Steric control of electron transfer. Changeover from outer-sphere to inner-sphere mechanisms in arene/quinone redox pairs

Article abstract of DOI:10.1021/ja9831002

The various aromatic hydrocarbons (Chart 2) constitute a sharply graded series of sterically encumbered (unhindered, partially hindered, and heavily hindered) donors in electron transfer (ET) to quinones (Chart 1). As such, steric effects provide the quantitative basis to modulate (and differentiate) outer-sphere and inner-sphere pathways provided by matched pairs of hindered and unhindered donors with otherwise identical electron-transfer properties. Thus the hindered donors are characterized by (a) bimolecular rate constants (k₂) that are temperature dependent and well correlated by Marcus theory, (b) no evidence for the formation of (discrete) encounter complexes, (c) high dependency on solvent polarity, and (d) enhanced sensitivity to kinetic salt effects - all diagnostic of outer-sphere electron-transfer mechanisms. Contrastingly, the analogous unhindered donors are characterized by (a) temperature-independent rate constants (k₂) that are 10² times faster and rather poorly correlated by Marcus theory, (b) weak dependency on solvent polarity, and (c) low sensitivity to kinetic salt effects - all symptomatic of inner-sphere ET mechanisms arising from the preequilibrium formation of encounter complexes with charge-transfer (inner-sphere) character. Steric encumbrances which inhibit strong electronic (charge-transfer) coupling between the benzenoid and quinonoid π systems are critical for the mechanistic changeover. Thus, the classical outer-sphere/inner-sphere distinction (historically based on coordination complexes) is retained in a modified form to provide a common terminology for inorganic as well as organic (and biochemical) redox systems.

Full text of DOI:10.1021/ja9831002

J. Am. Chem. Soc. 1999, 121, 617-626
617
Steric Control of Electron Transfer. Changeover from Outer-Sphere to
Inner-Sphere Mechanisms in Arene/Quinone Redox Pairs
Stephan M. Hubig, Rajendra Rathore, and Jay K. Kochi*
Contribution from the Department of Chemistry, UniVersity of Houston, Houston, Texas 77204-5641
ReceiVed August 28, 1998
Abstract: The various aromatic hydrocarbons (Chart 2) constitute a sharply graded series of sterically
encumbered (unhindered, partially hindered, and heavily hindered) donors in electron transfer (ET) to quinones
(Chart 1). As such, steric effects provide the quantitative basis to modulate (and differentiate) outer-sphere
and inner-sphere pathways provided by matched pairs of hindered and unhindered donors with otherwise identical
electron-transfer properties. Thus the hindered donors are characterized by (a) bimolecular rate constants (k2)
that are temperature dependent and well correlated by Marcus theory, (b) no evidence for the formation of
(discrete) encounter complexes, (c) high dependency on solvent polarity, and (d) enhanced sensitivity to kinetic
salt effectssall diagnostic of outer-sphere electron-transfer mechanisms. Contrastingly, the analogous
2
unhindered donors are characterized by (a) temperature-independent rate constants (k2) that are 10 times
faster and rather poorly correlated by Marcus theory, (b) weak dependency on solvent polarity, and (c) low
sensitivity to kinetic salt effectssall symptomatic of inner-sphere ET mechanisms arising from the preequilibrium
formation of encounter complexes with charge-transfer (inner-sphere) character. Steric encumbrances which
inhibit strong electronic (charge-transfer) coupling between the benzenoid and quinonoid π systems are critical
for the mechanistic changeover. Thus, the classical outer-sphere/inner-sphere distinction (historically based
on coordination complexes) is retained in a modified form to provide a common terminology for inorganic as
well as organic (and biochemical) redox systems.
Introduction
experimentally by charge-transfer (CT) interactions extant in
the donor/acceptor precursor or encounter complex prior to
electron transfer, and the degree of charge transfer as defined
by Mulliken theory can be taken as a measure of the donor/
In bimolecular electron transfer (ET) between freely diffusing
donors and acceptors in solution, the nuclear prearrangement
of the reactants in the transition stateswith its critical donor/
6
-8
9
acceptor bonding.
For example, we recently showed that
acceptor distance and orbital overlapslimits the intrinsic rate
of the electron exchange.1,2 As a result, all theoretical calcula-
electron transfer from arene donors to photoactivated quinones
occurs via encounter complexes (EC) with substantial charge-
transfer bonding, by the observation of the near-IR absorption
bands and relatively high formation constants (KEC). Further-
more, the resulting two-step quenching mechanism involving
encounter-complex formation (KEC) followed by electron transfer
(kET) leads to second-order rate constants (k2) that are composites
tions of ET rate constants invoke far-reaching assumptions on
the relative orientation and electronic interaction of the donor
and the acceptor in the transition state.3
Owing to the intrinsic lifetime of transition states, their direct
(
spectroscopic) observation constitutes an experimental chal-
4
lenge. However, attempts have been made to predict structures
and degrees of donor/acceptor bonding in various ET transition
1
0,11
of KEC and kET.
In this study, we show how sterically
5
encumbered electron donors can circumvent the kinetics com-
plications arising from the formation of encounter complexes.
The systematic comparison of the electron-transfer kinetics of
hindered versus unhindered electron donors will probe the
effects of steric hindrance on the ET transition state. 1
states by different theoretical methods. Electronic coupling that
promotes electron transfer between redox partners is revealed
(1) Eyring, H.; Polanyi, M. Z. Phys. Chem. 1931, B12, 279. (b) Glasstone,
S.; Laidler, K. J.; Eyring, H. The Theory of Rate Processes; McGraw-Hill:
2,13
New York, 1941.
2) Ultrafast rate constants of kET > 2 × 10 s-1 have been determined
12
(
for electron-transfer processes within electron donor/acceptor complexes.
(6) Mulliken theory7 describes the wave function (ΨAD) of a charge-
See: (a) Wynne, K.; Galli, C.; Hochstrasser, R. M. J. Chem. Phys. 1994,
transfer complex as principally the sum of the dative (bonding) function
-
1
00, 4797. (b) Asahi, T.; Mataga, N. J. Phys. Chem. 1989, 93, 6575. See
also: Hannappel, T.; Burfeindt, B.; Storck, W.; Willig, F. J. Phys. Chem.
997, B101, 6799.
3) Traditionally, electron-transfer reactions are classified as either “inner-
(ψ1) and the “no-bond” function (ψ0), i.e. ΨAD ) aψ0 (A, D) + bψ1 (A ,
+
D ) + ...
1
(7) Mulliken, R. S. J. Am. Chem. Soc. 1950, 72, 600. (b) Mulliken, R.
S. J. Am. Chem. Soc. 1952, 74, 811. (c) Mulliken, R. S.; Person, W. M.
Molecular Complexes; Wiley: New York, 1969.
(
sphere” (bonded) or “outer-sphere” (nonbonded) processes. See: (a) Taube,
H. Electron-Transfer Reactions of Complex Ions in Solution; Academic
Press: New York, 1970. (b) Cannon, R. D. Electron-Transfer Reactions;
Butterworth: London, 1980. (c) Zipse, H. Angew. Chem., Int. Ed. Engl.
7
(8) On the basis of Mulliken theory, the degree of charge transfer is
2
defined as the ratio (b/a) of the mixing coefficients a and b of the no-
bond and the dative wave functions, respectively. For the experimental
2
1
997, 36, 1697. (d) Eberson, L. New J. Chem. 1992, 16, 151. (e) Tributsch,
H.; Pohlmann, L. Science 1998, 279, 1891.
4) See: (a) Polanyi, J. C.; Zewail, A. H. Acc. Chem. Res. 1995, 28,
19. (b) Zhong, D.; Zewail, A. H. J. Phys. Chem. 1998, A102, 4031.
5) Sastry, G. N.; Shaik, S. J. Am. Chem. Soc. 1998, 120, 2131. (b)
determination of (b/a) , see: (a) Ketelaar, J. A. A. J. Phys. Radium 1954,
15, 197. (b) Tamres, M.; Brandon, M. J. Am. Chem. Soc. 1960, 82,
2134. See also: (c) Briegleb, G. Elektronen-Donator-Acceptor-Komplexe;
Springer: Berlin, 1961.
(9) Rathore, R.; Hubig, S. M.; Kochi, J. K. J. Am. Chem. Soc. 1997,
119, 11468.
(10) Hubig, S. M.; Kochi, J. K. J. Am. Chem. Soc. In press.
(11) Compare: Rehm, D.; Weller, A. Isr. J. Chem. 1970, 8, 259.
(
1
(
Eberson, L.; Shaik, S. S. J. Am. Chem. Soc. 1990, 112, 4484. (c) Bertran,
J.; Gallardo, I.; Moreno, M.; Sav e´ ant, J.-M. J. Am. Chem. Soc. 1996, 118,
5
737. (d) Su, J. T.; Zewail, A. H. J. Phys. Chem. 1998, A102, 4082.
1
0.1021/ja9831002 CCC: $18.00 © 1999 American Chemical Society
Published on Web 01/13/1999

618 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Hubig et al.
Chart 1. Quinone Acceptors
which reveal substantial electronic coupling of the donor/
acceptor orbitals comparable to that found in mixed-valence
2
6
metal complexes. Since the latter are used as prototypical
models for the bridged-activated complex in inner-sphere
2
3,27
electron transfers,
we adopt the term “inner-sphere” to also
describe the electron transfer between donors and acceptors in
the encounter complex that are not covalently bonded but are
2
8-30
nonetheless strongly coupled.
The critical experimental
evidence for inner-sphere character is the pronounced sensitivity
of the electron-transfer rates to steric hindrance and the
weakening of the electronic coupling between the donor and
We employ benzoquinones in their photoactivated state as
electron acceptors and monitor electron transfer from various
hindered and unhindered arene donors by time-resolved laser-
flash experiments. Chloranil (CA), 2,5-dichloroxyloquinone
(17) Orbital overlap is commonly described by the electronic coupling
matrix element V (or HAB), which is assumed (within the limit of weak
coupling) to exhibit an exponential falloff with the donor-acceptor distance
R, i.e., V ) V0 exp{-â (R - R0)} with R0 being the donor-acceptor distance
at van der Waals contact. (b) See: Endicott, J. F.; Kumar, K.; Ramasami,
T.; Rotzinger, F. P. Prog. Inorg. Chem. 1983, 30, 141 and references therein.
(CX), and 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ)
in Chart 1 are electron acceptors of choice since their long-
lived (µs) excited (triplet) states exhibit high reduction potentials
(
E*red > 1.6 V vs SCE).14 The series of methyl-substituted
-1
-1
(
c) Weak coupling (V < 100 cm or 0.3 kcal mol ) of the donor and
18
benzenes and their sterically encumbered analogues in Chart 2
are selected on the basis of similar one-electron oxidation
acceptor orbitals is found in solvent-separated ion-radical pairs and in
donor/acceptor couples separated by rigid spacers.19 (d) Strong coupling is
-1 18, 20
observed in contact ion-radical pairs (V = 800-1000 cm ),
exciplexes
V = 1300 cm ), cyclophane-derived charge-transfer complexes (V =
1300-1800 cm-1),22 and binuclear (mixed-valence) metal complexes23, 24a
1
5
potentials.
-1 21
(
The electron acceptors and donors in Charts 1 and 2 are
especially well-suited for delineating the effects of steric en-
cumbrance on the mechanism of electron transfer since (a) the
driving force can be tuned over a wide range without essential
changes in the size and orientation of the redox centers and (b)
steric hindrance can be introduced by bulky substituents without
affecting the driving force.16 Thus, we will demonstrate that
increases in the donor-acceptor distance caused by steric hin-
drance induce a changeover in the electron-transfer mechanism
owing to the substantial diminution of the donor/acceptor orbital
24
-1
25
(e.g., pyrazine-bridged: V = 300-700 cm ; cyano-bridged: V = 1500-
-
1
2
100 cm ).
18) Gould, I. R.; Young, R. H.; Moody, R. E.; Farid, S. J. Phys. Chem.
991, 95, 2068. (b) Gould, I. R.; Young, R. H.; Mueller, L. J.; Farid, S. J.
(
1
Am. Chem. Soc. 1994, 116, 8176. (c) Tachiya, M.; Murata, S. J. Am. Chem.
Soc. 1994, 116, 2434.
(
19) Miller, J. R.; Calcaterra, L. T.; Closs, G. L. J. Am. Chem. Soc.
1
984, 106, 3047. (b) Closs, G. L.; Calcaterra, L. T.; Green, N. J.; Penfield,
K. W.; Miller, J. R. J. Phys. Chem. 1986, 90, 3673. (c) Wasilewski,
M. R.; Niemczyk, M. P.; Svec, W. A.; Pewitt, E. B. J. Am. Chem.
Soc. 1985, 107, 1080. (d) Paddon-Row, M. N. Acc. Chem. Res. 1994,
2
7, 18.
20) Gould, I. R.; Noukakis, D.; Gomez-Jahn, L.; Goodman, J. L.; Farid,
S. J. Am. Chem. Soc. 1993, 115, 4405.
21) Gould, I. R.; Young, R. H.; Mueller, L. J.; Albrecht, A. C.; Farid,
17
overlap. Unhindered donors form distinct encounter complexes
(
with the quinone acceptorssthe charge-transfer absorptions of
(
(
12) Rathore, R.; Lindeman, S. V.; Kochi, J. K. J. Am. Chem. Soc. 1997,
S. J. Am. Chem. Soc. 1994, 116, 8188.
(22) Benniston, A. C.; Harriman, A.; Philp, D.; Stoddart, J. F. J. Am.
Chem. Soc. 1993, 115, 5298.
1
19, 9393. (b) For the effects of steric hindrance on exciplex formation,
see: Jacques, P.; Allonas, X.; Suppan, P.; Von Raumer, M. J. Photochem.
Photobiol. 1996, A 101, 183.
(23) Haim, A. Prog. Inorg. Chem. 1983, 30, 273.
(13) The precursor or encounter complex (prior to electron transfer) and
(24) Creutz, C. Prog. Inorg. Chem. 1983, 30, 1. (b) Creutz, C.; Taube,
H. J. Am. Chem. Soc. 1969, 91, 3988. (c) Goldsby, K. A.; Meyer, T. J.
Inorg. Chem. 1984, 23, 3002.
the ET transition state are assumed to be structurally similar and exhibit
more or less comparable donor/acceptor interactions. See: (a) Sutin, N.
Acc. Chem. Res. 1968, 1, 225. Compare also: (b) Marcus, R. A. J. Chem.
Phys. 1956, 24, 966. (c) Marcus, R. A. Angew. Chem., Int. Ed. Engl. 1993,
(25) Burewicz, A.; Haim, A. Inorg. Chem. 1988, 27, 1611.
(26) Compare the electronic coupling matrix elements of organic
2
1
3
2, 1111 and references therein.
14) The reduction potential of the photoactivated quinones (E*red) is
donor/acceptor exciplexes with those of mixed-valence metal com-
plexes.2
3-25
(
taken as the sum of the quinone triplet energy (ET = 2.2 eV) and the
reduction potential of the quinone in its ground state. For the triplet energies
(27) Astruc, D. Electron Transfer and Radical Processes in Transition-
Metal Chemistry; VCH: New York, 1995; p 30.
(ET) of the quinones, see: (a) Shcheglova, N. A.; Shigorin, D. N.; Yakobson,
(28) This view of “inner-sphere” electron transfer goes beyond its original
definition3a that is largely based on ionic (inorganic) coordination complexes
by including uncharged (organic) redox systems with measurable donor/
acceptor coupling. We believe it is highly desirable to retain the classical
inner-sphere/outer-sphere distinction in this modified form (to avoid
inventing new terms) so that a universal and common terminology can be
applied to describe electron-transfer mechanisms in all branches of inorganic
chemistry, organic chemistry, and biochemistry.
G. G. Y.; Tushishvili, L. Sh. Russ. Phys. Chem. 1969, 43, 1112. (b)
Trommsdorff, H. P.; Sahy, P.; Kahane-Paillous Spectrochim. Acta 1968,
2
4A, 785. (c) Herre, W.; Weis, P. Spectrochim. Acta 1973, 29A, 203. (d)
Koboyama, A. Bull. Chem. Soc. Jpn. 1962, 35, 295. For the reduction
potentials of the quinones (in the ground state), see: (e) Mann, C. K.; Barnes,
K. K. Electrochemical Reactions in Non-Aqueous Systems; Dekker: New
York, 1970. (f) Peover, J. E. J. Chem. Soc. 1962, 4540.
(15) (a) Howell, J. O.; Goncalvez, J. M.; Amatore, C.; Klasinc, L.;
(29) From the practical point of view, this distinction between inner-
sphere and outer-sphere electron transfers based on the (experimentally
observable) electronic coupling of donor and acceptor is rather straight-
forward. First, it circumvents the (quantitative) ambiguities inherent to
chemically based differentiations such as ligand exchange, isotopic labeling,
Wightman, R. M.; Kochi, J. K. J. Am. Chem. Soc. 1984, 106, 3968. (b)
Note that the steric encumbrance of hindered donors such as hexaethyl-
benzene (HEB) relative to hexamethylbenzene (HMB) is gauged by their
increased van der Waals thickness of 2r g 6.4 Å arising from the pendant
methyl groups (illustrated below)
2
3,27
bridged intermediate, etc. in inner-sphere electron transfers.
Second,
anomalies in the outer-sphere behavior (i.e., deviations from Marcus theory)
need not to be explained by approximate corrections of the work terms,
etc.,3b if they can be accounted for by electronic coupling terms in an inner-
sphere model.1
7b,41b
(30) Note also that an inner-sphere/outer-sphere distinction based on
orbital overlap allows for a continuum of intermediate cases to exist between
the two idealized models that depend on the degree of electronic coupling.
Moreover, the simultaneous occurrence of both mechanisms is readily
accounted for in medium-strong interactions. See: (a) Taube, H.; Myers,
H. J. Am. Chem. Soc. 1954, 76, 2103. (b) Melvin, W. S.; Haim, A. Inorg.
Chem. 1977, 16, 2016. (c) Connocchioli, T. J.; Hamilton, E. J.; Sutin, N. J.
Am. Chem. Soc. 1965, 87, 926. For the suggestion of a continuum, see:
(d) Fukuzumi, S.; Wong, C. L.; Kochi, J. K. J. Am. Chem. Soc. 1980, 102,
2928. (e) Rosseinsky, D. R. Chem. ReV. 1972, 72, 215. (f) Eberson, L.
AdV. Phys. Org. Chem. 1982, 18, 79.
that discourage any close cofacial approach to the benzenoid (π-) chro-
1
2
mophore (see Chart 3). In addition, a few “partially” hindered donors are
included in this study to demonstrate the effects of the (ring) position of
bulky substituents on the overall steric encumbrance of the arene.
(16) Furthermore, the use of uncharged redox partners allows the electron
transfer to be studied in aprotic polar as well as nonpolar solvents (to avoid
the rather unique ionic solvation by water). Note also that the charge-
delocalization and charge-transfer ability is optimized in such multiatom
(expanded) redox centers of the donor/acceptor pair.

Steric Control of Electron Transfer
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 619
Chart 2. Arene Donors
Table 1. Electron-Transfer Rate Constants and Ion Yields
the acceptor due to an increased distance. Most importantly, a
substantial increase in the donor-acceptor distance due to bulky
substituents in hindered donors ultimately leads to outer-sphere
electron transfer with no or very little sensitivity to steric
hindrance.31
6
-1
k
2
[10 M s-1]c
Q/ArHa
∆GET [eV]b
CH CN CH Cl
Φiond
5.3 (8.4) >0.3
3
2
2
1
2
3
4
5
6
7
8
9
0
1
2
CX/MES
CX/TTB
CX/XYL
CX/DTB
CX/TMB
CA/TOL
CX/DUR
CX/OMA
CX/PMB
CX/HMB
CX/HEB
CA/MES
+0.49
+0.39
+0.44
+0.41
+0.27
+0.25
+0.21
+0.22
+0.13
(0
-0.03
-0.04
-0.14
-0.09
-0.12
-0.32
-0.31
-0.53
-0.56
-1.02
-1.05
4.0 (3.9)
0.4
2.4 (9.1)
0.8
e
e
3.5 (4.6) >0.45
Results
e
15(18)
15(18)
300(380)
17
>0.54
>0.85
>0.72
0.99
e
40 (50)
10 (10)
2560 (2550)
50
I. Electron Transfer from Aromatic Donors to Photoac-
tivated Quinones. A. Determination of the Second-Order
Rate Constants (k2). Photoexcitation of the quinones (Q) in
Chart 1 with a 10-ns laser pulse at 355 nm spontaneously
generates their excited triplet states (Q*) with unit efficiency
in acetonitrile solution, and the characteristic absorption spec-
trum of Q* decays to the spectral baseline on the microsecond
4000
5140
970
1500 (2200)
5500 (7300)
160
e
1
1
1
1.03
1.03
0.94
1.02
0.95
1.15
0.95
e
4400
900
1150 (1080)
14
13 CA/TTB
14 CA/XYL
4
-1 9
time scale with kd < 5 × 10 s . However, Q* decays
significantly faster in the presence of the aromatic donors (ArH),
and the concomitant formation of the quinone anion radical
5400
1300
1200 (800)
3
1
1
1
1
1
5
6
7
8
9
CA/DTB
CA/DUR
CA/OMA
CA/HMB
CA/HEB
6000
2750
12000
1900
-
•
+•
(
Q ) and the arene cation radical (ArH ) is observed with
8000
6900
12000
8000
0.98
1.14
e
identical (first-order) rate constants for Q* decay and ion-radical
formation. Quantitative analysis of the time-resolved absorption
spectra (see the Experimental Section) establishes the formation
20 DDQ/HMB
21 DDQ/HEB
21000
16000
e
e
e
-
•
+•
of the ion radicals Q and ArH to occur in a 1:1 molar ratio
a
Quinone/arene combination (see Charts 1 and 2). b Electron-transfer
with unit efficiency, i.e. Φion = 1,32
driving force as determined from eq 2. c Second-order rate constant
for the electron transfer from the arene donor to the photoactivated
quinone. d Ion yield ((0.1) as determined by benzophenone actinometry
k2
-•
+•
e
Q* + ArH (CH3CN)8 Q + ArH
(1)
(see the Experimental Section). Not measured.
encumbrance of the arene donor. Thus the rate constants for
electron transfer from sterically hindered tert-butyl donors TTB
and DTB as well as the tetra- and hexa-substituted analogues
OMA and HEB are reduced by a factor of 100 (or more)
compared to k2 of the analogous unhindered donors MES, XYL,
DUR, and HMB, respectively, as paired mates with essentially
The kinetics of the bimolecular electron transfer in eq 1 is
examined under pseudo-first-order conditions by monitoring the
decay of Q* (or the simultaneous growth of Q-• and ArH ) as
a function of excess arene concentration ([ArH]). At low ArH
concentrations of [ArH] < 0.01 M, the observed (first-order)
rate constant (kobs) increases linearly with the arene concentra-
tion, and the slope of the pseudo-first-order plot yields the
second-order rate constant (k2) listed in Table 1 for electron
transfer from ArH to Q* in eq 1.
+•
0
the same donor properties (compare E ox values in Chart 2).
Such a trend is independent of the quinone acceptor, as seen in
Table 1 by the pairwise comparisons of the even/odd entries of
k2 at comparable driving forces (-∆GET).
B. Steric Effects. The k2 values for the various quinone/arene
combinations in Table 1 are strongly affected by the steric
(32) In some cases (entries 1-4 in Table 1), the ion-radical yields (Φion)
are less than unity, and k2 in Table 1 represents the upper limit of the second-
order rate constant for electron transfer.
(31) Juillard, M.; Chanon, M. Chem. ReV. 1983, 83, 425.

620 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Hubig et al.
Table 2. Solvent Dependence of the Electron-Transfer Rates
Table 3. Salt Effects on Electron-Transfer Rates
[10 M s-1]b
8
-1
k
2
[10 M s-1]c
8
-1
k
2
CH
(ꢀ ) 35.9)
3
CN
CH
(ꢀ ) 8.9)
2
Cl CHCl
2
3
CCl
(ꢀ ) 2.2)
4
Q/ArHb
no salta
0.1 M salta
Q/ArHa
c
c
(ꢀ ) 4.8)
c
c
CX/HEB
CA/TTB
CA/DTB
0.42
0.16
0.02
2.41
0.38
0.41
CX/HMB
CX/TEM
CX/PET
CX/HEB
51.4
32
17
55
33
20
1.6
23
17
14
0.4
23
7.6
1.7
0.04
a
b
Tetra-n-butylammonium hexafluorophosphate in CHCl3. Quinone/
9.7
arene combination (see Charts 1 and 2). c Second-order rate constant
for the electron transfer from the arene donor to photoactivated quinone.
CX/DUR
CX/OMA
25.6
<0.5
3.0
0.17
2.3
0.15
1.1
0.02
CA/MES
CA/DTT
CA/TTB
39
24
9
12
8.8
0.14
6.7
5.3
0.16
2.5
0.86
0.003
from hindered donors is quite sensitive to the presence of inert
salt. Thus the bimolecular rate constants (k2) increase by up to
a factor of 5 when 0.1 M tetra-n-butylammonium hexafluoro-
phospate is deliberately added to the solution of quinone and
hindered donor in chloroform.
CA/XYL
CA/DTB
54
13
12
0.03
6.6
0.02
1.8
0.002
a
Quinone/arene combination (see Charts 1 and 2). b Second-order
II. Driving-Force (-∆GET) Dependence of the Electron-
Transfer Rate Constant. The driving-force dependence of
electron transfer from ArH to Q*, as given by the free energy
rate constant for the electron transfer from the arene donor to
photoactivated quinone. Dielectric constant (see ref 33c).
c
1
1
change (∆GET), is based on eq 2,
C. Solvent Effect. Electron transfer from hindered donors
to quinone is also strongly affected by solvent polarity, whereas
the rate constants for electron transfer from the unhindered
0
ox
∆
GET ) E - E*red + constant
(2)
donors do not vary significantly from acetonitrile to dichlo-
0
romethane.33 The pronounced solvent effect on the rate constants
where E ox is the oxidation potential of the benzene donor
15
(
ArH) and E*red is the reduction potential of the photoactivated
quinone (Q*). The second-order rate constants (k2) in Table
vary over 4 orders of magnitude from the most endergonic
electron-transfer couple (CX*/MES) to the most exergonic
couple (DDQ*/HMB). Figure 1 shows that the rate constants
k2) do not increase linearly with the driving force (-∆GET) of
(k2) of the hindered donor/acceptor pairs becomes even more
1
4
evident with the solvent variation presented in Table 2. Thus,the
rate constants for electron transfer from the hindered donors
HEB, OMA, TTB, and DTB progressively decrease (up to 4
orders of magnitude) from the polar solvent acetonitrile (ꢀ )
1
(
3
5.9) to dichloromethane (ꢀ ) 8.9), chloroform (ꢀ ) 4.8), and
33
the electron transfer. A strong increase is observed over more
than 3 orders of magnitude in the endergonic and slightly
exergonic region (-0.5 eV < ∆GET < 0.5 eV), and this is
the nonpolar solvent carbon tetrachloride (ꢀ ) 2.2). In strong
contrast, the rate constants for the unhindered analogues hexa-
methylbenzene (HMB), durene (DUR), mesitylene (MES), and
p-xylene (XYL) undergo minor change (at most by a factor of
10
-1 -1
followed by a limiting value of k2 = 10
M
s
which remains
unchanged over the exergonic region (-1.5 eV < ∆GET < -0.5
eV). A closer scrutiny reveals that the hindered (filled circles)
and the unhindered (open circles) donor/acceptor pairs show
quite different driving-force dependences.
2
5) as a result of the same solvent variation. The effects of
partial hindrance on the electron-transfer kinetics are also shown
in Table 2 by the falloff in k2 in all solvents from the successive
replacement of the methyl substituents of hexamethylbenzene
by ethyl groups (see entries 1-4 in Table 2). It is noteworthy
that the rate constants do not decrease monotonically with the
number of ethyl substituents, but an abrupt drop in the rate
constants occurs in the interval from pentaethyltoluene (PET)
to hexaethylbenzene (HEB). Similarly, the difference in the rate
constants between tri-tert-butylbenzene (TTB) and di-tert-
butyltoluene (DTT) is much larger than that between DTT and
mesitylene (MES, see entries 7-9). The solvent dependence
of the rate constants follows the same patternsthe partially
hindered arene donors showing a moderate solvent effect sim-
ilar to that observed for the unhindered donors. On the other
hand, substantial changes of the ET rate constants are solely
observed for the hindered donors due to solvent polarity (vide
supra).
A. Hindered Donors. The ET rate constants of the hindered
donors (Table 1 and solid circles in Figure 1) can readily
be simulated by the Marcus free-energy correlation (dashed
1
3
line), i.e.
q
k2 ) k /[1 + A exp(∆G /RT)]
(3)
(4)
diff
A ) kdiff/ZK
q
2
∆
G ) (λ/4)[1 + (∆G - R)/λ)]
(5)
ET
where k2 is the second-order rate constant for the bimolecular
electron transfer, kdiff is the rate constant for diffusion,34 Z is
the frequency factor, K ) kdiff/k-diff is the equilibrium constant
for diffusional encounters, ∆G is the free activation enthalpy,
GET is the free energy as defined in eq 2, λ is the reorganiza-
tion energy, and R is a shift parameter applied to triplet
quenching. The close match between the experimental data
solid circles) and the Marcus correlation (dashed line) obtains
in Figure 1 between -0.5 eV < ∆GET < 0.5 eV.
D. Salt Effects. The effect of added salt on the rate constants
for electron transfer from variously hindered donors to photo-
activated quinones in chloroform is presented in Table 3. In
35
q
∆
contrast to the unhindered donors, which do not exhibit a
3
6e
significant kinetic salt effect,9
,33b
the electron-transfer kinetics
(
36
(33) In dichloromethane solution and other less polar solvents, the
quenching of Q* by polymethylbenzenes (ArCH3) results in the formation
•
•
of semiquinone (QH ) and benzyl (ArCH2 ) radicals. However, salt-effect
(34) Moore, J. W.; Pearson, R. G. Kinetics and Mechanisms, 3rd ed.;
Wiley: New York, 1981; p 239f.
(35) The equilibrium constant K for diffusional encounters depends on
the effective encounter distance R between the donor and the acceptor. For
the encounter complex between uncharged (spherical) species with R ) 7
Å, the formation constant is estimated to be K ) 0. 9 M . See: Eigen, M.
Z. Phys. Chem. N. F. (Leipzig) 1954, 1, 176.
3
3b
studies unambiguously show that the hydrogen transfer can occur via
rate-determining) electron transfer followed by fast proton transfer. (b)
Bockman, T. M.; Hubig, S. M.; Kochi, J. K. J. Am. Chem. Soc. 1998, 120,
826. (c) As a measure for the solvent polarity, the dielectric constants
(
2
-
1
were taken. See: Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of
Photochemistry, 2nd ed.; Dekker: New York, 1993.

Steric Control of Electron Transfer
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 621
Table 4. Temperature Dependence of Electron-Transfer Rate
Constants
Q/ArHa
CX/DUR
+0.21
CA/MES
-0.04
CA/TTB
-0.14
∆
2
G
ET [eV]b
[10 M s-1]c
T ) 239 K
T ) 253 K
T ) 277 K
T ) 295 K
8
-1
k
d
d
32
d
44
2.9
17.4
4.4
5.4
8.1
9.0
7.9
13.1
26.0
26.2
25.6
0
-
1 e
E
A
[kJ mol ]
q
G [kJ mol ]f
-1
30.8
∆
a
Quinone/arene combination (see Charts 1 and 2). b Electron-transfer
driving force as determined with eq 2. c Second-order rate constant for
the electron transfer from the arene donor to photoactivated quinone.
d
e
Not measured. Activation energy as determined from the Arrhenius
f
plot in Figure 2. Free activation enthalpy determined from the Marcus
formulation37 with λ ) 77 kJ mol
-1
.
Figure 1. Free-energy dependence of the second-order rate constants
of the electron transfer from hindered (b) and unhindered (O) arene
donors to photoactivated quinones. The dashed line represents the best
fit of the data points of the hindered donors to the Marcus correlation
9
-1 -1
11 -1
according to eqs 3-5 with kdiff ) 8 × 10 M
s , Z ) 10 s , K )
3
5
36e
0
.86, λ ) 1.2 eV, and R ) 0.4 eV.
B. Unhindered Donors. The kinetics data for the unhindered
donor/acceptor pairs (Figure 1, open circles) show significant
deviations from the Marcus prediction. Thus, the k2 values are
generally higher than those of the corresponding hindered donor/
acceptor pairs, and the points are too strongly scattered in the
endergonic ∆GET region between 0.25 and 0.49 eV for a
satisfactory Marcus simulation. To understand these striking
effects of steric hindrance, let us now turn to a comparative
analysis of the electron-transfer kinetics with hindered and
unhindered donorssincluding studies of the temperature, solvent
polarity, and salt effects. We focus on the endergonic and
slightly exergonic driving-force region (-0.5 eV < ∆GET <
Figure 2. Temperature dependence of the second-order rate constants
for electron transfer from TTB to CA* (b), from MES to CA* (2),
and from DUR to CX* (9) evaluated by the Arrhenius relationship.
-
1
0
.5 eV) since it shows the most pronounced effects of steric
The slopes yield activation energies (E ) of 7.9, 2.9, and 0 kJ mol
,
A
hindrance, and it is thus particularly informative of the structural
requirements in bimolecular electron transfers.
respectively.
from the ∆GET values according to Marcus theory.37 For the
two unhindered donor/acceptor pairs, we note a striking dis-
crepancy (up to 30 kJ mol ) between the experimentally
determined activation energies (EA) and the theoretically
predicted ∆G values, whereas EA for the hindered couple CA/
TTB is in close agreement with ∆G .
III. Temperature Dependence of the Electron-Transfer
Rate Constant. The temperature dependence of the bimolecular
ET rate constants for three donor/acceptor pairs that undergo
electron transfer with driving forces in the endergonic and
slightly exergonic region is presented in Table 4. The two
unhindered couples (CX/DUR) and (CA/MES) show no or very
little variation of k2 over a temperature range of more than 40
-
1
q
q
IV. Comparative Analysis of the Electron-Transfer Kinet-
ics for Hindered and Unhindered Aromatic Donors. A.
Curved Versus Linear Kinetics Plots. A striking difference
in the kinetic behavior of hindered versus unhindered arene
donors is observed when the plots of the observed (first-order)
rate constants (kobs) for Q* decay are extended to high arene
concentrations. Thus the kinetics plots of kobs versus [ArH] for
hindered donors remain linear over a wide range of concentra-
tions up to the solubility limits of the arene. By contrast, the
kinetics plots for unhindered donors show significant curvature
at relatively low donor concentrations, and the kobs values reach
limiting (plateau) values at concentrations above 0.2 M. The
effect of steric encumbrance on the electron-transfer kinetics is
underscored in Figure 3 by the strongly contrasting behavior
of durene and its sterically hindered analogue OMA in two
solvents.
°C. Contrastingly, the value of k2 for the hindered donor/acceptor
couple (CA/TTB) doubles over a similar temperature range.
As such, the Arrhenius plots of ln k2 versus the reciprocal
temperature yield ET activation energies (EA) of 0, 2.9, and
-
1
7
.9 kJ mol for CX/DUR, CA/MES, and CA/TTB, respec-
tively (see Figure 2). For comparison, Table 4 also contains
q
the activation enthalpies (∆G ) for electron transfer as calculated
(36) Deviations of the bimolecular ET rate constants from Marcus
behavior in the highly exergonic driving-force region (∆GET < 1.5 eV)
1
1
have been observed, and the plateau of diffusion-limited rate constants
for highly exergonic electron transfers is taken into account in the empirical
free-energy correlation by Rehm and Weller.11 Various theoretical explana-
tions of the “non-Marcus” behavior (i.e., the lack of the “Marcus-inverted”
region) have been reported: (b) Kakitani, T.; Yoshimori, A.; Mataga, N. J.
Phys. Chem. 1992, 96, 5385. (c) Kikuchi, K.; Takahashi, Y.; Katagairi, T.;
Niwa, T.; Hoshi, M.; Miyashi, T. Chem. Phys. Lett. 1991, 180, 403. (d)
Gould, I. R.; Young, R. H.; Mueller, L. J.; Farid, S. J. Am. Chem. Soc.
(37) According to Marcus theory,13 the activation enthalpy (∆G ) is a
q
1
994, 116, 8176. (e) To achieve satisfactory overlap between the experi-
mental data and the Marcus simulation, the experimental ∆GET values of
Table 1 were shifted by R ) -0.4 eV. For a theoretical explanation of the
shift parameter R, see: Tamura, S.-I.; Kikuchi, K.; Kokubun, H.; Usui, Y.
Z. Phys. Chem. N. F. (Wiesbaden) 1978, 111, 7.
function of the driving force (∆GET) and the reorganization energy (λ)
q
2
q
according to: ∆G ) (λ/4)(1 + ∆GET/λ) . The calculation of ∆G in Table
-
1
4 is based on λ ) 77 kJ mol , as extracted from the Marcus treatment of
the data in Figure 1.

622 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Hubig et al.
Figure 3. Curved versus linear kinetics plots for electron transfer from durene (filled markers) and its hindered analogue OMA (open markers) to
photoactivated dichloroxyloquinone in (A) acetonitrile (circles) and (B) dichloromethane (triangles).
Table 5. Driving-Force Dependence of the Formation Constant
(KEC) of the Encounter Complex
K
EC [M-1]c
Q/ArHa
∆GET [eV]b CH
CHCl
CCl
4
3 2
CN CH Cl
2
3
1
CX/MES
CX/XYL
CX/TMB
CA/TOL
CX/DUR
CX/PMB
CX/HMB
CA/MES
CA/XYL
CA/DUR
Ca/HMB
+0.49
+0.44
+0.27
+0.25
+0.21
+0.13
(0
-0.04
-0.09
-0.32
-0.53
1.7
4.3
2.7
2.6
4.0
2.4
4.0
8.7
3
5
6
7
9
0
2
4
6
8
15
14
18
24
d
d
d
d
d
d
39
67
30
<10
d
1
71
54
19
202
49
21
1
1
1
1
d
a
Quinone/arene combination (see Charts 1 and 2). b Electron-transfer
c
driving force as determined with eq 2. Formation constant ((10%)
for the encounter complex [Q*, ArH]. d Not determined due to the linear
kinetics plots (see text).
Figure 4. Double-reciprocal representation of the curved kinetics plots
in Figure 3 for selected donor/acceptor couples in dichloromethane
solution (b ) CA/MES, O ) CX/HMB, 2 ) CX/DUR, and 9 )
CX/PMB).
various unhindered donors in four solvents are compiled in Table
5. The most striking result of this kinetics evaluation is that the
KEC values deviate substantially from the unit value calculated
3
5
for purely diffusional encounters.
The saturation (asymptotic) behavior of kobs (solid circles in
Figure 3A and triangles in Figure 3B) is diagnostic of a
preequilibrium intermediate38 between the photoactivated quino-
ne and the aromatic donor prior to electron transfer, and it has
B. Interdependence of k2, KEC, and kET. At low arene
concentrations, eq 7 simplifies to a linear correlation between
kobs and [ArH] to reveal the direct relationship between k2 (in
39c
eq 1) and KEC and kET (in eq 6), i.e.
9
been previously identified as the encounter complex, i.e.
kobs ) K k [ArH] ) k [ArH]
(8)
EC ET
2
KEC
Q*, ArH] kET
[
-•
+•
Q* + ArH y z
8 Q + ArH
(6)
encounter
Thus, the bimolecular rate constants k2 for electron transfer from
ArH to Q* can be obtained either from the initial slope of the
curved pseudo-first-order plots in Figure 3 or from the slope of
the double-reciprocal plots in Figure 4 as the product KECkET
see k2 values in parentheses in Table 1). In other words, the
second-order rate constant (k2) in eq 1 represents a composite
complex
Thus, the limiting value of kobs at high donor concentrations
corresponds to the intrinsic rate constant (kET) of the electron
transfer within the encounter complex. Accordingly, the curved
kinetics plots in Figure 3 are evaluated in a double-reciprocal
(
(
linearized) representation in Figure 4, from which the preequi-
(38) Bunnett, J. F. In InVestigation of Rates and Mechanisms of
Reactions; Part 1; Bernasconi, C. F., Ed.; Wiley: New York, 1986; p 286f.
librium constant (KEC) and the intrinsic electron-transfer rate
(39) For a mechanistic description of the limiting kinetics behavior,
constant (kET) are extracted according to eq 7:39
see: Espenson, J. D. Chemical Kinetics and Reaction Mechanisms, 2nd
ed.; McGraw-Hill: New York, 1995; p 89f. (b) For the decay kinetics of
photoexcited quinones, see: Kobashi, H.; Okada, T.; Mataga, N. Bull. Chem.
Soc. Jpn. 1986, 59, 1975. (c) Note that back electron transfer (k-ET) is
neglected in eq 6 on the basis of unit ion yields in Table 1, and the natural
decay (k0) of Q* (in the absence of donors) is not included in eq 7 since k0
, kET.9
1
1
1
1
)
+
(7)
kobs kET K k [ArH]
EC ET
The equilibrium constants (KEC) for complex formation of the

Steric Control of Electron Transfer
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 623
encounter complex between arene donor and photoactivated
9
quinone (see eq 6). Thus, the double-reciprocal kinetic evalu-
ation in Figure 4 yields the equilibrium constants for the
formation of the encounter complexes which attain values up
-
1
9
to KEC ) 200 M (Table 5). Most importantly, the analysis
of the spectral features of these donor/acceptor precursors reveals
a strong charge-transfer interaction within the transient complex,
which has been quantified in terms of its degree of charge
2
6-9
transfer (b/a) = 0.3 or 30%.
To achieve such a significant electronic interaction between
donor and acceptor orbitals, the (nonbonded) donor/acceptor pair
must be arranged in a tight complex to allow optimum coupling
of the benzenoid π orbitals. When such a close donor/acceptor
association is discouraged by bulky substituents, the charge-
transfer interaction within the donor/acceptor pair is precluded.12
Indeed, chloranil and the unhindered hexamethylbenzene readily
form the tight donor/acceptor complex with the van der Waals
separation of 3.5 Å (Chart 3), as revealed by X-ray crystal-
4
5
lography of charge-transfer crystals. In strong contrast, the
closest cofacial approach of chloranil and the hindered hexa-
ethylbenzene of more than 4.5 Å precludes complex formation,
Figure 5. Bell-shaped free-energy dependence of the formation
constant KEC of the encounter complex [Q*, ArH] in dichloro-
methane.
46
and the characteristic charge-transfer absorption is not observed
9
,12a,15b
in the sterically hindered donor/acceptor pair.
quantity, each component of which can be affected by the
solvent or the driving force.
As a result, only linear kinetics plots are observed for all
sterically hindered donor/acceptor combinations, and the forma-
tion constants (KEC , 1) of their encounter complexes are
negligibly small. Such small equilibrium constants (KEC) relate
to short encounter lifetimes to reduce the overall electron-
C. Driving-Force Dependence of KEC. Figure 5 illustrates
the unique driving-force dependence of the formation constants
listed in Table 5, in which a bell-shaped correlation between
KEC and ∆GET is obtained (in dichloromethane). Particularly
47
transfer rates, and result in bimolecular ET rate constants
-
1
noteworthy is that a maximum value of KEC ) 67 M is
observed in the isergonic region around ∆GET ) 0 eV, and KEC
values close to unity are found in the free-energy regions ∆GET
2
which are at least 10 times slower than those of the corre-
48
sponding unhindered pairs (see Tables 1 and 2). In fact, most
rate constants of the hindered donor/acceptor pairs in Tables 1
>
0.3 eV and ∆GET < -0.1 eV. The data in Table 5 for
(41) According to our definition (vide supra), inner-sphere mechanisms
40
acetonitrile follow the same trend, and a similar bell-shaped
correlation is also observed in chloroform and carbon tetra-
should be considered in all cases where charge-transfer complexes,
exciplexes, or contact ion-radical pairs are actually observed as reaction
intermediates.3 In other words, we arbitrarily take electronic interactions
1,42
9
chloride solution.
-
1
-1
in the transition state of less than 1 kcal mol (350 cm ) to result in
outer-sphere electron transfer, whereas electronic interactions substantially
-1
43,44
Discussion
greater than 350 cm lead to inner-sphere electron transfer.
Accordingly,
the outer-sphere/inner-sphere distinction in this study is based on the
structure and the critical (spectroscopically established) donor/acceptor
orbital overlap in the precursor complex immediately preceding the ET
transition state.9 As a consequence, inner-sphere ET is readily recognized
by its unusually fast rates, significant deviations from the Marcus-predicted
driving-force dependence, and pronounced sensitivity to steric hindrance.
For earlier studies, see: (a) Fukuzumi, S.; Wong, C. L.; Kochi, J. K. J.
Am. Chem. Soc. 1980, 102, 2928. (b) Fukuzumi, S.; Kochi, J. K. Bull. Chem.
Soc. Jpn. 1983, 56, 969. (c) Kochi, J. K. Angew. Chem., Int. Ed. Engl.
The comparative study of bimolecular electron transfers from
hindered versus unhindered arene donors reveals several striking
effects of steric encumbrance which are most pronounced for
endergonic and slightly exergonic electron transfers. Electron
transfer involving sterically hindered donors meets the condi-
tions of “outer-sphere” electron transfer between weakly coupled
donors and acceptors, and can therefore be analyzed according
to Marcus theory. On the other hand, electron transfers from
unhindered donors do not follow the predictions for “outer-
sphere” electron transfer owing to substantial electronic coupling
between donor and acceptor, and they are hereinafter referred
,12a
1
988, 27, 1227.
1
3
(42) For inner-sphere descriptions of ion-pair states, see: (a) Hubig, S.
M.; Bockman, T. M.; Kochi, J. K. J. Am. Chem. Soc. 1996, 118, 3842. (b)
Bockman, T. M.; Karpinski, Z. J.; Sankararaman, S.; Kochi, J. K. J. Am.
Chem. Soc. 1992, 114, 1970. (c) H o¨ rmann, A.; Jarzeba, W.; Barbara, P. F.
J. Phys. Chem. 1995, 99, 2006.
(
43) Reynolds, W. L.; Lumry, R. W. Mechanisms of Electron Transfer;
Ronald Press: New York, 1966; p 12.
44) Typical examples for inner-sphere mechanisms involving organic
to as “inner-sphere” processes since the electron transfer occurs
via charge-transfer bonds.41 In other words, the degree of donor/
(
acceptor bonding in the ET transition state is effectively modu-
lated by steric encumbrance as follows.
substrates are as follows: (a) Oxidation of alkylbenzenes by nitrate radical
•
(NO3 ): Del Giacco, T.; Baciocchi, E.; Steenken, S. J. Phys. Chem. 1993,
9
7, 5451. (b) Reduction of nitroarenes by oxygen atom transfer to Ru-
carbonyl complexes: Skoog, S. J.; Gladfelter, W. L. J. Am. Chem. Soc.
997, 119, 11049. (c) Reduction of methyl iodide by HI: Holm, T.;
Crossland, I. Acta Chem. Scand. 1996, 50, 90.
I. Steric Effects on the Electron-Transfer Kinetics. The
kinetics behavior of hindered and unhindered donors is strongly
differentiated in Figure 3. The saturation (asymptotic) behavior
of the observed rate constants for electron transfer from
unhindered donors requires a preequilibrium intermediate prior
to electron transfer. This has been previously identified as the
1
(
45) Harding, T. T.; Wallwork, S. C. Acta Crystallogr. 1955, 8, 757.
(
(
46) Based on molecular-mechanics calculations in ref 12a.
47) For example, a unit equilibrium constant (K ) kdiff/k-diff ) 1 M-1)
10
-1 -1
and a diffusion-controlled rate constant (k ) 2 × 10
diff
M
s ) for the
formation of the encounter complex result in a dissociation constant of k-diff
10 -1
(40) In acetonitrile, significant curvature in the kinetics plots such as in
) 2 × 10
s , which corresponds to a lifetime of 50 ps. See also: Marcus,
Figure 3 are only obtained for ∆GET > 0.2 eV, which allowed us to reliably
extract values for KEC and kET using the reciprocal evaluation in eq 7. The
lack of sufficient curvature in the kinetics plots for ∆GET < 0.2 eV arises
from limiting values of kobs which severely exceeded the time resolution of
R. A. in ref 6a.
(48) For steric effects on other photoinduced electron transfers, see: (a)
Jones, G., II; Chatterjee, S. J. Phys. Chem. 1988, 92, 6862. (b) Gassman,
P. G.; DeSilva, S. A. J. Am. Chem. Soc. 1991, 113, 9870. (c) Gould, I. R.;
Farid, S. J. Phys. Chem. 1993, 97, 13067.
8
-1
the 10-ns laser pulse (kET . 10 s ). See ref 9.

624 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Hubig et al.
Chart 3
and 2 are several orders of magnitude slower than those for
In both cases, quantum-mechanical tunneling due to high-
frequency modes can account for the lack of solvent effects.
diffusion-limited processes (kdiff = 1010
M
-1
s ), which
indicates that only a small fraction of the quinone/arene
collisional) encounters endure long enough to undergo electron
transfer.
II. Steric Effects on the Temperature and Solvent Depen-
-1 34
50
By contrast, the striking difference in Figure 3 on hindered
donor/acceptor couples establishes the important role of the
solvent in differentiating electron-transfer pathways. Such a
critical involvement of the solvent molecules can only be
envisioned for “loose” encounters between donors and acceptors
that characterize the “outer-sphere” electron-transfer model.
(
dence of the Bimolecular Rate Constants. To further amplify
the difference between the ET mechanisms of hindered and
unhindered donors, let us now consider temperature and solvent
effects.
A. Temperature Effect. The absence of a significant
temperature dependence on the ET rate constants of unhindered
donors in Figure 2 and Table 4 indicates ET activation energies
C. Salt Effects. The significant effects of added salt on the
ET rates of hindered donor/acceptor couples in chloroform (in
Table 3) are similarly interpreted as electron transfers within
“loose” encounters of the donors being susceptible to ionic
5
2
strength for ion-radical stabilization in solution.
(
“
EA) close to zero. This finding is in striking contrast to the
III. Steric Effects on the Free-Energy Correlation of the
Electron-Transfer Rate and Role of the Encounter Complex.
The difference in the ET mechanisms between hindered and
unhindered donor/acceptor pairs is most evident in the com-
parison of the free-energy correlations in Figure 1. Thus, the
rate constants for the hindered donor/acceptor pairs are readily
accommodated by Marcus theory, particularly in the endergonic
outer-sphere” electron-transfer model that predicts a quadratic
q
variation of the activation enthalpy (∆G ) with the free-energy
change (∆GET) for electron transfer.13 Thus, on the basis of
reasonable reorganization energies extracted from the Marcus
37
q
simulation in Figure 1, we calculate activation enthalpies (∆G )
-
1
that are up to 30 kJ mol higher than the experimental data
for EA in Table 4. The lack of temperature effects and the low
activation energies follow from the preequilibrium formulation
in eq 6. In other words, unusual temperature effects on the
overall reaction kinetics are expected when complex formation
precedes electron transfer.49 As a result, even strongly ender-
gonic electron transfers (as described in this study) may exhibit
negligible activation energies due to an opposing temperature
dependence of complex formation (and the follow-up electron
transfer). For example, we previously demonstrated that the
temperature independence of k2 for the CX/DUR couple in
Table 4 results from an increase of KEC and compensating
3
6
and slightly exergonic free-energy region. Contrastingly, the
rate constants for unhindered donor/acceptor combinations are
scattered in the same free-energy range and are not well fitted
to reasonable Marcus parameters. Similar deviations from
Marcus behavior were previously observed for endergonic and
slightly exergonic electron transfers from photoactivated pyrene
4
9a
(or naphthalene) to cyanoarene acceptors. Furthermore, the
absence of significant temperature dependence of the ET rate
constants was evidence for strong (donor/acceptor) complexation
4
9a, 53
prior to electron transfer.
The ET rate constants of unhindered donor/acceptor couples
are faster than those of the hindered analogues. They are also
faster than those predicted by Marcus theory, and the greatest
deviation from Marcus behavior is observed for donor/acceptor
couples that form the strongest encounter complexes. This
discrepancy is illustrated in Figure 6 by the superposition of
the ET rate constants of the unhindered donors from Figure 1
with the formation constants for encounter complexes from
Figure 5 as a function of the driving force. It is particularly
noteworthy that the fast ET rate constants observed for
unhindered donors coincide with the maximum in the formation
constant. As a result, such tight encounter complexes with strong
9
decrease of kET. On the other hand, electron transfers with
sterically encumbered donors show normal temperature effects
and reasonable activation energies (see Table 4)ssupporting
the foregoing conclusion that donor/acceptor complexes are
unimportant. In other words, bimolecular electron transfers of
sterically encumbered donors follow the predicted behaVior of
“
outer-sphere” electron transfer between weakly coupled donors
1
3
and acceptors.
B. Solvent Effect. Solvent variation on the ET rate constants
of hindered versus unhindered donors provides further distinc-
tion of the reaction mechanisms. Thus the moderate solvent
effect on unhindered donor/acceptor pairs in Table 2 is similar
to that previously observed for back-electron transfer in contact
charge-transfer character must experience a significant predis-
position toward electron transfer,9 which allows even endergonic
5
0
51
ion pairs or for intramolecular electron-transfer processes.
(51) Asahi, T.; Ohkohchi, M.; Matsusaka, T.; Mataga, N.; Zhang, R. P.;
(
49) Baggott, J. E.; Pilling, M. J. J. Chem. Soc., Faraday Trans. 1 1983,
9, 221. (b) See also: Baggott, J. E. In Photoinduced Electron Transfer;
Fox, M. A., Chanon, M., Eds.; Elsevier: New York, 1988; Part B, p 385f.
50) Asahi, T.; Ohkohchi, M.; Mataga, N. J. Phys. Chem. 1993, 97,
3132.
Osuka, A.; Maruyama, K. J. Am. Chem. Soc. 1993, 115, 5665.
(52) Gordon, J. E. The Organic Chemistry of Electrolyte Solutions;
Wiley: New York, 1975; p 99f. (b) Masnovi, J. M.; Kochi, J. K. J. Am.
Chem. Soc. 1985, 107, 7880. (c) Yabe, T.; Kochi, J. K. J. Am. Chem. Soc.
1992, 114, 4491.
7
(
1

Steric Control of Electron Transfer
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 625
Summary and Conclusions
Steric effects on the kinetics of electron transfer from hindered
and unhindered arene donors to quinone acceptors and their
temperature, solvent, and driving-force dependence reveal a
structure-induced (mechanistic) changeover. Thus unhindered
28-30
donors undergo inner-sphere electron transfer
owing to the
strong electronic coupling of donor and acceptor in a well-
defined encounter complex preceding the ET transition state.
By contrast, hindered donors show no (kinetic or spectroscopic)
evidence for a discrete encounter complex in the preequilibrium
step, and the kinetics follows outer-sphere ET behavior expected
1
3
for weakly coupled donors and acceptors. Although this
comparative study of hindered and unhindered electron donors
establishes a clear-cut (experimental) distinction between outer-
sphere and inner-sphere electron transfers, we believe there will
generally be a broad borderline region between the two
3
0e
mechanisms. Thus the idealized descriptions of inner-sphere
and outer-sphere should be taken as the two extreme ends of a
continuum of electron-transfer behavior that is tuned by the
magnitude of electronic coupling of the donor/acceptor pairs.30
Figure 6. Superposition of Figures 1 and 5 to demonstrate the
coincidence of the maximum for encounter-complex formation (b) and
the maximum deviation of the data with unhindered donors (O) from
the Marcus simulation (dashed line).
Experimental Section
electron transfers to occur at diffusion-limited rates (see Figure
Materials. Hexaethylbenzene (Acros), hexamethylbenzene, penta-
methylbenzene, durene, 1,3,5-tri-tert-butylbenzene, 3,5-di-tert-butyl-
toluene, and 1,4-di-tert-butylbenzene (Aldrich) were recrystallized from
ethanol and heptane. Mesitylene, p-xylene, and 1,2,4-trimethylbenzene
(Aldrich) were purified by distillation. Tetrachloro-p-benzoquinone
(chloranil, Aldrich) was sublimed in vacuo and recrystallized from
benzene. Toluene (reagent grade) was distilled from sodium and
benzophenone under an argon atmosphere. 1,1,4,4,5,5,8,8-Octamethyl-
5
4
1
and Table 1). In contrast, sterically encumbered donors are
subject to rather “loose” diffusive encounters with the photo-
activated quinones at intermolecular distances greater than 4.5
Åssufficient to discourage any significant charge-transfer inter-
action prior to electron transfer.1 Thus, these electron-transfer
couples represent optimum donors for purely “outer-sphere”
2a
5
5
electron transfer. On the other hand, the electron donors with
little or no steric encumbrance cannot meet the criterion for
1
,2,3,4,5,6,7,8-octahydroanthracene was synthesized according to the
58
9
literature procedure. The synthesis of 2,5-dichloroxyloquinone, as
well as that of pentaethyltoluene and triethylmesitylene, was reported
earlier. Dichloromethane, acetonitrile, chloroform, and carbon tetra-
chloride were purified according to standard procedures. The laser
photolysis experiments were carried out with the third harmonic (355
nm) output of a Q-switched Nd:YAG laser (10-ns fwhm, 22 mJ) for
the generation of the triplet quinones, and the details have been
described earlier.60
“outer-sphere” electron transfer since they form encounter com-
12a
plexes with a considerable degree of charge-transfer or inner-
sphere bonding. As such, we envisage steric encumbrances as
an important modulating influence on the changeover from
outer-sphere to inner-sphere pathways in electron-transfer mech-
anisms of aromatic donors.56 Both types of electron transfers
ultimately result in the same electron-transfer products (viz. ion-
radical pairs) without any overall changes in intermolecular
chemical bonding.5 Thus, we expect the mechanistic difference
in the degree of bonding in the precursor complexes immediately
preceding the transition states to greatly affect the electron-
transfer rate constant and its dependence on the driving force,
the temperature, and the solvent.
5
9
Determination of Ion-Radical Yields. The ion-radical yields in
7
61
Table 1 were obtained with benzophenone as the transient actinometer.
Samples of quinone/donor combinations in acetonitrile and benzophe-
none in benzene with matching absorbance at 355 nm were exposed
to the 10-ns laser pulses, and the transient absorbance of triplet
(56) In the strongly exergonic driving-force region, the electron-transfer
behavior of hindered versus unhindered donors cannot be distinguished
experimentally (i.e., all rate constants are at the diffusion limit), which may
indicate the merging of inner-sphere and outer-sphere mechanisms. Under
these energetic conditions, electron transfer is apparently faster than diffusion
even at donor/acceptor distances of 4.5 Å and more. This observation
suggests that, with increasing driving force of the electron transfer, the
effective electron-transfer distance also increases, and electron transfer
remains competitive with diffusion. [The lack of “Marcus-inverted” behavior
in bimolecular electron transfers has been explained on the basis of this
effect.3 ] Accordingly, the observation of steric effects on the rate constants
for bimolecular electron transfers in the highly exergonic driving-force
region (and possibly the observation of the Marcus-inverted region) is only
expected at driving forces and donor/acceptor distances at which the electron
transfer is the rate-limiting step following the diffusional formation of the
encounter complex. Other explanations for the lack of an inverted driving-
force dependence of bimolecular electron-transfer rate constants in the
exergonic region are cited in ref 36. The possibility of observing the inverted
region in extremely hindered organic donors/acceptors is under investigation.
(57) The electron transfer does not effect any net bond formation or bond
cleavage in either reactant. Moreover, back electron transfer in acetonitrile
solution completely restores the starting donor/acceptor pairs.
(
53) However, the electron-transfer rate constants of the pyrene/cy-
anoarene systems were found to be slower than those predicted by Marcus
theory for “outer-sphere” electron transfer. As such, reaction pathways other
than electron transfer were invoked to account for the low rate constants of
fluorescence quenching of pyrene and naphthalene by cyanoarenes. In our
study, the ion-radical yields of unity for most donor/acceptor couples (see
Table 1) rule out such alternative pathways.
(
54) A more physical view accounts for the fast rate constants in inner-
6b
sphere electron transfer by its adiabatic pathway. Thus, strong electronic
coupling between the donor and the acceptor in close (van der Waals)
distance leads to an avoided crossing of the potential surfaces. A substantial
energy gap of ∆ꢀ ) 2HAB results in adiabatic electron transfer with unit
probability. See: Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985,
8
11, 265.
55) In these electron transfers, the weak electronic coupling between
(
the sterically encumbered redox partners leads (in its extreme) to a
nonadiabatic (diabatic) electron transfer that (in the isergonic and endergonic
driving-force region) results in very slow rates. For examples of nonadiabatic
electron transfers due to steric hindrance, see: (a) Rau, H.; Frank, R.;
Greiner, G. J. Phys. Chem. 1986, 90, 2476. (b) Sandrini, D.; Maestri, M.;
Belser, P.; Von Zelewski, A.; Balzani, V. J. Phys. Chem. 1985, 89, 3675.
(58) Bruson, H. A.; Kroeger, J. W. J. Am. Chem. Soc. 1940, 62, 36.
(59) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals, 3rd ed.; Pergamon: Oxford, 1988.
(60) Bockman, T. M.; Karpinski, Z. J.; Sankararaman, S.; Kochi, J. K.
J. Am. Chem. Soc. 1992, 114, 1970.
(
c) Koval, C. A.; Pravata, R. L. A.; Reidsema, C. M. Inorg. Chem. 1984,
2
1
9
3, 545. (d) Koval, C. A.; Ketterer, M. E. J. Electroanal. Chem. 1984,
75, 263. (e) Kitamura, N.; Rajagopal, S.; Tazuke, S. J. Phys. Chem. 1987,
1, 3767.

626 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Hubig et al.
benzophenone at 525 nm (ꢀ525 ) 7,220 M- cm-1)61 was quantitatively
1
The transient decay was fitted to first-order kinetics, and the observed
rate constants (kobs) were plotted against the donor concentration. The
slope of the linear portion of the pseudo-first-order plots in Figure 3
compared with that of the chloranil anion radical at 450 nm (ꢀ450
700 M cm ) or that of the CX anion radical at 430 nm (ꢀ430
800 M cm ). For comparison, the absorption bands of the arene
)
)
-
1
-1 62
9
6
-
1
-1 9
2
yielded the second-order rate constants (k , (10%) in Tables 1-4
cation radicals were also examined to ensure sufficient separation from
the quinone anion-radical absorption.
according to eq 8. The kinetics of the curved plots of the unhindered
donors was evaluated with eq 7 by considering the preequilibrium step
in eq 6 to obtain the equilibrium constants KEC in Table 5 and Figure
5. For additional details of the kinetics analysis, see Rathore et al. in
ref 9 and references therein.
Kinetics Measurements. A 0.005 M solution of quinone in
acetonitrile (or dichloromethane, chloroform, or carbon tetrachloride)
was exposed to 355-nm (10-ns) laser excitation, and the decay of the
triplet quinone was observed at 500 nm on the ns/µs time scale in the
-4
-1
presence of varying concentrations (10 to 10 M) of aromatic donors.
Acknowledgment. We thank the National Science Founda-
(
61) Hurley, J. K.; Sinai, N.; Linschitz, H. Photochem. Photobiol. 1983,
tion and the R. A. Welch Foundation for financial support.
3
8, 9.
62) Andr e´ , J. J.; Weill, G. Mol. Phys. 1968, 15, 97.
(
JA9831002