the open-shell singlet states (1B2) one π electron occupies the
2b1 MO and another π electron occupies the a2 MO. Therefore, neither of these
states can take full advantage of the lower energy of the 2b1 MO, relative to the a2
MO in OXA. That is why substitution of O in OXA for CH2 in TMM is computed
to stabilize the 1A1 state, relative to the 3B2, state by 17.5 kcal/mol.
One way to represent the greater contribution of the |…2b12> configuration,
relative to that of |…a22>, to the wave function for the singlet ground state of OXA
is by writing the zwitterionic resonance structure b in Figure 20 as a contributor to
this 1A1 wave function. However, the NIPE spectrum shows a C=O progession of
1680 ± 50 cm-1 in the 1A1 state (8, 9), indicating that resonance structure b in Figure
20 is not a significantly greater contributor to the singlet ground state of OXA than
similar resonance structures are to the singlet ground states of ordinary ketones.
The high C=O stretching frequency in the 1A1 state, found in the NIPE spectrum
of OXA (8, 9), provides experimental confirmation for predictions, made 15 (90),
19 (6), and 25 (91) years earlier, that the dominant contributor to the lowest singlet
state of OXA is resonance structure a in Figure 20, not resonance structure b.

287
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 20. Two possible resonance contributors to the structure of the singlet
ground state of OXA. The high C=O stretching frequency in this state, seen in
the NIPE spectrum of OXA•–, indicates that structure a is dominant and that the
contribution of structure b is not significantly greater in the singlet ground state
of OXA than in the singlet ground states of other ketones.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Nevertheless, structure b suggests that the lowest singlet state of OXA

might undergo an orbital symmetry-allowed (36), disrotatory ring closure
to cyclopropanone, with only a modest barrier. Indeed, calculations found
this mode of ring closure actually to be barrierless (8, 9, 89). Therefore,
the 1A1 state of OXA is actually the transition structure for ring opening of
cyclopropanone. Specifically, it is the Cs transition structure by which optically
active trans-cyclopropanone-2,3-d2 would racemize.
The fact that the 1A1 state of OXA is a transition structure explains why the
bands for formation of this state in the NIPE spectrum of OXA•– are so broad. The
Heisenberg uncertainty principle can be written

Therefore, if the lifetime of a state is very short (i.e., Δt is very small), the
uncertainty in the energy of the state (ΔE) must be large. Consquently, it can be
said the lines in the NIPE spectrum for the formation of the 1A1 state of OXA are
“uncertainty broadened.”
Despite the very short lifetime of the 1A1 transition structure, as already noted,
a progression of 1680 ± 50 cm-1 for C=O stretching can be seen in this state in the
NIPE spectrum of OXA•–. In fact, NIPES makes it possible to obtain spectroscopic
information about not only transition structures (8, 9, 15, 92) but also mountain
tops (93) on potential energy surfaces.

The Effect of the Substitution of O for CH2 on ΔEST in Going from MBQDM
to MBQ
If the substitution of the oxygen in OXA for a CH2 group in TMM reduces
the value of ΔEST by 17.5 kcal/mol, one night expect an even larger reduction
when the two oxygens in MBQ are substituted for the two methylene groups in
MBQDM. However, from the NIPE spectra of MBQDM and MBQ, the actual
change in ΔEST is only 0.6 kcal/mol, from ΔEST = 9.6 kcal/mol in MBQDM (87)
to ΔEST = 9.0 kcal/mol in MBQ (10, 11).
Interestingly, the prediction, that this would be found to be the case (7), was
published more than 20 years before the NIPE spectrum of MBQ•– was first
288
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 reported by Xuebin Wang and coworkers (10). They prepared this radical anion
by electrospraying the dianion of m-cresol into the gas phase. When stripped of
a sufficient number of solvent molecules, the dianion loses an electron and forms
the MBQ•– radical anion.
As shown in Figure 21, the NIPE spectrum of MBQ•– is complicated,
consisting of many peaks (10). Hence, it is not amenable to easy interpretation.
However, Dr. Bo Chen in my research group was able to use electrionic structure
and Franck-Condon calculations, in order to provide the simulation, also shown
in Figure 21, of the experimental NIPE spectrum (11).
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Figure 21. The NIPE spectra of MBQ•– (10). Shown under the experimental
spectra are the predicted spectra, obtained by using ezSpectrum (81) to calculate
the Franck-Condon factors for the vibrational progressions in the formation of
each electronic state of MBQ from MBQ•– (11). Reprinted with permission from
Chen, B.; Hrovat, D. A.; Deng, S. H. M.; Zhang, J.; Wang, X.-B.; Borden, W. T. J.
Am. Chem. Soc. 2014, 136, 3589.. Copyright 2014 American Chemical Society.

Comparison of the NIPE Spectra in Figures 18 and 21 shows that the lowest
singlet state changes from being 1A1 in MBQDM to 1B2 in MBQ. The reason for
this change is easy to see from the depictions of the bonding in these two singlet
states in Figure 19. As already discussed, in the hydrocarbon the aromaticity of the
π system of the benzene ring in the 1A1 state of MBQDM makes this the lowest
singlet state. However, on substitution of oxygen for CH2, the strengths of the
C=O π bonds in the 1B2 state of MBQ apparently overcome the aromaticity of the
benzene ring in the 1A1 state; so 1B2 becomes the ground state of MBQ.
Consequently, despite the change in ΔEST of only 0.6 kcal/mol on going from
MBQDM to MBQ, the substitution of O for CH2 actually has a very large effect on
289
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 the relative energies of the 1A1 and 1B2 states. As shown schematically in Figure
22 and in the NIPE spectrum of MBQDM•– in Figure 18, 1B2 is found to be 12.5
kcal/mol higher in energy than 1A1 (87). However, on going from MBQDM to
MBQ there is a 20.1 kcal/mol stabilization of 1B2, relative to 1A1, so that in the
NIPE spectrum of MBQ•– in Figure 21, 1B2 is lower in energy than 1A1 by 7.6
kcal/mol (11). Nevertheless, as shown in Figure 18, despite the huge change in
the relative energies of these two singlet states, the energy difference between the
lower of these two states and the 3B2 ground state changes by only 0.6 kcal/mol
on going from MBQDM to MBQ.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Figure 22. Relative energies (kcal/mol) of the low-lying electronic states of

MBQDM (87) and MBQ (11), obtained from their NIPE spectra. Calculated
C-C bond lengths (Å) for each state are also shown.

Electronic structure calculations were not only essential for deconvoluting the
NIPE spectrum of MBQ•– (11), but also for understanding why the substitutition
of the oxygens in MBQ for the CH2 groups in MBQDM has almost no effect on
ΔEST. As already mentioned, the prediction, that this would be found to be the
case (7), was published more than 20 years before the NIPE spectrum of MBQ
provided experimental confirmation of this prediction (11).

Violations of Hund’s Rule in Diradicals (14)

Fifty years ago, when I first began doing calculations on diradicals, one of
the things that most interested me was the PPP prediction that in square CBD and
in planar allene, ΔEST = 0. When I subsequently discovered that, with inclusion
of dynamic electron correlation, the singlet is actually predicted to be the ground
state of both diradicals (37, 40, 41, 45), I began to think about how these predicted
violations of Hund’s rule might be confirmed experimentally.

290
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Unfortunately, square CBD and planar allene are both transition structures, the
former for interconversion of the two rectangular minina on the potential energy
surface for singlet CBD and the latter for racemization of an optically active allene.
Therefore, measuring the energy difference between these two singlet transition
structures and the corresponding triplet energy minima would pose a formidable
challenge.

ΔEST in TME

Of course, in addition to CBD and planar allene, there are other diradicals
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

that have disjoint NBMOs. For example, the NBMOs of tetramethyleneethane

(TME), which are shown in Figure 12, can be chosen to have no atoms in common.
Consequently, to a first approximation, the singlet and the triplet have the same
energies. Inclusion of dynamic spin polarization, which creates some weak π
bonding between the central carbons of the two allylic radical subsystems, results
in the lowest singlet state of planar TME being predicted to be lower in energy
than the triplet state (94).
However, the conformational flexibility of TME complicates calculating
ΔEST in this diradical (94, 95). The relative energies of the in-phase and
out-of-phase combinations of the localized allyl NBMOs in Figure 12 depend
on the dihedral angle between the allyl groups in TME, as does the π bonding
between the two central carbons, which results from dynamic spin polarization
in the lowest singlet state.
The conformational flexibility of TME may be at least partly responsible for
the apparently contradictory experimental findings about whether a singlet or the
triplet is the ground state of TME. A linear Curie plot of the intensity of the EPR
signal versus 1/T indicates that the triplet is the ground state (96), but the NIPE
spectrum of TME•– shows that the singlet is the ground state (97). A possible
resolution of this apparent paradox has been proposed (4, 82).

ΔEST in TMB

If TME is viewed as being created by joining the nodal carbons of two allyl
radicals, 1,2,4,5-tetramethylenebenzene (TMB) can be seen to result from joining
the two pair of nodal carbons in two pentadienyl radicals. The two C-C single
bonds that join the pentadienyl radicals in TMB make it far less conformationally
flexible than TME, where the two allyl radicals are connected by just one, single,
C-C bond.
As shown in Figure 23, in TMB n* = n. Therefore, each of the two NBMOs
of TMB can be taken to be the NBMO of one of the two pentadienyl radicals.
These localized, non-bonding orbitals are, of course, disjoint; and, with inclusion
of dynamic spin polarization, TMB has been predicted to have a singlet ground
state (98).

291
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 23. Starred and unstarred carbons in 1,2,4,5-tetramethylenebenzene
(TMB) and its a2u and b1g NBMOs. The GVB orbitals of TMB are the sum
and difference of the a2u and b1g NBMOs, Therefore, each GVB orbital is the
non-bonding π orbital of one of the two pentadienyl radicals, of which TMB
is comprised. Therefore, the GVB MOs are disjoint, as can be predicted from
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

the fact that n* = n in TMB.

The localized pair of pentadienyl NBMOs of TMB are the GVB orbitals of
TMB. These GVB orbitals are the sum and the difference of the a2u and b1g TMB
NBMOs in Figure 23, which have the full D2h molecular symmetry. However,
the a2u and b1g MOs are not degenerate. With inclusion of the effects of 1,4-
interactions between the 2p-π AOs, a2u is slightly lower in energy than b1g.
In a two-configuration wave function for singlet TMB, which takes the form
of the wave function in eqn. 34, the coefficient of |…a2u2> will be larger than
that of |…b1g2>. Therefore, the wave function for the lowest singlet state of TMB
will have a larger contribution from the lower energy of these two NBMOs. In
contrast, in the triplet state, one electron occupies each of these NBMOs. The
greater occupancy of the lower-energy, a2u MO in the singlet than in the triplet
should tend to make the singlet the ground state of TMB.
Thus, there are two different effects that contribute to the prediction that
TMB should have a singlet ground state. The first is the effect of dynamic spin
polarization, which creates some additional π bonding in the singlet between the
pairs of pentadienyl nodal carbons (C1-C2, and C4-C5) at which negative spin
density appears. The second is the effect of the 1,4-π interactions between C3 and
C6 and between pairs of methylene carbons.
The relative sizes of these two effects can be estimated by first computing ΔEST
with a two-configuration wave function, like that in eqn. 34, and then recomputing
ΔEST with a multi-configurational wave function. In the former wave function.
the effects of the 1,4-π interactions are captured, but in the latter the effects of
dynamic spin polarization are also included. Such a comparison shows that 1,4-π
interactions contribute about 30% to the calculated value of ΔEST = 6.0 kcal/mol
in TMB, and dynamic spin polarization contributes the rest (98). Experiments by
Jerry Berson and coworkers have confirmed the prediction that TMB has a singlet
ground state (99, 100).
However, since the a2u and b1g NBMOs of TMB have slightly different
energies, can TMB really be claimed to violate Hund’s rule? A violation of the
strictest form of Hund’s rule can only be really claimed if (a) the two half-occupied
MOs in a molecule have exactly the same energies; and (b) the ground state of
the molecule is, nevertheless, a singlet.
292
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The only way to ensure that two MOs have exactly the same energies is if
symmetry makes them degenerate. That is what would make measurement of ΔEST
at the square (D4h) geometry of CBD very interesting; the two half-filled MOs
would be guaranteed to have the same energy. Unfortunately, I could not think of
an experiment that would allow ΔEST at the D4h geometry of CBD to be measured.

The NIPE Spectrum of COT•–

However, I was able to design an experiment that would measure ΔEST at the
D8h geometry of cyclooctatetraene (COT) (15). Like the D4h geometry of singlet
CBD, the D8h geometry of COT is the transition structure (TS) for bond shifting
in singlet COT. As is the case at the D4h geometry of CBD, at the D8h geometry
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

of COT dynamic spin polarization results in the singlet state being calculated to
be lower in energy than the triplet state (16, 17), in violation of Hund’s rule (14).
However, there is an important difference between CBD and COT. The
equilibrium geometry of singlet CBD is a highly reactive, planar D2h rectangle,
but the equilibrium geometry of singlet COT is the kinetically stable, non-planar,
D2d tub, which is shown in Figure 1.
The reaction path down from the D8h TS for bond shifting in COT to the
D2d equilibrium geometry bifurcates at the D4h TS for ring inversion, since either
of two tub geometries can be formed after bond shifting. In fact, the reaction
path from the D8h TS for bond shifting to the D4h TS for ring inversion contains
a valley-ridge inflection point, where the force constant for maintaining planarity
in singlet COT changes from positive to negative (101). Therefore, in descending
from the D8h geometry of the TS for bond shifting to the D2d equilibrium geometry,
a COT molecule need not stay on the reaction path that leads to the D4h TS for
ring inversion. Instead, a planar, D4h COT molecule can fall off the D4h ridgeline
into either of the D2d valleys below.
Conversely, a D2d COT molecule that undergoes bond shifting need not
pass through the D4h transition structure for ring inversion but can begin shifting
its four double bonds before complete planarity is reached. Nevertheless, one
can rigorously divide the energy required for bond shifting in D2d COT into the
energy required for passage through the D4h TS for ring inversion plus the energy
required for bond shifting, starting from the D4h TS for ring inversion. No D2d
COT molecules may undergo bond shifting by exactly this pathway. However,
it is convenient to think about the energy that is necessary to reach the D8h TS
for bond shifting as being the sum of the energy that is necessary to planarize the
eight-membered ring, plus the energy that is necessary to shift the double bonds
in the planarized, D4h COT ring.
In fact, in separate NMR experiments, Anet elegantly measured the activation
energy that is necessary, starting from D2d COT, to reach both the D8h transition
structure for double bond shifting (102) and the D4h transition structure for
ring inversion (103, 104). The difference of ca. 4 kcal/mol between these two
activation energies is, of course, equal to the barrier for double bond shifting
in planar D4h COT. Therefore, if one could measure the difference between the
energy of the singlet at the D4h TS for ring inversion and the energy of the triplet
at its D8h equilibrium geometry, addition of 4 kcal/mol to the energy of the singlet
293
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 would give the energy difference between the D8h singlet and the D8h triplet states
of COT.
I thought that it should be possible to use NIPES to measure the energy
difference between the D4h singlet and the D8h triplet. The COT radical anion
(COT•–) is known to be planar (105); so the Franck-Condon factors for formation
of non-planar singlet geometries from the planar radical anion should be zero. Of
course, planar COT•– could not only lose a photoelectron to form the D4h singlet,
it could also lose a photoelectron to form the D8h equilibrium geometry of the
triplet. The difference between the kinetic energies of these two photoelectrons
would be exactly equal to the energy difference between D4h singlet COT and
D8h triplet COT.
At a Gordon conference I told Paul Wenthold, who was then a post-doc in Carl
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

Lineberger’s research group, about my proposed NIPES experiment on COT•–;

and he agreed to try it when he returned to Boulder. He did the experiment, and it
worked (15). In fact, it worked even better than I had expected.
As I had hoped, the NIPE spectrum of COT•– showed a peak for formation of
the equilibrium geometry of the planar triplet state and another peak for formation
of the D4h TS for ring inversion in the lowest singlet state of COT. However,
in addition, 4.4 kcal/mol above the peak for formation of the D4h TS for ring
inversion was a pair of peaks. Their energy was in good agreement with the
energy, expected from Anet’s experiments (102–104), for the D8h TS for double
bond shifting (16, 17). In addition, as Carl Lineberger pointed out, at energies
near or above that of this D8h TS, there should actually be two vibrational wave
functions for antisymmetric C-C stretching, with slightly different energies (15).
Thus, the NIPE spectrum provided direct measurements of the energies of not only
the D8h equilibrioum geometry of the triplet state and the D4h TS for ring inversion
of the singlet state, but also of the D8h TS for double bond shifting in the singlet
state.
From the separation of 0.36 eV between the lower energy of the two peaks
for the D8h TS for double bond shifting and the 0→0 peak for formation of the
D8h triplet state, the NIPE spectrum of COT•– gives ΔEST = -8.3 kcal/mol in D8h
COT. The negative sign indicates that the singlet lies below the triplet in D8h COT.
At this geometry the NBMOs of COT are degenerate in energy. Therefore, the
strictest version of Hund’s rule is violated in D8h COT, and it is violated by a very
substantial amount of energy, > 8 kcal/mol (15).

(CO)4 – An Apparently Closed-Shell Molecule with a Triplet

Ground State
The story of how I became interested in cyclobutane-1,2,3,4-tetraone, (CO)4
(106), starts with the finding by Dr. Xin Zhou, a post-doc in my research group,
of a very large imaginary frequency in a molecule on which she was carrying out
calculations. The explanation of her observation led to her performing calculations
on other molecules that we expected would have very high-lying, highest-occupied
(HO)MOs and very low-lying, lowest unoccupied (LU)MOs (107, 108).

294
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Our papers on these molecules were read by Professor Rolf Gleiter. Rolf had
been my host in Heidelberg, when I spent three, three-month sabbaticals there, as
a Humboldt Senior Scientist. Rolf had published the results of some calculations
on (CO)4, which he had also found to have a very high-lying HOMO and a low-
lying LUMO (109). He had also found (CO)4 to be a computationally challenging
molecule; and he, therefore, suggested that my group might consider performing
some higher level calculations on it.
Professor Haijun Jiao had also carried out calculations on (CO)4, and the
results of his calculations were very surprising in predicting that this very ordinary-
looking tetraketone might have a triplet ground state (110). The results of our own
calculations, at several different levels of theory, also predicted a triplet ground
state for (CO)4 (111). However, like the results of Rolf (109) and Haijun (110), the
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

lower-level calculations performed by us gave such widely different results that we

were skeptical of whether the results of our highest-level calculations [CCSD(T)]
could be trusted. Therefore, we took a different approach.
(CO)4 has a very high-lying σ HOMO (b2g) and a low-lying π LUMO (a2u).
These are shown in Figure 24. The singlet electronic state, in which a2u is doubly
occupied, is only slightly higher in energy than the singlet state in which b2g is
doubly occupied. Thus, it is understandable that some calculations predict a triplet
ground state in which b2g and a2u are each singly occupied.

Figure 24. The two MOs that are singly occupied in the triplet ground state
of (CO)4. Reprinted with permission from Bao, X.; Zhou, X.; Lovitt, C. F.;
Venkatraman, A.; Hrovat, D. A.; Gleiter, R.; Hoffmann, R.; Borden, W. T. J. Am.
Chem. Soc. 2012, 134, 10259. Copyright 2012, American Chemical Society.

The b2g and a2u MOs obviously have atoms in common. Therefore, if b2g and
a2u had exactly the same energy, then Hund’s rule should be obeyed (12, 13); and
the ground state of (CO)4 should be the triplet. We calculated the energies of these
orbitals in several different ways. For instance, we computed the IEs of the singly
occupied MOs in the triplet at two different levels of theory. Based on the results
of our calculations, our conclusion was that the triplet is the ground state of (CO)4
(18).
Our paper was apparently read by Dr. Xuebin Wang, who tested our prediction
by obtaining the NIPE spectrum of (CO)4•– (19). He generated the radical anion
by electrospraying a solution of the dianion of the commercially available squaric
acid [C4O2(OH)2] into the gas phase. In the gas-phase, the squarate dianion (CO)4-2
loses an electron and forms (CO)4•–.
295
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 The NIPE spectrum of the radical anion is shown in Fig. 25 (19, 20). Based
on the large intensity of the lowest energy peak, Xuebin assigned it as belonging to
the triplet, giving a value of ΔEST = 2.5 kcal/mol, with the triplet state lower than
the singlet state. A value of this size was consistent with the results of high-level
calculations, done by ourselves (111)and by others (112).
Unlike (CO)4, (CO)3, (CO)5, and (CO)6 are all calculated to have singlet
ground states (113), as is (CS)4 (114). Xuebin obtained the NIPE spectra of (CO)3
(115), (CO)5 and (CO)6 (79), and (CS)4 (116), and the spectra confirmed our
prediction that each of these molecules has a singlet ground state. We collaborated
with Xuebin on assigning all of the peaks in his NIPE spectra by computing
the positions of the band origins and the frequencies and the Franck-Condon
intensities of the peaks in the vibrational progressions in each band. The results
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

of our calculations on the vibronic peaks in (CO)4 are shown in Figure 25, under
the experimental spectrum (20). They confirm Xuebin’s original assignment of
the triplet as the ground state of (CO)4 (19).

Figure 25. The NIPE spectrum of (CO)4•-, with the simulated vibrational
structure superimposed onto the experimental spectrum. Note that only two
electronic transitions, i.e., to the 3B2u ground state (blue lines) and to the
low-lying 1A1g excited state (green lines) were used in the simulation. The
simulation shows that the B peaks are not due to a third electronic state (19), but
are part of a vibrational progression in the symmetrical C-C stretching mode in
the closed-shell, 1A1g state. Reprinted with permission from Bao, X.; Hrovat, D.
A.; Borden, W. T.; Wang, X. B. J. Am. Chem. Soc. 2013, 135, 4291. Copyright
2013, American Chemical Society.

Asking why the ground state of (CO)4 is the triplet is equivalent to asking
why the b1g and a2u MOs are accidentally degenerate in energy (18). Both MOs
can be viewed as being the in-phase combinations of the C-O π* orbitals of four
296
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 CO molecules (113). Normally, one would expect the σ overlaps of the 2p-π AOs
in b1g to provide more C-C bonding than the π overlaps of 2p-π AOs in a2u; and
this probably is the case in (CO)4 too.
However, in (CO)4, in addition to the 1,2 interactions between AOs on
adjacent carbons, there are also substantial 1,3-interactions between 2p AOs on
pairs of carbons across the diagonals of the four-membered ring. As can be seen
in Figure 24, these 1,3-interactions are bonding in a2u, but antibonding in b2g. It
appears that the difference between the favorable 1,3-interactions in a2u and the
unfavorable 1,3-interactions in b2g almost exactly balances the greater strength of
the 1,2-σ interactions in b2g, compared to the 1,2-π interactions in a2u. It is this
accidental balance between the sizes of two different types of C-C interactions
that leads to the b2g and a2u MOs of (CO)4 having almost exactly the same energy
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

(113) and, hence, to this molecule being both predicted (18) and found (19, 20) to
have a triplet ground state.

Future Research on Diradicals

It is traditional to end a scholarly treatise on a field of chemistry with some

predictions about the future evolution of the field. However, there is a Danish
saying, “It is difficult to make predictions, especially about the future.”
These words of wisdom have been attributed to a diverse group of famous
people, including Nostradamus, Mark Twain, Niels Bohr, Samuel Goldwyn, and
Yogi Berra (117). I believe in the truth of this saying, even if none of these
wise men ever said it. In fact, I just authored a journal article entitled, “The
Unpredictability of Research Directions… in Physical-Organic Chemistry” (106).
Therefore, I would not only be foolish, but also hypocritical, if I ventured to make
any predictions about the future evolution of the field of diradicals.
However, I do feel secure in making some predictions about the near-term
future of my own research in this area of chemistry. These predictions involve
collaborative projects with Xuebin Wang that we have either discussed or have
already begun.
By electrospraying solutions of the dianions into the gas-phase, Xuebin has
the ability to generate the corresponding radical anions and to obtain their NIPE
spectra. He has successfully generated the following radical anions in this manner:
(CO)n•–, with n = 3 (115), 4 (19), 5 and 6 (20), (CS)4•– (116), and MBQ•- (10, 11).
Adding to this list of organic radical anions is easy; one has only to think of acids
that are strong enough to form dianions. Three that come to mind immediately are
the bis-enol of 1,2,4,5-cyclohexanetetraone, oxalic acid, and carbonic acid.
Electrospraying the corresponding dianions into the gas-phase, should result
in electron loss, leading to formation of TOTMB•–, TOTME•– and TrOTMM•–.
Ejection of a photoelectron from the radical anions thus formed, would give the
oxygen containing diradicals shown in Figure 26. They are, respectively the
tetraoxo (TO) derivatives of TMB, TME, and the trioxo (TrO) derivative of
TMM.
297
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Figure 26. Three diradicals in which oxygens replace the CH2 groups of,
respectively, TMB, TME, and TMM.

Xuebin has already obtained the first NIPE spectra of TOTMB•– and
TrOTMM•– radical anions, and my group has completed calculations on both
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

the TOTMB and TrOTMM diradicals. The results of our calculations and
Xuebin’s experiments may have already been published by the time that this
chapter appears.

Acknowledgments
I thank the National Science Foundation for supporting my research over
the past 50 years, first with an NSF graduate Fellowship and then with a long
succession of NSF research grants. I also want to acknowledge the guidance
and inspiration provided to me by my mentors – Professors E. J. Corey, H.
C. Longuet-Higgins, and William von Eggers Doering, – and by my senior
collaborators Jerry Berson, Ernest R. Davidson, Carl Lineberger, and, most
recently, Xuebin Wang. I am indebted to all the members of my research group,
who, over the years have done calculations on diradicals. Their names appear
on the papers that we coauthored, most of which are cited in the references for
this chapter. I am particularly grateful to Dr. David A. Hrovat, who has been my
friend and collaborator since he joined my group in 1984, for what he warned me
would be “only one year.” I dedicate this chapter to the memory of my college
roommate, Robert David Joffee. I will always be grateful to Bob for his friendship
and for permanently loaning me his copy of Streitwieser’s Molecular Orbital
Theory for Organic Chemists.

References
1. Salem, L.; Rowland, C. Angew. Chem., Int. Ed. Engl. 1972, 11, 92.
2. Borden, W. T. In Diradicals; Borden, W. T., Ed.; Wiley-Interscience: New
York, 1982; pp 1−72 (Review).
3. For those chemists who care more about linguistic purity than about
consistent nomenclature, it must be admitted that “bi” and “radical” are both
Latin in origin; whereas, “di” is, of course, Greek.
4. Lineberger, L. C.; Borden, W. T. Phys. Chem. Chem. Phys. 2011, 13, 11792.
5. Actually, I believe that the discovery, which has been most valuable to me
over the course of my 50 year career as a chemist, is that, with enough
298
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 persistence, almost any problem, no matter how puzzling it may initially
seem, can eventually be solved.
6. Coolidge, M. B.; Yamashita, K.; Morokuma, K.; Borden, W. T. J. Am. Chem.
Soc. 1990, 112, 1704.
7. Fort, R. C., Jr.; Getty, S. J.; Hrovat, D. A.; Lahti, P. M.; Borden, W. T. J. Am.
Chem. Soc. 1992, 114, 7549.
8. Ichino, T.; Villano, S. M.; Gianola, A. J.; Goebbert, D. J.; Velarde, L.;
Sanov, A.; Blanksby, S. J.; Zhou, X.; Hrovat, D. A.; Borden, W. T.;
Lineberger, W. C. Angew. Chem., Int. Ed. 2009, 48, 8509.
9. Ichino, T.; Villano, S. M.; Gianola, A. J.; Goebbert, D. J.; Velarde, L.;
Sanov, A.; Blanksby, S. J.; Zhou, X.; Hrovat, D. A.; Borden, W. T.;
Lineberger, W. C. J. Phys. Chem. A 2011, 115, 1634.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

10. Fu, Q.; Yang, J.; Wang, X.-B. J. Phys. Chem. A 2011, 115, 3201.
11. Chen, B.; Hrovat, D. A.; Deng, S. H. M.; Zhang, J.; Wang, X.-B.; Borden, W.
T. J. Am. Chem. Soc. 2014, 136, 3589.
12. Hund, F. Linienspektren und Periodisches System der Elemente; Springer-
Verlag: Berlin, 1927; p 124.
13. Kutzelnigg, W. Angew. Chem., Int. Ed. Engl. 1996, 35, 4001 (Review).
14. Borden, W. T.; Iwamura, H.; Berson, J. A. Acc. Chem. Res. 1994, 27, 109
(Review).
15. Wenthold, P. G.; Hrovat, D. A.; Borden, W. T.; Lineberger, W. C. Science
1996, 272, 1456.
16. Hrovat, D. A.; Borden, W. T. J. Am. Chem. Soc. 1992, 114, 5879.
17. Dewar, M. J. S.; Merz, K. M. J. Phys. Chem. A 1985, 89, 4739.
18. Zhou, X.; Hrovat, D. A.; Borden, W. T. J. Phys. Chem. A 2010, 114, 1304.
19. Guo, J.-C.; Hou, G.-L.; Li, S. D.; Wang, X.-B. J. Phys. Chem. Lett. 2012, 3,
304.
20. Bao, X; Hrovat, D. A.; Borden, W. T.; Wang, X. B. J. Am. Chem. Soc. 2013,
135, 4291.
21. Chichibabin, A. E. Ber. Dtsch. Chem. Ges. 1907, 40, 1810.
22. Thiele, J.; Balhorn, H. Chem. Ber. 1904, 37, 1463.
23. Schlenck, W.; Brauns, M. Ber. Dtsch. Chem. Ges. 1915, 48, 661.
24. For reviews of quinodimethanes, see (a) Platz, M. In Diradicals; Borden, W.
T., Ed.; Wiley-Interscience: New York, 1982; pp 195−258. (b) Berson, J. A.
In The Chemistry of the Quinoid Compounds; Patai, S., Rappoport, Z., Eds.;
John Wiley & Sons: New York, 1988; Vol. II, Chapter 10.
25. Longuet-Higgins, H. C. J. Chem. Phys. 1950, 18, 265.
26. Watts, L.; Fitzpatrick, J. D.; Pettit, R. J. Am. Chem. Soc. 1966, 88, 623.
27. Dowd, P. J. Am. Chem. Soc. 1966, 88, 2587.
28. Crawford, R. J.; Cameron, D. M. J. Am. Chem. Soc. 1966, 88, 2589 (Thermal
deazatation affords methylenecyclopropane).
29. Dowd, P. Acc. Chem. Res. 1972, 5, 242 (Review).
30. Baseman, R. J.; Pratt, D. W.; Chow, M.; Dowd, P. J. Am. Chem. Soc. 1976,
98, 5726.
31. Streitwieser, A. Molecular Orbital Theory for Organic Chemists; John Wiley
& Sons: New York, 1961.
32. Borden, W. T. J. Org. Chem. 2111, 76, 2943 (Perspective).

299
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 33. Pioneers of Quantum Chemistry; Strom, E. T., Wilson, A. K., Eds.; ACS
Symposium Series; American Chemical Society: Washington, DC, 2013; pp
155−195.
34. Parr, R. G. Quantum Theory of Molecular Structure; W. A. Benjamin, Inc.:
1963, New York.
35. Borden, W. T. Chem. Commun. 1969, 881.
36. Woodward, R. B.; Hoffmann, R. Angew. Chem., Int. Ed. Engl. 1969, 8, 781.
37. Borden, W. T. J. Chem. Phys. 1966, 45, 2512.
38. However, my first published paper was experimental. It was titled,
“Separation of Surfactants in Admixture by Thin-Layer Chromatography,”
and it was published in the Journal of the American Oil Chemical Society.
39. Angus, R. O., Jr.; Schmidt, M. S.; Johnson, R. O. J. Am. Chem. Soc. 1985,
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

107, 532 (CASSCF calculations place the singlet state of planar allene 4.9
kcal/mol below the triplet).
40. Craig, D. P. Proc. R. Soc. London, Ser. A 1950, 202, 498.
41. Buenker, R. J.; Peyerimhoff, S. J. Chem. Phys. 1968, 48, 354.
42. Goddard, W. A., III; Dunning, T. H.; Hunt, W. J. Acc. Chem. Res. 1973, 6,
368 (Review).
43. Borden, W. T. Modern Molecular Orbital Theory for Organic Chemists;
Prentice-Hall: Englewood Cliffs, NJ, 1975.
44. Review: Bally, T.; Borden, W. T. In Reviews in Computational Chemistry;
Lipowitz, K. B., Boyd, D. B., Eds.; Wiley: New York, 1999; Vol. 13, pp
1−99.
45. Borden, W. T. J. Am. Chem. Soc. 1975, 97, 5968.
46. Kollmar, H.; Staemmler, V. Theor. Chim. Acta 1978, 48, 223.
47. Borden, W. T. Tetrahedron Lett. 1967, 259.
48. Longuet-Higgins, H. C.; Abrahamson, E. W. J. Am. Chem. Soc. 1965, 87,
2045.
49. Hoffmann, R.; Woodward, R. B. J. Am. Chem. Soc. 1965, 87, 2046.
50. Doering, W. von E.; Roth, H. D. Tetrahedron 1970, 26, 2825.
51. Gilbert, J. C.; Butler, J. R. J. Am. Chem. Soc. 1970, 92, 2168.
52. Jones, M., Jr.; Hendrick, M. E.; Gilbert, J. C.; Butler, J. R. Tetrahedron Lett.
1970, 845.
53. Dewar, M. J. S.; Wasson, J. W. J. Am. Chem. Soc. 1971, 93, 3081.
54. Borden, W. T.; Salem, L. J. Am. Chem. Soc. 1973, 95, 932.
55. Berson, J. A.; Salem, L. J. Am. Chem. Soc. 1972, 94, 8917.
56. Borden, W. T.; Davidson, E. R. J. Am. Chem. Soc. 1977, 99, 4587.
57. Yarkony, D. R.; Schaefer, H. F., III. J. Am. Chem. Soc. 1974, 96, 3754.
58. Borden, W. T. J. Am. Chem. Soc. 1975, 97, 2906.
59. A wonderful example of a diradical in which different sets of MOs are
optimal for the lowest singlet and triplet states is provided by phenylnitrene.
In the open-shell singlet state one nonbonding electron is localized in a
2p-σ AO on nitrogen, while the electron of opposite spin is localized in
the π system of the benzene ring. In the lowest triplet state the unpaired π
electron is much more delocalized into the 2p-π AO on nitrogen than in the
open-shell singlet state. This difference between the orbitals that are optimal

300
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 for these two states explains many aspects of the chemistry and spectroscopy
of phenylnitrene (60, 61).
60. Borden, W. T.; Gritsan, N. P.; Hadad, C. M.; Karney, W. L; Kenmitz, C. R.;
Platz, M. S. Acc. Chem. Res 2000, 33, 765 (Review).
61. Karney, W. L.; Borden, W. T. In Advances in Carbene Chemistry; Brinker,
U. H., Ed.; Elsevier: Amsterdam, 2001; Vol. 3, pp 206−251 (Review).
62. Borden, W. T. J. Am. Chem. Soc. 1976, 98, 2695.
63. Davidson, E. R.; Borden, W. T. J. Chem. Phys. 1976, 64, 663.
64. Davidson, E. R.; Borden, W. T. J. Phys. Chem. 1983, 87, 4783.
65. Davidson, E. R.; Borden, W. T. J. Am. Chem. Soc. 1977, 99, 2053.
66. Borden, W. T.; Davidson, E. R.; Hart, P. J. Am. Chem. Soc. 1978, 100, 388.
67. Coulson, C. A.; Rushbrooke, G. S. Proc. Cambridge Philos. Soc. 1940, 36,
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

193.
68. Ovchinnikov, A. A. Theor. Chim. Acta 1978, 47, 297.
69. Yamaguchi, K.; Jensen, F.; Dorigo, A.; Houk, K. N. Chem. Phys. Lett. 1988,
149, 537.
70. Cramer, C. J. Essentials of Computational Chemistry; John Wiley & Sons:
New York, 2002; p 212.
71. Borden, W. T.; Davidson, E. R. Acc. Chem. Res. 1996, 29, 87.
72. Andersson, K.; Malmqvist, P.-Å.; Roos, B. O. J. Chem. Phys. 1992, 96,
1218.
73. Andersson, K.; Aquilante, F.; Barysz, M.; Bernhardsson, A.; Blomberg, M.
R. A.; Carissan, Y.; Cooper, D. L.; Cossi, M.; DeVico, L.; Ferré, N.; Fülscher,
M. P.; Gaenko, A.; Gagliardi, L.; Ghigo, G.; de Graaf, C.; Gusarov, S.; Hess,
B. A. Hagberg, D.; Holt, A.; Karlström, G.; Lindh, R.; Malmqvist, P.-Å.;
Nakajima, T.; Neogrády, P.; Olsen, J.; Pedersen, T.; Pitonak, M.; Raab,
J.; Reiher, M.; Roos, B. O.; Ryde, U.; Schimmelpfennig, B.; Schütz, M.;
Seijo, L.; Serrano-Andrés, L.; Siegbahn, P. E. M.; Stålring, J.; Thorsteinsson,
T.; Veryazov, V.; Widmark. P.-O. MOLCAS, version 7; Lund University:
Sweden, 2008.
74. For more detailed discussions of Curie plots, see the reviews by Platz and by
Berson in ref (24).
75. Kothe, G.; Denkel, K. H.; Sümmermann, W. Angew. Chem., Int. Ed. 1970,
9, 906.
76. Luckhurst, G. R.; Pedulli, G. F. J. Chem. Soc. B 1971, 329.
77. Wright, B. B.; Platz, M. S. J. Am. Chem. Soc. 1983, 105, 628.
78. Wenthold, P. G.; Hu, J.; Squires, R. R.; Lineberger, W. C. J. Am. Chem. Soc.
1996, 118, 475.
79. (a) Franck, J. Trans. Faraday Soc. 1926, 21, 536. (b) Condon, E. Phys. Rev.
1926, 28, 1182.
80. Bao, X.; Hrovat, D. A.; Borden, W. T.; Wang, X.-B. J. Am. Chem. Soc. 2013,
135, 4291.
81. Mozhayskiy, V. A.; Krylov, A. I. ezSpectrum, version 3.0; USC,
http://iopenshell.usc.edu/downloads.
82. Berson, J. A. In Reactive Intermediate Chemistry; Moss, R. A., Platz, M. S.,
Jones, M., Jr., Eds.; Wiley Interscience: Hoboken, NJ, 2004; pp 165−203
(Review of Non-Kekulé Molecules).

301
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 83. Wenthold, P. G.; Hu, J.; Squires, R. R. J. Am. Chem. Soc. 1994, 116, 6961.
84. Feller, D.; Tanaka, K.; Davidson, E. R.; Borden, W. T. J. Am. Chem. Soc.
1982, 104, 967.
85. Cramer, C. J.; Smith, B. A. J. Phys. Chem. 1996, 100, 9664.
86. Slipchenko, L. V.; Krylov, A. I. J. Chem. Phys. 2005, 123, 84107.
87. Wenthold, P. G.; Kim, J. B.; Lineberger, W. C. J. Am. Chem. Soc. 1997, 119,
1354.
88. Hrovat, D. A.; Murcko, M. A.; Lahti, P. M.; Borden, W. T. J. Chem. Soc.,
Perkin Trans. 2 1998, 1037.
89. Mozhayskiy, V.; Goebbert, D. J.; Velarde, L.; Sanov, A.; Krylov, A. I. J.
Phys. Chem. A 2010, 114, 6935–6943.
90. Lim, D.; Hrovat, D. A.; Borden, W. T.; Jorgensen, W. L. J. Am. Chem. Soc.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

1994, 116, 3494.

91. Osamura, Y.; Borden, W. T.; Morokuma, K. J. Am. Chem. Soc. 1984, 106,
5112.
92. Neumark, D. M. Acc. Chem. Res. 1993, 26, 33 (Review).
93. Chen, B.; Hrovat, D. A.; West, R.; Deng, S. H. M.; Wang, X. B.; Borden, W.
T. J. Am. Chem. Soc. 2014, 136, 12345.
94. Du, P.; Borden, W. T. J. Am. Chem. Soc. 1987, 109, 930.
95. (a) Nachtigall, P.; Jordan, K. D. J. Am. Chem. Soc. 1992, 114, 4743. (b)
Nachtigall, P.; Jordan, K. D. J. Am. Chem. Soc. 1993, 115, 270.
96. Dowd, P.; Chang, W.; Paik, Y. H. J. Am. Chem. Soc. 1977, 99, 8507.
97. Clifford, E. P.; Wenthold, P. G.; Lineberger, W. C.; Ellison, G. B.; Wang, C.
X.; Grabowski, J. J.; Jila, F.; Jordan, K. D. J. Chem. Soc., Perkin Trans. 2
1998, 1015.
98. Du, P.; Hrovat, D. A.; Borden, W. T.; Lahti, P. M.; Rossi, A. R.; J. A.
Berson, J. A. J. Am. Chem. Soc. 1986, 108, 5072.
99. (a) Reynolds, J. H.; Berson, J. A.; Kumashiro, K. K.; Duchamp, J. C.; Zilm,
K. W.; Rubello, A.; Vogel, P. J. Am. Chem. Soc., 1992, 114, 763. (b)
Reynolds, J. H.; Berson, J. A.; Scaiano, J. C.; Berinstain, A. B. J. Am. Chem.
Soc. 1992, 114, 5866. Reynolds, J. H.; Berson, J. A.; Kumashiro, K. K.;
Duchamp, J. C.; Zilm, K. W.; Scaiano, J. C.; Berinstain, A. B.; A. Rubello,
A.; Vogel, P. J. Am. Chem. Soc. 1993, 115, 8073.
100. See also Hrovat, D. A.; Borden, W. T. J. Am. Chem. Soc. 1994, 116, 6327.
101. Castaño, O.; Palmeiro, R.; Frutos, L. M.; Luisandrés, J. J. Comput. Chem.
2002, 23, 732.
102. Anet, F. A. L. J. Am. Chem. Soc. 1962, 84, 671.
103. Anet, F. A. L.; Boum, A. J. P.; Lin, Y. S. J. Am. Chem. Soc. 1964, 86, 3576.
104. See also Oth, J. F. M.Pure Appl. Chem., 1971, 25, 573.
105. Hammons, J. A.; Hrovat, D. A.; Borden, W. T. J. Am. Chem. Soc. 1991, 113,
4500and references therein.
106. Borden, W. T. Israel J. Chem. 2015in press.
107. Zhou, X.; Hrovat, D. A.; Borden, W. T. Org. Lett. 2008, 5, 893.
108. Zhou, X.; Hrovat, D. A.; Borden, W. T. J. Phys. Chem. A 2010, 114, 3683.
109. Gleiter, R.; Hyla-Kryspin, I.; Pfeifer, K.-H. J. Org. Chem. 1995, 60, 5878.
110. Jiao, H.; Frapper, G.; Halet, J.-F.; Saillard, J.-Y. J. Phys. Chem. A 2001, 105,
5945.

302
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 111. Zhou, X.; Hrovat, D. A.; Gleiter, R.; Borden, W. T. Mol. Phys. 2009, 107,
863.
112. Perera, A.; Molt, R. W., Jr.; Lotrich, V. F.; Bartlett, R. J. Theor. Chem. Acc.
2014, 133, 1514.
113. Bao, X.; Zhou, X.; Lovitt, C. F.; Venkatraman, A.; Hrovat, D. A.; Gleiter, R.;
Hoffmann, R.; Borden, W. T. J. Am. Chem. Soc. 2012, 134, 10259.
114. Bao, X.; Hrovat, D. A.; Borden, W. T. Chem. Eur. J. 2013, 19, 5687.
115. Chen, B.; Hrovat, D. A.; Deng, S. H. M.; Zhang, J.; Wang, X.-X.-B.;
Borden, W. T. J. Am. Chem. Soc. 2014, 136, 3589.
116. Zhang, J.; Hrovat, D. A.; Sun, Z.; Bao, X.; Borden, W. T.; Wang, X.-B. J.
Phys. Chem. A 2013, 117, 7841.
117. See http://quoteinvestigator.com/2013/10/20/no-predict/.
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ch011

303
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Subject Index
D planar allene, violation of Hund’s rule,
266
Diradicals planar singlet TMM, wave functions,
cyclobutane-1,2,3,4-tetraone, (CO)4, 294 271
diradicals, future research, 297 relative PPP energies, schematic
three diradicals, CH2 groups, 298f depiction, 263f
diradicals, history, 254 singlet TMM versus singlet CBD, 270
chemistry professor, 256 singlet wave function, 261
Chichibabin’s (1) hydrocarbons, 254f TMM and CBD, PPP calculations,
Diels-Alder trapping, generation and 259
stereospecific, 254f trimethylenemethane, singlet and
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ix002

triplet TMM, generation, 255f triplet states, 268

ΔEST, measurements in diradicals, 277 triplet state, schematic depiction, 267f
(CO)5•−, NIPE spectrum, 281f two components, schematic depiction,
negative ion photoelectron 260f
spectroscopy (NIPES), 278 radical anions, NIPE spectra, 283
NIPES experiment, schematic bonding, schematic depiction, 286f
depiction, 279f Lineberger’s experimental values, 287
non-zero overlaps, vibrational wave MBQ•−, NIPE spectra, 289f
functions, 282f MBQDM, NIPE spectrum, 285f
vibrational progressions, 280 MBQDM, schematic depiction, 284f
ΔEST, size, 276 planar TMM, three electronic states,
ground state of a diradical, prediction, 284f
272 possible resonance contributors, 288f
Ovchinnikov’s equation, 275 simulated vibrational structure, NIPE
square CBD and tetramethyleneethane spectrum, 296f
(TME), NBMOs, 274f two MOs, 295f
TMM and meta-benzoquinodimethane Doering, William von Eggers, 199
(MBQDM), NBMOs, 273f Doering, Bill, 200f
Hund’s rule in diradicals, violations, 290 bullvalene story, 213
cyclooctatetraene (COT), geometry, Chinese graduate program (CGP), 219
293 departmental seminars, 207
1,2,4,5-tetramethylenebenzene deuterium-exchange fellowship, 215
(TMB), starred and unstarred diazomethane, 214
carbons, 292f diazomethane, quantities, 201
introduction dihalocarbene addition,
D4h transition structure (TS), 253f stereochemistry, 208
diradicals/biradicals, 251 dissertation, 209
singlet-triplet energy differences, Doering's nickname, 212
experimental techniques, 252 NMU/KOH reaction, 216
PPP calculations on diradicals, results, nuclear non-proliferation, 220
257 organic chemistry, 205
dynamic spin polarization, schematic organic chemistry, laboratory
depiction, 268f technique, 213
Hund’s rule, 264 quantitative analysis, 202
lowest singlet and triplet states, 265 reaction mechanismwise, 218
lowest singlet state of TMM, spiropentane, carbon-carbon bonds,
geometry, 269 205
low-lying electronic states, 262 Stanford chemists, 210
NBMOs, 270f weekly group seminars, 210
planar allene, nonbonding (NB)MOs, wet chemistry, 201
264f young Doering, Bill, 204f

311
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 F early optically active deuterium
compounds, 80f
Free radical physical organic chemistry, iodine isotopes, SN2 exchange, 82f
223 optically active 1-butanol-1-d,
CKI and KUI at 12 years of age, 227f preparation, 81f
CKI and KUI at 70 years of age, 227f smaller methyl, favored transition
CKI skiing, 224f structure, 81f
electron paramagnetic resonance, 235 secondary α-deuterium isotope effects,
garden in Sussex, 225f 84
great water-ski caper, 224f cycloaddition of nitrostyrene to
hydrocarbon oxidation, absolute rate cyclopentadiene, isotope effects,
constants, 230 89f
hydrocarbon oxidation by phenols, cycloaddition reactions, synchronicity,
mechanism of inhibition, 228 88f
2,6-di-tertbutyl-4-methylphenol, 229 out-of-plane bending motions, 86f
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ix002

KUI skiing, 225f solvolysis, α-d isotope effects, 88f

nanosecond laser flash photolysis (LFP) some bending vibrations, 85f
and solvent effects, 245 vibration frequency, effect, 87f
1998 NSERC medal, 249f tritium, 89
christened sequential proton loss
electron transfer, 248s
hydrogen bond acceptor (HBA) N
solvent, 246
quantitative explanation, 247
Norbornyl cation isomers, 139
NOCT, invention of a nitric oxide
conclusions, 166
cheleotropic, 244
historical background, 142
Norris Award, 226
camphenyl cation, representation,
peroxyl radical trapping antioxidants,
143f
238
2-norbornyl cation, 144t
physical organic chemistry,
norbornyl compounds vs. cyclohexyl,
competitive kinetic methods,
solvolysis rates, 145f
241
infrared spectroscopy, classical vs.
α-tocopherol, bioactivity, 239s
nonclassical structure, 153
α-tocopherol, chemical significance,
Born-Oppenheimer molecular
238c
dynamics (BOMD) simulation,
α-tocopherol, sold as vitamin E, 240
158f
persistent carbon-centered radicals, 234
C7H11+, experimental spectrum, 155f
pro-oxidant, vitamin E, 242
C7H11+ from protonation of
quantum mechanical tunneling, 236
norbornene, experimental spectrum,
reduction of alkyl halides, absolute rate
156f
constants, 231
clarifying wagner-meerwein reaction
synthesis via a chain reaction,
products, 161f
advantage, 233
1,3-dimethylcyclopentenyl, computed
SOTS-1, invention of the first superoxide
spectra, 157f
thermal source, 243
Duncan’s IR-PD apparatus, 154f
2-endo-norbornyl cation, 158
endo-2-norbornyl cation, norpinyl
H product, 160f
excerpts from Friedrich and Winstein,
Hydrogen isotopes, 77 164f
deuterium, discovery, 78 exo vs. endo 2-norbornyl solvolyses,
hydrogen-deuterium asymmetry, 79 159f
1-bromobutane-1-d, racemization, 83f L2-endo-norbornyl (norpinyl) cations,
chlorosulfite, SN2 reaction, 84f 165f
low energy isomers, 156f

312
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 pinene-HCl to bornyl chloride (endo) O
rearrangement, 162f
pinyl to 2-endo-norbornyl dyotropic Optical activity, study
rearrangement, 163f computational studies, 25
potential experiment solution, 153f gas phase measurements, 36
puzzling wagner-meerwein product cavity ring-down polarimeter, 37f
stereochemistry, 161f optical rotation and solvent dielectric
introduction, 140 constant, relationship, 38f
Schleyer, Paul von R., 140f introduction, 23
2-norbornyl cation, direct observation, minimal basis sets, use, 35
146 logarithmic scale, effect of basis set,
excerpts from Roberts, Lee, 149f 36f
excerpts from Saunders, Schleyer, norbornenone, 40
148f optical activity, 26
excerpts from Schleyer, Watts, 147f Rosenfeld’s equation, 27
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ix002

2-norbornyl cation, 150f optical activity, experimental studies, 28

2-norbornyl cation, 1H NMR spectra, calculated conformational energies,
149f 29f
overview, 141 calculated optical rotations,
norbornyl cation citations, 142f relationship, 31f
norbornyl citations, 141f 3-chloro-2-butene, calculated optical
symmetrical bridged structure, classical rotation profile, 30f
ions, 145 3-Chloro-1-butene conformers, 29f
exo vs. endo 2-norbornyl solvolyses, conformational mixtures, calculated
146f and observed specific rotations, 32t
nonclassical ion vs. rapid equilibrium oxidation, mechanisms, 25
model, 145f small ring compounds, 24
X-ray crystallography, adventures, 150 spectroscopic studies, 26
[C7H11]+[Al2Br7]- • CH2Br2 crystal sum-over states calculations, 38
structure, asymmetric unit, 152f chloropropionitrile, sum-over-states
The Nonclassical Ion Problem, cover, study for 1500 excited states, 39f
152f chloropropionitrile, sum-over-states
2-norbornyl cation, disorder, 151f study for first 125 excited states, 39f
unsubstituted 2-norbornyl cation XRD norbornenone model and its C=C-C-C
structure, 153f torsion angle, relationship, 40f
Norris, James Flack twisted achiral compounds, 32
brief biography, 3 B3LYP calculated specific rotations,
Dr. Norris, James Flack, 4f effect of basis set, 33t
leadership qualities, 5 difference density plot, 34f
National Research Council Section, 6 twisted butane, calculated optical
Norris's right eye, cataract surgery, 7 rotations, 32f
dedication, 18
James Flack Norris awards, 16
physical organic chemistry, early days, 1
quantitative organic chemistry, P
contributions, 12
chemical theory, early differences, 13 Physical organic chemistry
Columbia University, presentation, 14 antiaromaticity, 62
ethanolysis, rates, 15t increasing antiaromatic
Remsen, Ira, influence, 8 cyclobutadiene character, 64f
selenium, tellurium and molecular parent antiaromatic compounds, 64f
addition compounds, early scientific symmetric cations, triplet ground
research, 9 states, 65f
tricoordinate carbon, 10 aromatic and antiaromatic molecules,
electrical conductivity, 66
aromaticity, 62

313
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 cyclopropenyl cation and departmental seminars, 112
cyclopropenone, 63f graduate school, 104
biomimetic chemistry and enzyme graduate student Havill, Martha, 101f
mimics, 67 granddaughter Kylie, 110f
catalysis by polymers, 69 MS in nuclear chemistry,
combat cancer, mechanism, 72 UC-Berkeley, 105
suberoylanilide hydroxamic acid, 73f Ph.D, 114
enzyme mimics, cyclodextrins, 68 Russell and Martha, married on June
enzyme mimics, reaction of steroids, 67 6, 1953, 100
formose reaction and prebiotic chemistry, Russell family, 106
70 Russell symposium, 109f
importance, 61 skiing trip, 116
my studies, 62 successful academic researcher, 117
physical organic chemistry, future, 73 Sun Valley, 107f
selective stereochemistry, creation, 71 Trahanovsky, Walter, 113
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ix002

some enzyme reactions, mechanisms, 70 wedding day, 102f

thiamine pyrophosphate, mechanism of Russell’s research, 118
catalysis, 66 semidione story, 124
thiazolium zwitterions, 66f acyloin, oxidation, 125f
transition state geometries, 69 anion radicals, 126f
ESR spectra, 127f
p-benzoquinone, one electron
reduction, 125f
R proton splitting, 129f
semidiones, early synthetic methods,
Russell, Glen A. 128f
carbanions, oxidation, 123
Russell’s suggested chain reaction,
124s
electron transfer, 130 S
unsaturated systems, electron transfer,
131f Schleyer, Paul
exploration, 133 from adamantane to dodecahedrane, 175
introduction, 93 Diels-Alder product, 177s
DPPH, resonance structures, 95f research facilities, Princeton, 176
Koelsch’s radical, resonance sulphuric acid treatment, 175s
structures, 96f applied spectroscopy and hydrogen
Russell, Glen, Iowa State faculty bonding, 185
member, 94f carbocation rearrangement mechanisms,
triphenyl methyl radical, resonance 178
structures, 95f computational chemist, beginnings, 191
Iowa State, early work, 121 early years, 170
NBS story, 122 epilogue, 195
organic synthesis, research, 132 Schleyer, Paul and Streitwieser,
Purdue, early research, 119 Andrew, 195f
Russell mechanism, 120f graduate student, 172
Russell’s life and career, 97 interminable norbornyl controversy, 189
1993 Gordon conference, 108f introduction, 169
1963 nobel prizes, 115 Schleyer, Paul, Erlangen, 170f
academic gowns, Russell, 99f move to Erlangen, 180
age 6 photo, 98f quantum chemistry, Pople, John, 192
associate professor, 103 solvolysis reactions, 187
career, overview, 111 undergraduate years, Princeton, 171
chemistry department parking lot, Solvation, thermochemical studies
110f physical organic chemistry, beginnings,
dance skills, Russell, 107f 49

314
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.
 Arnett group, typical solution complete thermodynamic analysis,
calorimeter, 53f problem 1, 53
organic chemistry, solvation, 51 preparation of o-di-t-butyl benzene,
organic reactions, elucidation, 50 schematic, 58f
solution calorimeters, 52 synthesis and strain energy, project 4,
solution calorimetry, application 57
adduct, presumed structure, 57f t-butyl chloride solvolysis, problem 3,
alkyl halide solvolysis, problem 2, 54 55
aqueous-ethanol solvents, mole
fraction water, 56f
Publication Date (Web): November 24, 2015 | doi: 10.1021/bk-2015-1209.ix002

315
In The Foundations of Physical Organic Chemistry: Fifty Years of the James Flack Norris Award; Strom, et al.;
ACS Symposium Series; American Chemical Society: Washington, DC, 2015.