Comments on Inorganic Chemistry

ISSN: 0260-3594 (Print) 1548-9574 (Online) Journal homepage: http://www.tandfonline.com/loi/gcic20

Assigning Reactive Excited States in Inorganic

Photochemistry

Arthur W. Adamson

To cite this article: Arthur W. Adamson (1981) Assigning Reactive Excited States
in Inorganic Photochemistry, Comments on Inorganic Chemistry, 1:1, 33-45, DOI:
10.1080/02603598108078078

To link to this article: http://dx.doi.org/10.1080/02603598108078078

Published online: 19 Dec 2006.

Submit your article to this journal

Article views: 14

View related articles

Citing articles: 9 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=gcic20

Download by: [Nationwide Childrens Hospital] Date: 14 October 2017, At: 10:37
 Assigning Reactive Excited States in Inorganic
Photochemistry
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

The matter of assigning reactive excited states in inorganic photochemistry turns out to have
complexities as more detailed and more varied information is obtained. Three cases are
considered. For Cr(1II) ammines there is still much controversy in the assignment of chemical
reactivity to doublet and quartet states. In the case of Rh(NH3)5X2+complexes, it may be
necessary to invoke three different reactive or emitting ligand-field excited states, while for
W(CO),L species, at least two, including a charge transfer state, are needed.

Introduction

We are seeing currently the development of an extended research field for that
special breed of physical inorganic chemist, the excited-state kineticist. Studies
of excited-state rate processes are an increasingly important adjunct to
conventional quantum yield and product characterizations. Increasingly
complex and intimate excited-state reaction schemes are being constructed.
The identification of the reactant species, ordinarily obvious in thermal
reactions, turns out not to be so obvious in excited-state chemistry. We limit
ourselves here to three examples involving transition metal complexes in
solution.
The general picture, and some vocabulary, should be presented first. In the
case of mononuclear complexes having monodentate or simple bidentate
ligands, to which this discussion will be confined, the visible U V absorption
spectra show two principal types of transitions, ligand field (LF) and charge
transfer (CT). There may be several LF absorptions, and we designate these as
L,, L,, etc., with a left superscript to indicate the spin multiplicity of the
terminal state. Thus the 4A2g+4T2g transition for a d 3 Ohcomplex is labelled
4L,. Already there is a difficulty; spin and orbital angular moment increasingly
mix on going to the heavier transition metals, and the use of spin-only
designations, while convenient, is suspect.
LF bands are usually broad, and it is now generally accepted that the optical
Comments Inorg. Chem. 0 Gordon and Breach Science Publishers, Inc., 1981
1981, Vol. 1, pp. 3 3 4 5 Printed in Great Britain
026Cr-3594/81/0101-0033$06.50/0

33
 transitions are Franck-Condon in type, so that absorption of light gives a
distribution of vibrationally excited molecules, which rapidly relax or
thermally equilibrate to the temperature of the medium. The broad absorption
band is essentially the Franck-Condon envelope for the transition. The thexi
state, as we will call the thermally equilibrated excited state, may have not only
different metal-ligand bond lengths, but also different bond angles.’ The
classic indication of such distortion is the presence of a very large Stokes’ shift
in those Cr(II1) complexes that show f l ~ o r e s c e n c e .Similar
~ ’ ~ broad emission
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

bands and larges Stokes’ shifts are observed for spin forbidden transitions as
well, although the d3 case is an exception. In the 2Ll transition, there has
merely been rearrangement of electrons in the t,, set of orbitals; these are non-
bonding, and little excited-state distortion occurs so that the phosphorescence
spectrum is narrow and vibronically structured, and not much displaced from
the ’L, band.
Because of the distortion problem, we will avoid the use of orbital symmetry
symbols in labelling L F thexi states. Rather, we designate them by their
nominal spin multiplicity: S (singlet),D (doublet), T (triplet), and Q (quartet or
quintet), with a right subscript to give the energy ordering, and a right
superscript zero to denote thermal equilibration. The ,T,, ligand-field state is
labelled QFC(as a Franck-Condon state), and becomes Q10 after thermal
equilibration. States related by descent of symmetry from 0, may be
designated by primes.
Again as noted earlier,’ thexi states are good thermodynamic species. An
ensemble of such a species has entropy, free energy, and a standard redox
potential, as well as energy. It is a topological isomer of the ground state, just
as square planar and tetrahedral ML, complexes are isomers. The thexi state
is a good kinetic species. Its reactions can be activated, stereospecific, subject
to ionic strength effects, etc. Its intimate reaction mechanism should be
treatable by conventional theories for rate processes. The identification of the
reactive thexi state in inorganic photochemistry is thus a matter of serious
chemistry.
The typical excited-state processes with which we deal are illustrated in
Figure 1, for the d 3 system. QFc may thermally equilibrate to QlO,or may
undergo prompt intersystem crossing (pisc), to D,’. The thexi states may exit
by emission, of rate constant k,, by non-radiative return to the ground state, k,,
and kh,, or by chemical reaction, k,, and khr They may interconvert by
intersystem crossing (isc) and back intersystem crossing (bisc).A non-classical
chemical reaction is that which occurs during non-radiative relaxation -the
so-called hot ground-state reaction. These have been very difficult to establish,
and will not be considered here.
We turn now to specific systems to examine what progress has been made in
identifying reactive states and in obtaining actual rate constant values.
34
 Cr(ZZZ)Ammines

The photochemistry, essentially substitutional, has been studied exten-

~ - ~interesting situation is that after some 20 years of activity in the
~ i v e l y . The
field, there are still questions as to the roles of D, ', QI0,and, in the case of non-
0, complexes, any Q;'. An early supposition was that all reaction was from
D10;7,8 it was known as relatively long-lived state, typically with millisecond
phosphorescence lifetimes at 77 K, and the spin pairing could free an orbital to
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

facilitate a concerted substitution process. In seeking to test the doublet

hypothesis, we came to the conclusion that the Q 1 0state could not be ignored,
and produced the photolysis rules:'
1) Consider the six ligands to lie in pairs at the ends of three mutually
perpendicular axes. That axis having the weakest average crystal field will be
the one labilized, and the total quantum yield will be about that for an 0,
complex of the same average field.
2) If the labilized axis contains two different ligands, then the ligand of
greater field strength preferentially aquates.
The rules are approximately obeyed, although there are e x ~ e p t i o n s , ~ ~ ~ ~ "
and make theoretical sense if Q10is the reactive state. From the ligand-field
point of view, an electron has been promoted to an e, antibonding orbital and,
in a non-0, complex, the antibonding axis could be expected to be the one for
which the average ligand strength was the weaker. In Figure 1, provided that
distortion has not significantly disturbed the octahedral framework, Q1
would usually be the axially labilized thexi state in a Clvcomplex, and Q1'', the
equatorially labilized one. More quantitative and more elaborate ligand-field
explanations have been made which adequately predict both the rules and the
exceptions to them.''-" At one point, it became widely thought that all
reaction occurred from QIo,but now the pendulum is swinging back a little.
Complexes of the type Cr(NH&XZ+ show two photoreaction modes,
ammonia aquation and aquation of the X - group. The absorption spectrum
for X - = NCS- is shown in Figure 2; this case provided an early indication
that both D,' and Q1' are r e a ~ t i v e , ' ~both
" ~ in that the ratio of reaction
modes changed with wavelength, and from sensitization results. A point of
relevance later is that the ammonia photoaquation, the rules-predicted mode,
is antithermal, i.e., the labilized ligand is not the thermally reactive one. In the
case of trans-Cr(en),(NH,)(NCS)' +, three reaction modes have been re-
ported,15 aquation of one end of an ethylenediamine ligand, ammonia
aquation, and (presumably) thiocyanate aquation. Are the excited states just
being sloppy or are three different ones now involved?
'
A new leverage became available when it was found that D, emission could
be observed in room temperature solutions with lifetimes now in the micro- or
35
 t
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

Distortion ---+

FIGURE i Energy vs. distortion diagram for a d 3 system. Bars locate thexi states, indicated in
square frames. Vertical lines denote radiative and wavey lines, non-radiative processes. The light
horizontal lines indicate successive vibrational wells as the solvent cage adjust to geometry
changes (for clarity, shown only for the thermal equilibration of QFC),l

200 300 400 500 600 700

A, nm
FIGURE 2 Absorption spectrum for Cr(NH,),(NCS)Z t . 5

nano-second range (indicating that k , was no longer important).16- l 9 The

emission could be quenched,6 and an important observation was that, on
quenching the emission, much of the photochemistry is also quenched in the
cases of trans-Cr(NH,),(NCS); 2o and Cr(en): + , l7 , 2 1 ~ 2 2 but not for
Cr(CN)2-.23D,' is clearly implicated in the first two cases, and the simplest
36
 explanation of D,' involvement is that it is chemically reactive and furnishes
part of the overall quantum yield. Alternatively, however, promptly formed
D,' could be returning to Q , by bisc, so that all reaction is still from Q , '.
In an effort to probe these alternatives, we monitored the rate of appearance
of primary photoproduct, in the case of Cr(en);+, and found that indeed 30'X of
photoproduct, Cr(en),(enH)(H,0)4+, appeared in less than a few nano-
seconds, with the remainder growing in with the D,' emission lifetime.24
Because of ground-state bleaching, it was also possible to determine the
'
efficiency of D, formation and thence &, the efficiency of the 'slow' product
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

formation. This was significantly greater than the overall quantum yield,
indicating that chemical reaction was indeed occuring from D,' as well as
from QIo,see further below:
Further support for D,' as a reactive state has come from emission rules for
room temperature solution^,'^ of which the second is:
Rule 2 If two different kinds of ligands are coordinated, the emission lifetime will be relatively
short if that ligand which is preferentially substituted i n the thermal reaction lies o n the weak-
field axis of the complex.

Implied in the rule is that the emission lifetime, z, is determined mainly by k,,,
rather than by k,, or kbisc.A rationale for the rule is that the reactivity of D,'
tends to parallel that of the the groud state; that the paralleling is in reaction
rate, as well as in reaction mode, is indicated by some more recent
observations. *
Our hypothesis at this point is that where the photolysis rules predict the
thermal reaction mode [as for any 0, complex and for trans-
Cr(NH,),(NCS), -], the photoreaction may partly be from Q,' and partly
from D,', the latter being the quenchable portion. Where the photolysis rules
predict an antithermal reaction, this is from Q,', while any thermal reaction
mode present is from D,'.
At present there is no consensus, and some specific contentions on the
matter of D,' reactivity.17,21,22.26-30 In addition, an alternative attempt to
account for two reaction modes has been to invoke a Q1", with ligand-field
analysis as to how axial and equatorial labilization should behave."

The Case of Cr(en); +

We turn now to the specific case of aqueous Cr(en):+ to see how emission life
time measurements can be helpful in assessing D,' reactivity (see also
Gutierrez and Adamson3' for the case of trans-Cr(NH,),(NCS), 32). If our
interpretation is correct l/z gives kcr; on the alternative hypothesis, 1 / ~is
related to kbisc.
37
 We are dealing with a coupled reaction scheme, Eq. (1):

Following an excitation pulse producing unit concentration of QFC,pisc

occurs with efficiencyf, so that at small times (on the nanosecond time scale),
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

we have (DlO)’=fand (Ql0)O=(1 -f).The exiting rate constants are k2, =k,
+ +
+ k,, k,, =k,, + k,,, and k3, = k: kb, + kh, . 7 kkr
~ + kh,, and the efficiency of

chemical reaction from D, is cpb = k,,/k2 1, while that from Q1 is pb = khr/k3,.

The intersystem crossing rate constants are denoted by k23 (bisc)and k32 (isc).
In the case of quenching of D l o emission, the exiting rate constant k,, is
augmented by the term k,(A), where A is the acceptor or quenching species.
The solution of Eq. (1)has been and we proceed to the aspect
of interest here. The rate at which exiting occurs from D,’ is k2,(DI0),and the
total exiting up to time t is

ED=k21](D10)dt.
0

Analysis gives

where

and ,Il and A2 are the two observable decay constants which maybe
measured.34 Similarly, the total exiting from Q I 0is given by

38
 TABLE I
Excited state kinetics for Cr(en):+
- __ ~ -~ ~ _ _ _ ~_ _
_ ~_

1 0 01 1000 087 026

~
03 - 0297 0700 0003 - 030
001 1000 I01 016 011
I 0 01 1000 029 026
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

- - 0 89 0881 0110 0009 ~ - 088

001 1000 101 100 011
I 429 I429 087 026
- ~

00 0298 0700 0002 .- 0 30

001 1000 101 016 011
001 001 1000
4 - - 07 - 0007 0300 0694 - 026
- 070
1 1000 101 037 011
-~
The quantity A, Eq. (3,is always a small number in these regimes.

If I l and I, correspond to the positive and negative roots, respecitvely, of Eq.

(3), then it follows from Eq. (3) that I,,<p,<I,. Eq. (2) and (4)may be
abbreviated thus

and the total yield of photoproduct, cp, is just E,+E,.

In the present case, 1, is a large number and 2, is unity, if time is measured in
units of 1.8 psec (since the experimental emission lifetime given by 1/12is 1.8
psec). At a time small compared to 1/2,, but large compared to 1/11, we have

Further product formation then grows in, the eventual additional yield being

The quenchable fraction of the total yield is ~p,~,,/cp. Experimentally, cp =0.37,5

cprast=O.l1, and cpSlOw= 0.26.In addition, our monitoring results gave f=0.30
[assuming that (D,')' is formed by pis^].^^
Suppose, as regime Ia, that kbisc= 0. A possible set of k values is given in case
1 Table I. Case 2 is for the maximum possible f value, if cpfastis not to drop below
0.11 and cpb is at its minimum allowable value of 0.29, which means that kcr/(kcr
39
 + kn,)= 0.29 or k,,= 2.4 kcr.The emission lifetime is given essentially by l / k z 1
= l/(knr+ k,,), and we note that experimentally z showed linear Arrhenius plots
with an apparent activation energy of 10 kcal mol-' Good Arrhenius
behavior seems unlikely if two quite different kinds of rate constants are
making comparable contributions. Case 1, with q', = 0.87 or, were f slightly
smaller, with q',= 1, is more acceptable in this respect. Analysis of this regime
thus suggests that 5 is indeed controlled mainly by kcr.
Regime Ib is one of which (D, ")" appears via isc, ,f being zero. Case 3 in Table
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

I gives an acceptable calculation in terms of the observables, but a numerical

difficulty arises. If D," is to appear within a few picoseconds, as it apparently
k 3 2 must be about 10" sec-' and k,,must also beofthis magnitude
"
if qrastis to be as large as 0.11. The energy difference between Q1 and D l ohas
been estimated to be about 13 kcal mol-1.37 Neglecting entropy contri-
butions, k,,/k,, = 3 x lo-'" so that k,, =30 sec-', an uncomfortably small
value. Regime Ia is to be preferred over Ib, but the latter cannot be ruled out.
In regime IIa, we suppose that k,,= 0 ( k Z l= k,,), so that all chemical reaction
is via bisc and QIo;this is the alternative hypothesis. The emission lifetime is
now related to k,,, which is therefore taken to be about unity in case 4 of Table
I. This case is acceptable as to qfaSt and qqlowr but requires the yield of (DlO)"to
be 0.7 contrary to our observation of 0.3; no lower value for this yield can be
found in this regime. There are again numerical difficulties. The regime
requires that k,, not be much greater than k , , since k,, is a wasting process.
This is unlikely since both reactions are non-radiative transitions, but with less
geometry change for k,, than for k,,. In addition, little activation
energy is expected for k,,, while k , , must now be assigned the 10 kcal mol-
barrier obtained from the temperature dependence of z.Also, since k23 N l/z
~ 0 . x5 lo6 sec-', we find k , , . ~ 0 . 5x 106/3x 10-'O--2 x l O I 5 sec-'. This is
much too fast. Even if the energy gap between Q10and D I ois reduced to the
minimum value of 10 kcal mol- I (set by the temperature dependence of T ) , k , ,
is still about 1 x 10' sec-', which in turn implies an equally uncomfortably
large value of k34 Regime IIb, in which (D,')" is produced by isc rather than
by pisc, runs into similar difficulties.
Although the above presentation has been sketchy, it illustrates the point
that quantitative considerations can limit the type of kinetic scheme that is
acceptable. In the case of Cr(en);+, the result is a distinct favoring of regime Ia,
namely chemical reaction from D l 0 formed by pisc.

R h(N H 3 ) 5 X z Complexes
+

We take up briefly two more types of complexes. The aqueous photochemistry

of Rh(II1) ammines is mostly s ~ b s t i t u t i o n a l- .4~1 ~ Rh(NH,)5C12 photo-
+

40
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

Rh(NH,), CI2+ Rh(NH,), Br2+

FIGURE 3 Energy vs. distortion diagram for Rh(NH,),X2 +. (a) Two reactive or emitting states.
(b)Three reactive or emitting states.

aquates C1- with q=0.16 at 350 nm, while Rh(NH,),BrZf shows both
ammonia photoaquation, (qNH, = 0.18) and bromide aquation (qBr= 0.019).42
Emission from aqueous solutions has recently been o b ~ e r v e d ~and , . ~may
~ be
quenched by O H - On quenching the emission, we found that (pa was
85% quenched, in the case of Rh(NH,),C12f,45 and that qNH,was fully
quenched, but qBrnot at ull in the case of Rh(NH3)5Br2+.46
An interesting and important problem is now posed. Conventional wisdom
assigns the emitting state as a triplet and since the emission spectra and
lifetimes are very similar for the two complexes, it would seem that it is the
same triplet state in both cases, which we call TIo.A simple explanation of the
quenching results is that T I ois chemically reactive, but if this is so, then Rh-Cl
bond-breaking occurs in the one case and Rh-NH, bond-breaking in the
other. The bromide mode of reaction can then be assigned to some other state,
say T, O', as illustrated in Figure 3a. Although ligand-field rationalizations can
be made, it seems awkward to ask that the chemical reaction mode makes so
complete a change between X = C1- and X = Br -.
41
 An alternative scheme, shown in Figure 3b, invokes three excited states. The
bromide yield is assigned to a quintet state, QIo, and the quencbable
photochemistry to T,' or to T,", whichever is the lower lying, but with the
emission from T I nin both cases. [The quintet state is not necessarily high in
energy in CdV ligand-field theory.47] Q1'is placed with less distortion than T, '
because of its more symmetric arrangement of antibonding electrons and
might be similar to D,' in d3 systems in having ground state-like reactivity.
At present, no decision seems possible between the two and the three
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

reactive state schemes. The former has been ~uggested,~' but from data that do
not rule out the alternative.

W(CO),L Complexes

Another d6 situation is that of group VI carbonyl compounds. W(CO),L

complexes, where L is a n-electron donor, undergo photodissociation of the L
ligand and, in the presence of a second ligand, L', the intermediate W(CO), is
scavenged to yield W(CO),L'.49,50The quantum yield for such reactions is
ordinarily several tenths, but as L becomes more electron-withdrawing, a CT
band moves to the long wavelength side of the first LF band and the yield
drops to around 0.0'15'
For the case of L =4-cyanopyridine, we observed emission at room
temperature in methylcyclohexane solution with 0.1 M ethanol present as L'
(as well as without the ethanol).52 The emission was quenchable by
anthracene, for example, and on quenching the emission, the photoproduction
of W(CO), (ethanol) was also quenched. Clearly the emitting state, possibly a
triplet charge transfer state, %T, is implicated in the photochemistry. We can
+
write cp = k,,/(k, k,,+ kcr)for this state and, since cp is small, cp N k,,/k,,; k, is
taken to be negligible. The temperature dependence of cp gives an apparent
activation energy of 7.6 kcal mol-', so if 3CT is both the reactive and the
emitting state, 7.6 = E,*,- E,U,. In the one-reactive state scheme, Figure 4a, l/z
+ +
= ( k , k,, kcr)N k,,, and from the temperature dependence of z, E,*,N 1.5 kcal
mol-', whence E2=9.1 kcal mol '.This may be high for what is thought to
be a simple bond dissociation reaction.
An attractive alternative is a two-state scheme in which chemical reaction
occurs from a LF state lying above the emitting CT one, and in steady-state
equilibrium with it, as shown in Figure 4b. The observed E$ is now attributed
primarily to the energy difference, AE, between the two states. In this scheme,
excitation leads through intersystem crossings to 3CT. This state is emitting
but not highly chemically reactive. The higher LF state, presumably TIo,does
react efficiently, but the overall quantum yield is low because of the
competition with non-radiative relaxation of 3CT. If L is such that TIois the
42
 isc)
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

(a) (b)
FIGURE 4 Energy vs. distortion diagram for W(CO),L. (a) Reactive and emitting state the
same. (b) Emission from 3CT and reaction from back-populated T,'.

lower lying state, then the quantum yield should be large, as observed.
. ~ a~ similar
The two-state scheme has been f a ~ o r e d ; ~ ~also . ~ ' one has been
proposed for Ru(NH,),L2 ' c ~ m p l e x e s .Quantitative
~~ considerations in-
dicate at least some caution, however. In the two-state scheme, kblscmust
compete with k,,, and the emission lifetime of 360 nsec at 25°C gives k,, 'v 2.8
x lo6 sec-'. We have AE=E,*+E,*,-E,*,.=9.1-E,*,., where E,*,. is the
activation energy for reaction from TIo,which is probably small. If we take AE
to be the full 9 kcal mol-', and suppose that kbisc=O.l k,,, then 2.8 x lo6
whence Abisc= 1 x 10" sec-'. No allowance has been
x 0.1 =Abisc e-9000'RT,
made for entropy change, and this Abiscvalue, while marginally acceptable, is a
bit large. Further study of the W(CO),L complexes is needed.

Summary and Conclusion

The three cases described here illustrate one of the problems for the excited-
state kineticist. His colleagues who deal with ground-state reactions usually
know what the reactant species is. The photochemist spends much effort in
trying to find out what his reactants actually are, and in few cases so far has this
effort been 'unambiguously successful. That is, while reasonable guesses
provide good working hypotheses, it has been a difficult matter to be sure
whether one is dealing with a one reactive state scheme, or a two or three
reactive state one. Determining the thexi-state spin multiplicity and its actual
structure is yet more difficult. Fast magnetic susceptibility methods may help
on the former question, and also photochemistry in high magnetic fields. The
structure problem may yield to excited-state resonance Raman
spectroscopy.
43
 An important reason for establishing at least the number of reactive thexi
states in a given system is that ligand-field theoreticians have been interested in
explaining thexi-state reactivity in terms of specific metal-ligand bond
labilization leading to dissociation and a five-coordinate intermediate." - l 2
In such analyses, it makes a difference in d3 if the state is D l oor QI0,and, in d6,
whether it is Slo, TI*, QI0, or 'CT. It can be embarrassing to provide a
theoretical explanation for the wrong scheme! The rnechunisrn of thexi-state
reactions has not been a focus of this paper. My personal opinion, however, is
that the solvation reactions are more likely to be concerted with solvent rather
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

than being limitingly dissociative in type.

Acknowledgement

Our unpublished work described here was supported in part by the US National Science
Foundation and the Office of Naval Research.

ARTHUR W. ADAMSON
Depurfment of Chemistry,
University of Southern Cali$ornia,
Culifornia 90007

References

1. A. W. Adamson, Adv. Chem. 150, 128 (1976).

2. G. B. Porter and H. L. Schlafer, Z. Physik. Chem. 37, 109 (1963).
3. A. D. Kirk and G . B. Porter, J. Phys. Chem. 84, 887 (1980).
4. V. Balzani and V. Carassit, Photochemistry of Coordinution Compounds, (Academic Press,
New York 1970).
5. E. Zinato in Concepts of Inorganic Photochemistry, ed. by A. W. Adamson and P. F.
Fleischauer, (John Wiley and Sons, Inc., 1975).
6. V. Balzani, L. Moggi, M. F. Manfrin, F. Bolletta and G . S. Lawrence, Coor. Chem. Rev. 15,
321 (1975).
7. H. L. Schlafer, Z. Physik. Chem. 11, 65 (1957).
8. R. A. Plane and J. P. Hunt, J. Am. Chem. SOC.79, 3343 (1957).
9. A. W. Adamson, J. Phys. Chem. 71, 798 (1967).
10. (a) J. 1. Zink, J. Am. Chem. SOC.94, 8039 (1972). (b) N. J. Incorvis and J. I. Zink, Inorg.
Chem. 13, 2489 (1974).
11. M. Wrighton, H. B. Gray and G . S. Hammond, Mol. Photochem. 5, 164 (1973).
13. (a) L. G. Vanquickenborne and A. Ceulemans, J. Am. Chem. SOC.100,475 (1978). (b) idem,
Inorg. Chem. 17, 2730 (1978).
13. E. Zinato, R. D. Lindholm and A. W. Adamson, J. Am. Chem. SOC.91, 1076 (1969).
14. J. E. Martin and A. W. Adamson, Theor. Chim. Acta 20, 119 (1971).
15. A. D. Kirk and T. L. Kelly, Inorg. Chem. 13, 1613 (1974).
16. N. A. P. Kane-Maguire and C. H. Langford, J. Chem. SOC.Chem. Commun., 895 (1971).
17. R. Ballardini, G. Varoni, N. F. Wasgestian, L. Moggi and V. Balzani, J. Phys. Chem. 77,
2947 (1973).

44
 18. A. W. Adamson, C. Geosling, R. Pribush and R. Wright, Inorg. Chim. Acta 16, L5 (1976).
19. R. T. Walters and A. W. Adamson, Acta Chem. Scand. A33, 53 (1979).
20. S. Chen and G. Porter, Chem. Phys. Lett. 8, 41 (1970).
21. N. A. P. Kane-Maguire, D. E. Richardson and C. G. Toney, J. Am. Chem. Soc. 98, 3996
(1976).
22. N. A. P. Kane-Maguire, J. E. Pfifer and C. G. Toney, Inorg. Chem. 15, 593 (1976).
23. H. F. Wasgestian, J. Phys. Chem. 76, 1947 (1972).
24. R. Fukuda, R. T. Walters, H. Macke and A. W. Adamson, J. Phys. Chem. 83,2097 (1979).
25. A. W. Adamson, H. Macke, J. P. Puaux, E. Zinato, R. Riccieri and C. K. Poon, X X I
International Confer'ence on Coordination Chemistry, Toulouse, July, 1980.
26. M. Maestri, F. Bolletta, L. Moggi, M. S. Henry and M. Z. Hoffman, J. Am. Chem. Soc. 100,
Downloaded by [Nationwide Childrens Hospital] at 10:37 14 October 2017

2694 (1978).
27. (a) N. J. Shipley and R. G. Linck, J. Phys. Chem. 84,2490 (1980).(b) A. W. Adamson and A
R. Gutierrez, ibid, 2492 (1980).
28. Y. S. Kang, F. Castelli and L. S. Forster, J. Phys. Chem. 83, 2368 (1979).
29. A. D. Kirk, private communication.
30. N. A. P. Kane-Magiure, G. M. Clonts and R. C. Kerr, Inorg. Chim. Acta 44, L157 (1980).
31. A. D. Kirk, Inorg. Chem. 18, 2326 (1979).
32. A. R. Gutierrez and A. W. Adamson, J. Phys. Chem. 82, 902 (1978).
33. S. Benson, Foundations of Chemical Kinetics, (McGraw-HiII, New York, 1960), p. 39.
34. N. A. P. Kane-Maguire, C. G. Toney, B. Swiger, A. W. Adamson and R. E. Wright, Inorg.
Chim. Acta 22, L11 (1977).
35. A. D. Kirk, T. E. Hoggard, G. B. Porter, M. G. Rockley and M. W. Windsor, Chem. Phys.
Lett. 37, 199 (1976).
36. F. Castelli and L. S. Forster, J. Phys. Chem. 81, 403 (1977).
37. P. D. Fleischauer, A. W. Adamson and G . Sartori, Inorgunic Reuction Mechanisms, Part 11,
ed. by J. 0. Edwards, (John Wiley and Sons, Inc., New York. 1972).
38. J. I. Zink, Inorg. Chem. 12, 1018 (1973).
39. L. H. Skibsted and P. C. Ford, Inorg. Chem. 19, 1828 (1980).
40. K. F. Purcell, S. F. Clark and J. D. Peterson, Inorg. Chem. 19, 2183 (1980).
41. E. Martins, E. B. Kaplan and P. S. Sheridan, Inorg. Chem. 18, 2195 (1979).
42. P. C. Ford Concepts of Inorganic Photochemistry, ed. by A. W. Adamson and P. F.
Fleischauer, (John Wiley and Sons, Inc., 1975).
43. M. A. Bergkamp, R. J. Watts, P. C. Ford, J. Brannon and D. Magde, Chem. Phys. Lett. 59,
125 (1978).
44. A. W. Adamson, Pure Appl. Chem. 51, 313 (1979).
45. M. Larson, A. W. Adamson and R. C. Rumfeldt, Inorg. Chim. Acta 44, 213 (1980).
46. M. Larson and A. W. Adamson, unpublished work.
41. See E. Konig and S. Kremer, Ligand Field Energy Diagrams, (Plenum Press, New York,
1977).
48. T. L. Kelly and J. F. Endicott, J . Phys. Chem. 76, 1937 (1977); idrtn, J. Am. Chem. Soc. 94,
278 (1972).
49. M. S. Wrighton, Chem. Rev. 74, 401 (1974).
50. R. M. Dahlgren and J. I. Zink, Inorg. Chem. 16, 3154 (1977).
51. For another example of excited state tuning see M. J. Incorvia and J. I. Zink, Inorg. Chem.
17, 2250 (1978).
52. A. Lees and A. W. Adamson, J. Am. Chem. Soc. 102,6874 (1980).
53. See M. S. Wrighton, H. B. Abrahamsonand D. L. Morse, J. Am. Chem. Soc. 98,4105(1976).
54. M. S. Wrighton, private communication.
55. G. Malouf and P. C. Ford, J. Am. Chem. SOC.96,601 (1974).
56. R . F. Dallinger and W. H. Woodruff, J. Am. Chem. SOC.101, 4391 (1979).

45