J. Am. Chem. Soc.

1995, 117, 9357-9358 9357

Communications to the Editor

High-Precision Simultaneous Determination of proportion of a minor isotopic component in recovered material

Multiple Small Kinetic Isotope Effects at Natural compared to the original starting material (R/Ro) is related to
Abundance the fractional conversion of reactants (F) and the KIE (relative
rate for major/minor isotopic components) by eq l.6 As a
reaction approaches completion (F—T), RIRq approaches °», and
Daniel A. Singleton* and Allen A. Thomas KIEs become greatly magnified in the observable R/Ro. For
example, a KIE of 1.05 results in ~25% enrichment of a slower-
Department of Chemistry, Texas A&M University reacting isotopic component at 99% conversion. It has long
College Station, Texas 77843 been recognized that this ordinary effect of kinetic fractionation
can improve the precision of KIE determinations,7 but the degree
Received May 17, 1995
of improvement possible depends critically on the magnitude
Small heavy-atom and secondary hydrogen kinetic isotope of the KIE, the precision of the analytical technique ( R/Ro),
effects (KIEs) can provide detailed information about the and the uncertainty in F (AF).
mechanism of an exceptional range of chemical reactions.1·2
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
Downloaded via CALIFORNIA INST OF TECHNOLOGY on March 3, 2023 at 00:03:18 (UTC).

However, there are significant general limitations in methods R/R0

=
(1
-

F)(1/KIE)~‘ (1)
for the determination of these KIEs. Because absolute rate
measurements are rarely sufficiently precise, small KIEs are
usually determined in competition reactions of isotopically calcd
—
_1_J
labeled and unlabeled materials. This is possible only in systems ln[(l -

F)R/R0\
carefully chosen to allow the precise measurement of isotopomer
ratios with appropriate analytical techniques, such as scintillation
counting for 3H and l4C KIEs. The synthesis of isotopically 9KIE —ln(R/R0)
WTC
AKIEf =
—

AF
AE,_ =
AF (3)
labeled materials can be arduous, often prohibitively so, and a
new synthesis, competition reaction, and analysis are required
(1-/0 ln1(l -

F)R/R0]
for each KIE of interest. A broadly useful alternative, particu-
dKIE
larly for l3C KIEs, is to employ the high precision of isotope AKIES =
A(R/R0)
ratio mass spectrometry to study KIEs in materials labeled only d(R/R0)
at natural abundance.3 A major restriction is that each site of
interest must be selectively degradable without isotopic frac- -ln(l -

F)
A(R/R0) (4)
tionation into an analyzable small molecule, most often CO2. (R/R0) ln2[(l -

F)R/R0]
The isotope- and position-specific information inherent in
NMR techniques seems ideally suited to measuring KIEs at
The uncertainties in calculated KIEs (eq 26) due to AR/Ro
natural abundance. The utility of 2H NMR for determining large
2H KIEs at natural abundance has been established,4 and in
( ) and AF ( / ) are shown in eqs 3 and 4, respec-
tively.8 Most commonly, either the KIE is large or the
theory, all of the individual KIEs in reactions of complex natural measurement of R/Ro is highly precise (AR/Ro small). In these
abundance materials can be determined simultaneously!5 In
cases, the uncertainty in the KIE is dominated by / (not
practice, however, NMR quantitation has not been sufficiently
counting any systematic error), and no advantage is gained at
precise to be useful with small KIEs, the uncertainty in the few
cases tried generally rivaling or exceeding the size of the isotope
high conversion. However, for the relatively low precision of
NMR integrations, the uncertainty in the KIE is dominated by
effects.4·5 We report here a simple general method for attaining
/?, which decreases greatly as F increases, and /· is
chemically significant precision while simultaneously measuring
all of the KIEs for reactions at natural abundance. nearly negligible. For example, with a KIE of 1.02 and a 1.0%
±0.1 measurement of unreacted starting material (1 F = —

As any reaction proceeds, the starting materials are fraction-

0.01 ± 0.001), / is only 0.0004, and a 1.5% uncertainty
atively enriched in isotopically slower-reacting components. The in NMR integrations results in an uncertainty (« /?) of only
(1) For some varied recent applications of heavy-atom isotope effects, 0.003 in the KIE.
see: Burger, R. M.; Tian, G.; Drlica, K. J. Am. Chem. Soc. 1995, 117, To test this idea, the Diels—Alder reaction of isoprene with
1167. Jankowski, S.; Quin, L. D.; Paneth, P.; O’Leary, . H. J. Am. Chem.
Soc. 1994,116, 11675. Holm, T. J. Am. Chem. Soc. 1994,116, 8803. Zhang,
maleic anhydride (eq 5) was studied. The relative proportion
Y.; Bommuswamy, J.; Sinnott, M. L. J. Am. Chem. Soc. 1994, 116, 7557.
Kupczyk-Subotkowska, L.; Saunders, W. H., Jr.; Shine, H. J.; Subotkowski,
W. J. Am. Chem. Soc. 1993, 115, 5957. Yamataka, H.; Nagareda, K.; 25 »C h3C-
Takatsuka, T.; Ando, K.; Hanafusa, T.; Nagase, S. J. Am. Chem. Soc. 1993, "
115, 8570. Rao, M.; Barlow, P. N.; Pryor, A. N.; Paneth, P.; O’Leary, M. xylenes
H.; Quinn, D. M.; Huskey, W. P. J. Am. Chem. Soc. 1993, 115, 11676. O
(2) Some recent applications of secondary 2H or 3H isotope effects:
Balasubramanian, S.; Coggins, J. R.; Abell, C. Biochemistry 1995,34, 341.
Ankianiec, B. C.; Christou, V.; Hardy, D. T.; Thomson, S. K.; Young, G. of 2H and 13C at the various positions of isoprene, compared to
B. J. Am. Chem. Soc. 1994, 116, 9963. Barta, N. S.; Kirk, B. A.; Stille, J. the original starting material, could be readily determined by
R. J. Am. Chem. Soc. 1994, 116, 8912. Bauld, N. L.; Yueh, W. J. Am.
Chem. Soc. 1994, 116, 8845. Gajewski, J. J.; Brichford, N. L. J. Am. Chem. 2H and 13C NMR9 using the methyl group as an “internal
Soc. 1994, 116, 3165. O’Hair, R. A. J.; Davico, G. E.; Hacaloglu, J.; Dang, standard” on the assumption that its isotopic composition does
T. T.; DePuy, C. H.; Bierbaum, V. M. J. Am. Chem. Soc. 1994, 116, 3609.
Fitzpatrick, P. F. J. Am. Chem. Soc. 1994, 116, 1133. Nagorski, R. W.; (6) Melander, L.; Saunders, W. H., Jr. Reactions Rates of Isotopic
Slebocka-Tilk, H.; Brown, R. S. J. Am. Chem. Soc. 1994, 116, 419. Molecules; Wiley: New York, 1980; pp 95-102.
(3) O’Leary, . H. Methods Enzymol. 1980, 64, 83. (7) Raaen, V. F.; Ropp, G. A.; Raaen, . P. Carbon-14; McGraw-Hill:
(4) Pascal, R. A., Jr.; Baum, M. W.; Wagner, C. K.; Rodgers, L. R.; New York, 1968; pp 57-59. Bigeleisen, J.; Wolfsberg, M. Adv. Chem.
Huang, D.-S. J. Am. Chem. Soc. 1986, 108, 6477 (see especially footnote Phys. 1958, 1, 15. Ropp, G. A.; Danby, C. J.; Dominey, D. A. J. Am. Chem.
9b). Soc. 1957, 79, 4944. Jones, W. M. J. Chem. Phys. 1951, 19, 78.
(5) Zhang, B.-L. Magn. Reson. Chem. 1988, 26, 955. (8) See ref 6, pp 319—322.

0002-7863/95/1517-9357$09.00/0 © 1995 American Chemical Society

9358 J. Am. Chem. Soc., Vol. 117, No. 36, 1995 Communications to the Editor

expectations, prior experimental results, and theoretical calcula-

tions. The KIEs for ¿2, C3, and H3 are very small, as would be
expected for nonreacting centers. The individual 2H KIEs for
the 1 and 4 positions of isoprene may be compared with
aggregate KIEs from previous Studies of the reactions of
[1,1,4,4-D4]butadiene or [1,1,4,4-D4]isoprene with symmetrical
dienophiles.1415 At 25 °C, previous studies found KIEs per
deuterium (=KIE025) of 0.935,14 0.943,15 0.948,15 and 0.938,15
compared to a geometric mean of 0.942 in the current study. It
is notable that there have been no previous studies of either
l3C or l4C KIEs for the diene in a Diels—Alder reaction.16
However, the l3C KIEs for Ci and C4 are in line with recent
predictions of l4C KIEs for the concerted Diels—Alder reaction
of butadiene with ethylene based on theoretical calculations.17
The same calculations predicted that significantly more pro-
nounced 2H isotope effects would be observed for a diene’s
“inside” hydrogens (H]Z and H4Z) than for the “outside”
hydrogens (Hie and Hie). The results experimentally verify
that prediction.
The more pronounced KIEs for 2H substitution on C; over
C4 indicate some asynchronicity in bond formation to Ci versus
C4 at the transition state. In Gajewski’s elegant study of the
reaction of [4,4-Da]- and [l,l,4,4-D4]isoprene with symmetrical
dienophiles, some of the KIE results had also suggested
asynchronicity but were always consistent, within 2—5% error
Figure 1. (a) 2H and l3C isotopic composition of isoprene recovered limits, with a perfectly synchronous mechanism.15 The indica-
from a reaction taken to 98.9% completion, relative to starting isoprene, tion of asynchronicity in the current results is well outside of
with standard deviations in parentheses (n =11 (3 samples) for 13C experimental error.
data; = 5 (2 samples) for 2H data), (b) 2H and l3C KIEs (kn/ko and The major limitation on the determination of KIEs by this
btc/k<>c) calculated from the results in (a) and eq 2. method is that the reaction must be sufficiently scalable to allow
recovery of adequate starting material for the 2H NMR analysis.
not change.10 As the reaction proceeds, the relative proportion The method also requires that the starting material be isolable
of 13C at positions 1 and 4 of isoprene increases, and the from a large amount of product. The reaction of interest must
proportion of deuterium in these positions decreases. Isoprene be irreversible, and the reaction mechanism must not change
recovered by distillation* 11 from a reaction taken to 98.9(1)%
over the course of the reaction. However, the KIEs should be
completion12 was analyzed in detail, and the results are shown unaffected by secondary conversions of the products. This
in Figure la. The 2H and l3C KIEs (kn/ko and ¿i2C/ki3C) could
method should also be relatively insensitive to impurities in the
be calculated from eq 2 and are shown in Figure lb.
The KIEs determined in this manner are consistent with KIEs analyzed material. Indeed, it is completely insensitive to
estimated at lower conversion (with concomitantly decreased impurities unless they overlap in the NMR, and a 2% natural
abundance impurity causes only ~0.004 error in a KIE measured
precision).13 Within experimental error, the calculated KIEs
are independent of F. They are also in line with general
at 99% conversion. It should be noted that such contamination
in the 2H NMR can be detected from a careful NMR
(9) As described in the supporting information, a number of precautions integration.
were taken to minimize both random and systematic errors in the NMR
analysis. See: Rabenstein, D. L.; Keire, D. A. In Modem NMR Techniques By this method, a level of detailed KIE information often
and Their Application in Chemistry·, Popov, A. I., Hallenga, K., Eds.; Marcel unattainable in conventional ways can be quickly obtained with
Dekker: New York, 1991; pp 323—369. Samples were prepared identically,
a T¡ determination was carried out for each sample, and the results from a
routine instrumentation, and the determination of KIEs can
single sample that exhibited significantly different TVs were excluded. 13C become a routine part of the characterization of chemical
spectra were obtained with gated decoupling and a 63 s delay between reactions. We are currently applying this technique to a wide
calibrated 2 /9 pulses. Variations in the data workup (integration regions,
baseline correction methods, phasing), when applied consistently for both variety of classical and topical reactions.
original and recovered material, did not significantly affect the calculated
KIEs. The major source of variability in integrations appeared to be random
effects from signal-to-noise limitations and baseline variations. Acknowledgment. We thank the NIH and The Robert A. Welch
Foundation for support of our programs.
(10) Because of this assumption, the resulting KIEs are “relative”, erring
systematically to whatever extent the assumed 2H and l3C KIEs for the
methyl group are incorrect. A small systematic error would not affect the Supporting Information Available: Procedures for reactions, NMR
interpretation of the current results. In cases where a reasonable internal
standard is absent, the isotopic composition can be referenced to an external measurements, and calculations (3 pages). This material is contained
standard (see ref 5) or, better, groups introduced by chemical derivatization. in many libraries on microfiche, immediately follows this article in
(11) Because the NMR analysis compares relative amounts of mono- the microfilm version of the journal, and can be ordered from the ACS,
deuterated or mono-l3C materials, the effect of isotopic fractionation in the and can be downloaded from the Internet; see any current masthead
reisolation should be negligible.
(12) Based on a determination of 1.1(1)% remaining isoprene by GC page for ordering information and Internet access instructions.
analysis versus heptane as an internal standard. JA9516017
(13) The calculated l3C KIEs for reactions taken to 72(2), 81(2), 90(1),
95(1), and 96.0(0.4)% conversions were, respectively: for Ci, 1.022(7),
1.025(8), 1.019(4), 1.020(5), and 1.023(4); forC2, 0.999(9), 1.000(6), 1.001- (14) Van Sickle, D. E.; Rodin, J. O. J. Am. Chem. Soc. 1964, 86, 3091.
(5), 1.000(3), and 1.000(4); for C3, 1.007(8), 1.004(4), 0.999(4), 1.004(3), (15) Gajewski, J. J.; Peterson, K. B.; Kagel, J. R.; Huang, Y. C. J. J.
and 1.001(4); for C4, 1.020(8), 1.017(8), 1.014(5), 1.017(5), and 1.019(3), Am. Chem. Soc. 1989, 111, 9078.
respectively. The 2H KIEs for reactions taken to 81(2) and 90(1)% (16) For a study of l4C KIEs in the dienophile, see: Kupczyk-
conversions were, respectively: for Hie, 0.961(9), 0.966(13); for Hiz, 0.909- Subotkowska, L.; Shine, H. J. J. Am. Chem. Soc. 1993, 115, 5296.
(13), 0.899(10); for H3, 0.982(19), 0.994(9); H4£, 0.968(12), 0.959(9); H4Z, (17) Storer, J. W.; Raimondi, L.; Houk, K. N. J. Am. Chem. Soc. 1994,
0.940(11), 0.947(10). 116, 9675.