Direct observation and structural characterization of the encounter complex in bimolecular electron transfers with photoactivated acceptors

Article abstract of DOI:10.1021/ja971188y

The encounter complex between photoexcited quirtones Q* and various aromatic donors (ArH) is observed directly by time-resolved ps spectroscopy immediately before it undergoes electron transfer to the ion-radical pair [Q(°-) ArH(°+)]. The encounter complex (EC) is spectrally characterized by distinctive (near IR) absorption bands, and its temporal evolution is established by quantitative kinetics analysis. The structural characterization of the 1:1 encounter complex [Q*, ArH] identifies the cofacial juxtaposition of the donor and acceptor moieties for optimal overlap of their π-orbitals. Further comparisons of the (excited-state) encounter complex with the corresponding (ground-state) EDA complex of aromatic donors and quinones establish its charge-transfer character, which directly relates to electron transfer within the encounter complex. The mechanistic significance of the encounter complex to bimolecular electron transfer is discussed (Scheme 1).

Full text of DOI:10.1021/ja971188y

11468
J. Am. Chem. Soc. 1997, 119, 11468-11480
Direct Observation and Structural Characterization of the
Encounter Complex in Bimolecular Electron Transfers with
Photoactivated Acceptors
R. Rathore, S. M. Hubig, and J. K. Kochi*
Contribution from the Department of Chemistry, UniVersity of Houston,
Houston, Texas 77204-5641
ReceiVed April 14, 1997X
Abstract: The encounter complex between photoexcited quinones Q* and various aromatic donors (ArH) is observed
directly by time-resolved ps spectroscopy immediately before it undergoes electron transfer to the ion-radical pair
[Q•-, ArH•+]. The encounter complex (EC) is spectrally characterized by distinctive (near IR) absorption bands,
and its temporal evolution is established by quantitative kinetics analysis. The structural characterization of the 1:1
encounter complex [Q*, ArH] identifies the cofacial juxtaposition of the donor and acceptor moieties for optimal
overlap of their π-orbitals. Further comparisons of the (excited-state) encounter complex with the corresponding
(ground-state) EDA complex of aromatic donors and quinones establish its charge-transfer character, which directly
relates to electron transfer within the encounter complex. The mechanistic significance of the encounter complex to
bimolecular electron transfer is discussed (Scheme 1).
Introduction
exciplexes,7 contact ion-radical pairs,8 solvent-separated ion-
radical pairs,9 free ion radicals, etc., which have been detected
by time-resolved absorption and emission measurements. How-
ever, the relationships among the various intermediates and their
temporal sequence in the electron-transfer mechanism are not
clear. For example, Gould, Farid, and co-workers10 showed
convincingly by quantitative fluorescence measurements that
exciplexes or excited charge-transfer complexes are common
intermediates in a series of bimolecular electron-transfer reac-
tions in different solvents. However, the emission data did not
allow the authors to establish the temporal sequence in the
formation of exciplexes and solvent-separated ion pairs in polar
solvents.
Bimolecular electron transfer between a freely diffusing
electron donor (D) and an acceptor (A) requires the intervention
of a precursor or encounter complex immediately prior to
electron exchange,1 and theoretical analyses uniformly incor-
porate such an intermolecular complex [D,A] to accommodate
the Franck-Condon requirement for ultrafast electron jump.2,3
However, experimental confirmation has lagged the theoretical
constructs and little is definitively known about the postulated
encounter complexes4sin particular, how their formation,
lifetime, structure, and electronic configuration affect the overall
electron-transfer rates.
Time-resolved picosecond spectroscopy now provides the
means to detect even short-lived (transient) species and to follow
their temporal evolution in real time.5 Indeed, we believe the
exploitation of laser-flash techniques will allow the collisional
course of a photoactived electron acceptor A* to be monitored
directly as it evolves with an electron donor (D) to the transient
complex prior to electron transfer, i.e.
The discussion of whether exciplexes or excited charge-
transfer complexes are involved in electron-transfer quenching
reactions reached its first critical stage about three decades ago,
when the Rehm-Weller relationship between the fluorescence
quenching rate constants and the electron-transfer driving force
in polar solvents was introduced.11 This free-energy correlation
was based on a purely “outer-sphere” electron-transfer model
that explicitly excluded the consideration of an excited charge-
transfer complex as an intermediate, as postulated earlier by
Mataga et al.12 The role of exciplexes or excited charge-transfer
complexes as intermediates in electron transfer processes has
been actively discussed since then,10,13,14 and the question of
A* + D h [complex] f A•- + D•+
(1)
Various experimental approaches to elucidate photoinduced
electron transfer have led to the discovery of a variety of short-
lived intermediates such as excited charge-transfer complexes,6
X Abstract published in AdVance ACS Abstracts, November 1, 1997.
(1) (a) Libby, W. F. J. Phys. Chem. 1952, 56, 863. (b) Laidler, K. J.
Can. J. Chem. 1959, 37, 138. (c) Rabinovich, E. Trans. Faraday Soc. 1937,
33, 1225. (d) For the concept of encounter-complex formation in bimolecular
reactions, see: Eigen, M. Angew. Chem., Int. Ed. Engl. 1964, 3, 1.
(2) (a) Marcus, R. A. J. Chem. Phys. 1956, 24, 966. (b) Hush, N. S.
Trans. Faraday Soc. 1961, 57, 557.
(3) For reviews, see: (a) Cannon, R. D. Electron Transfer Reactions;
Butterworths: London, 1980. (b) Bolton, J. R.; Mataga, N.; McLendon, G.
Electron Transfer in Inorganic, Organic and Biological Systems; American
Chemical Society: Washington, DC, 1991.
(8) (a) Asahi, T.; Mataga, N. J. Phys. Chem. 1989, 93, 6575. (b) Asahi,
T.; Mataga, N. J. Phys. Chem. 1991, 95, 1956. (c) Peters, K. S. In AdVances
in Electron-Transfer Chemistry; JAI Press: New York, 1994; Vol 4, p 27.
(9) (a) Gould, I. R.; Farid, S. J. Phys. Chem. 1992, 96, 7635. (b) Gould,
I. R.; Young, R. H.; Moody, R. E.; Farid, S. J. Phys. Chem. 1991, 95,
2068.
(10) Gould, I. R.; Young, R. H.; Mueller, L. J.; Farid, S. J. Am. Chem.
Soc. 1994, 116, 8176.
(11) (a) Rehm, D.; Weller, A. Ber. Bunsenges. 1969, 73, 834. (b) Rehm,
D.; Weller, A. Isr. J. Chem. 1970, 8, 259.
(4) Sutin, N. Acc. Chem. Res. 1968, 1, 225.
(12) (a) Mataga, N.; Okada, T.; Yamamoto, N. Chem. Phys. Lett. 1967,
1, 119. (b) Okada, T.; Matsui, H.; Oohari, H.; Matsumoto, H.; Mataga, N.
J. Chem. Phys. 1968, 49, 4717.
(13) (a) Weller, A.; Zachariasse, K. Chem. Phys. Lett. 1971, 10, 590.
(b) Weller, A. Z. Phys. Chem. N.F. 1982, 130, 129. (c) Baggott, J. E. In
Photoinduced Electron Transfer; Fox, M. A., Chanon, M., Eds.; Elsevier:
New York, 1988; Part B, p 385.
(5) For example see: (a) West, M. A. In Creation and Detection of the
Excited State; Ware, W. R., Ed.; Marcel Dekker: New York, 1976; p 218.
(b) Simon, J. D. ReV. Sci. Instrum. 1989, 60, 3597.
(6) Ottolenghi, M. Acc. Chem. Res. 1973, 6, 153.
(7) Gordon, M.; Ware, W. R., Eds. The Exciplex; Academic Press: New
York, 1975.
S0002-7863(97)01188-8 CCC: $14.00 © 1997 American Chemical Society

The Encounter Complex in Photoinduced Electron Transfers
J. Am. Chem. Soc., Vol. 119, No. 47, 1997 11469
Chart 1
transfers with photoactivated quinones lead to a sequence of
spectral transients that span the distinctive time scales from
picoseconds to microseconds. Time-resolved spectroscopy
enables us to establish in a conclusive way not only their
existence, but also their temporal sequence within the electron-
transfer manifold. In addition to the time-resolved absorption
measurements, the kinetic experiments and the studies of the
effects of the solvents (as well as the temperature, the electron-
transfer driving force, and the steric hindrance) on the overall
electron-transfer dynamics are utilized to characterize (spec-
troscopically, kinetically, and thermodynamically) the reaction
intermediates. A particular focus will be directed to the initial
encounter complex of the donor and the acceptor prior to
electron transfer.
Results
their relationship to ground-state electron donor-acceptor
(EDA) complexes has been raised.7,15
I. Quinones as Photoactivated Acceptors (Q*). Upon the
10-ns laser excitation at 355 nm of an argon-purged solution
of chloranil (CA) in acetonitrile, a transient absorption spectrum
with double maxima at 380 and 510 nm and a shoulder at 480
mm was observed (Figure 1A) and readily assigned to the triplet
state of chloranil (CA*).16,24 The absorption bands decayed on
the microsecond time scale with lifetimes of τ > 10 µs
(depending on the chloranil concentration). A similar transient
absorption spectrum (with maxima at 370 and 500 nm and a
shoulder at 470 nm, see Figure 1B) was obtained upon laser
excitation of 2,5-dichloroxyloquinone (CX) under comparable
conditions. On the basis of its similarity with the triplet
chloranil spectrum and its oxygen-sensitive lifetime, we assigned
this transient spectrum to the triplet state of dichloroxyloquinone
(CX*). Similar transient absorption spectra were observed in
other solvents such as dichloromethane, chloroform, and carbon
tetrachloride upon the laser excitation of CX and CA. For
acetonitrile, the maximum extinction coefficient of the dichlorox-
For our studies on the mechanism of bimolecular electron
transfer, we used excited quinones as electron acceptors and
examined their diffusional interaction with the polymethylben-
zene electron donors in Chart 1 by time-resolved (ps/µs)
absorption spectroscopy. Chloranil and the related 2,5-
dichloroxyloquinone were chosen as transient acceptors for the
following reasons: (i) Upon photoexcitation, both quinones form
long-lived (µs) excited triplet states16 which function as powerful
one-electron oxidants for a variety of aromatic donors.17,18 (ii)
For chloranil, both exciplex formation19,20 and electron transfer
to the excited state16-18 have been established. (iii) Both
quinones also form EDA complexes with arene donors in the
ground state, as established by UV-vis spectroscopy.21 The
use of (a) polymethylbenzenes of various donor strengths22 and
(b) a pair of quinones of different acceptor strength23 allowed
us to vary the driving force for electron transfer over a range
of about 10 kcal mol-1. In this study, we show that electron
yloquinone triplet (CX*) at 500 nm was ꢀ500 ) 5300 M-1 cm-1
,
(14) (a) Kikuchi, K.; Niwa, T.; Takahashi, Y.; Ikeda, H.; Miyashi, T.;
Hoshi, M. Chem. Phys. Lett. 1990, 173, 421. (b) Kikuchi, K.; Takahashi,
Y.; Katagiri, T.; Niwa, T.; Hoshi, M.; Miyashi, T. Chem. Phys. Lett. 1991,
180, 403.
(15) (a) Kobashi, H.; Okada, T.; Mataga, N. Bull. Chem. Soc. Jpn. 1986,
59, 1975. (b) Kawai, K.; Shirota, Y.; Tsubomura, H.; Mikawa, H. Bull.
Chem. Soc. Jpn. 1972, 45, 77.
and a triplet quantum yield of ΦT ) 1.0 ( 0.05 was determined
by transient actinometry25 (see Experimental Section). The
spectral and kinetic data for the triplet states of chloranil and
dichloroxyloquinone are summarized in Table 1.
II. Electron Transfer of Q* with Aromatic Donors.
Spectral Observation of the Transient Intermediate. In the
presence of high concentrations (0.03-0.3 M) of aromatic
donors (ArH), the transient spectra obtained upon 10-ns laser
excitation of chloranil or dichloroxyloquinone showed not only
the absorption bands of the corresponding triplet states (Vide
supra) but, in addition, broad absorptions were observed in the
wavelength region above 700 nm and extending beyond 900
nm. Picosecond time-resolved experiments revealed that the
growth of these new absorptions at long wavelengths was not
concomitant with the appearance of the absorption bands of Q*
at 500 nm. For example, upon the 25-ps (laser) excitation of a
solution of CA (0.005 M) and mesitylene (0.1 M) in acetonitrile,
we observed the slower growth of a broad absorption band at
λ > 800 nm to occur over a time period of about 1 ns (see
Figure 2A). In marked contrast, the 500-nm absorption band
of CA* was observed immediately attendant upon the 25-ps
laser excitation.26 Similar broad absorptions in the wavelength
region above 700 nm were observed upon laser excitation of
(16) (a) For chloranil, see: Gschwind, R.; Haselbach, E. HelV. Chim.
Acta 1979, 62, 941. (b) CX and 1,4-benzoquinone have similar reduction
potentials. However, the short triplet lifetime (<10 ns)16c of benzoquinone
precluded its use in this study. (c) Kemp, D. R.; Porter, G. Proc. R. Soc.
London, Ser. A 1971, 326, 117.
(17) (a) Kawai, K.; Shirota, Y.; Tsubomura, H.; Mikawa, H. Bull. Chem.
Soc. Jpn. 1972, 45, 77. (b) Creed, D. In Organic Photochemistry and
Photobiology; Horspool, W. M., Song, P.-S., Eds; CRC Press: Boca Raton,
FL, 1995; p 737. (c) Bockman, T. M.; Kochi, J. K. J. Chem. Soc., Perkin
Trans. 2 1996, 1633. (d) Johnston, L. J.; Schepp, N. P. J. Am. Chem. Soc.
1993, 115, 6564.
(18) Jones, G., II; Haney, W. A. J. Phys. Chem. 1986, 90, 5410.
(19) (a) Kobashi, H.; Funabashi, M.-A.; Kondo, T.; Morita, T.; Okada,
T.; Mataga, N. Bull. Chem. Soc. Jpn. 1984, 57, 3557. (b) Kobashi, H.;
Kondo, T.; Funabashi, M.-A. Bull. Chem. Soc. Jpn. 1986, 59, 2347. (c)
Kobashi, H.; Hiratsuka, K.-I.; Motegi, K. Bull. Chem. Soc. Jpn. 1988, 61,
298. (d) Levin, P. P.; Tatikolov, A. S.; Kuz’min, V. A. Bull. Acad. Sci.
USSR, DiV. Chem. Sci. 1982, 31, 890. (e) Levin, P. P.; Kuz’min, V. A.
Bull. Acad. Sci. USSR, DiV. Chem. Sci. 1986, 35, 1303. (f) Levin, P. P.;
Kuz’min, V. A. Russ. Chem. ReV. 1987, 56, 307. (g) Levin, P. P.;
Pluzhnikov, P. F.; Kuz’min, V. A. Chem. Phys. 1989, 137, 331. (h) Levin,
P. P.; Raghavan, P. K. N. Chem. Phys. Lett. 1991, 182, 663.
(20) Tahara, T.; Hamaguchi, H.-O. J. Phys. Chem. 1992, 96, 8252.
(21) Foster, R. Organic Charge-Transfer Complexes; Academic Press:
New York, 1969; p 40.
(22) Howell, J. O.; Goncalves, J. M.; Amatore, C.; Klasinc, L.;
Wightman, R. M.; Kochi, J. K. J. Am. Chem. Soc. 1984, 106, 3968.
(24) (a) Kemp, D. R.; Porter, G. J. Chem. Soc. (D) 1969, 1029. (b) Porter,
G.; Topp, M. R. Proc. R. Soc. 1970, A315, 163. (c) Kobashi, H.; Gyoda,
H.; Morita, T. Bull. Chem. Soc. Jpn. 1977, 50, 1731.
(25) Hurley, J. K.; Sinai, N.; Linschitz, H. Photochem. Photobiol. 1983,
38, 9.
(26) (a) Hubig, S. M.; Bockman, T. M.; Kochi, J. K. J. Am. Chem. Soc.
1997, 119, 2926. (b) Hilinski, E. F.; Milton, S. V.; Rentzepis, P. M. J. Am.
Chem. Soc. 1983, 105, 5193.
(23) (a) The reduction potential of chloranil is E° ) 0.02 V Vs SCE;
red
see ref 23c,d. (b) The reduction potential of dichloroxyloquinone is Er°ed
)
-0.51 V Vs SCE in dichloromethane containing 0.1 M TBA+ PF6-. (c)
Mann, C. K.; Barnes, K. K. Electrochemical Reactions in Non-Aqueous
Systems; Dekker: New York, 1970. (d) Peover, J. E. J. Chem. Soc. 1962,
4540.

11470 J. Am. Chem. Soc., Vol. 119, No. 47, 1997
Rathore et al.
Figure 1. Triplet absorption spectra of (A) chloranil and (B) 2,5-dichloroxyloquinone observed in the time interval of (top-to-bottom) 1.3-15 µs
upon the 10-ns laser excitation at 355 nm.
Table 1. Spectral and Kinetic Characterization of the Triplet
States of Chloranil (3CA) and Dichloroxyloquinone (3CX) in
Acetonitrile
excitation at 355 nm of dichloroxyloquinone/durene in aceto-
nitrile decreased simultaneously with the decay of the absorption
band of CX* with a first-order rate constant of kdec ) 5.5 ×
107 s-1 (see Figure 4), and new narrow absorption bands
centered at 330 and 430 nm grew in at the same rate. The new
absorptions were readily assigned to the anion-radical of
dichloroxyloquinone (CX•-) by spectral comparison with that
generated independently.28 Subsequently, the 330- and 430-
nm absorption bands of the anion radical of dichloroxyloquinone
slowly decayed by second-order kinetics on the microsecond
time scale.
parameter
3CA
3CX
λ
T-T, max [nm]a
510
500
5300
1.00 ( 0.05
>2
b
ꢀT-T, max [M-1 cm-1
]
7570c
d
ΦT
0.98 ( 0.05c
>10
τT [µs]e
a Wavelength maximum of the triplet absorption band. b Triplet
extinction coefficient. c Reference 16. d Triplet quantum yield. e Triplet
lifetime.
Similar time-resolved spectra on the nanosecond time scale
were obtained upon laser excitation of dichloroxyloquinone in
the presence of hexamethylbenzene (0.01 M) in acetonitrile, and
upon photoexcitation of chloranil in the presence of either
mesitylene or xylene in acetonitrile. In all cases, the initial
spectrum upon the 10-ns laser excitation showed the diagnostic
absorption band (above 700 nm) of the transient intermediate
of the quinone and the arene (ArH). This absorption decayed
by first-order kinetics on the nanosecond time scale concomitant
with the 500-nm absorption of Q* to generate the spectrum of
the quinone anion radical (Q•-). In all cases, we assigned the
decay of the quinone/arene intermediates to the coproduction
of ion radicals (Q•- and ArH•+). [The absorption bands of the
accompanying cation radicals of the arene donors (ArH•+) were
not directly monitored owing to their low extinction coefficients
and spectral overlap with the strongly absorbing quinone anion
radicals.29] On the basis of the extinction coefficients of
CX in the presence of the various aromatic donors in Chart 1.
These new infrared (NIR) bands were not only observed in
acetonitrile, but also in other less polar solvents such as
dichloromethane, chloroform, and carbon tetrachloride. For
example in chloroform, the 25-ps laser excitation of CX and
durene led to new absorption in the NIR region above 700 nm,
as shown in Figure 2B. Strikingly, the slow growth of the broad
NIR absorption at 70, 570, and 1070 ps occurred under
conditions in which the local band of CX* centered at 500 nm
remained unchanged27sas underscored by the enlargements in
the insets for both absorption bands. Although the spectral
maxima of the near-IR absorptions were beyond our detector
limit (see Experimental Section), Figure 3 clearly shows the
bathochromic shifts in the absorption onsets to occur in the
following order: CX/DUR < CX/HMB < CA/MES < CA/
XYL. The new absorptions at wavelengths above 700 nm are
hereinafter ascribed to the transient intermediate (TI) between
the excited quinone and the arene donor.
From the rise time of the near-infrared absorption band in
Figure 2A, we determined the first-order rate constant for the
formation of the transient intermediate from chloranil and
mesitylene as kobs = 1 × 109 s-1. For a mesitylene concentra-
tion of 0.1 M, we calculated the bimolecular rate constant for
TI formation to be kTI = 1 × 1010 M-1 s-1, which is on the
time scale of diffusion-controlled processes. TI formation was
also observed on the early nanosecond time scale in CX
solutions containing durene (Figure 2B), and the diffusion-
controlled values for kTI were found to be the same for both
quinones and other donors (DUR, HMB, and XYL).
dichloroxyloquinone (ꢀ430 ) 6800 M-1 cm-1 28
and chloranil
)
(ꢀ450 ) 9700 M-1 cm-1),30 the ion radical yields of Φion ) 1.0
( 0.05 were determined for various quinone/arene combinations
in acetonitrile with benzophenone as the transient actinometer25
(see Experimental Section).
IV. The Fate of the Transient Intermediate in Solvents
of Low Polarity. In the less polar solvents such as dichlo-
romethane, chloroform, or carbon tetrachloride, the decay of
(28) The CX•- anion radical was generated by electron-transfer quenching
of CX* with 4,4′-dimethylbiphenyl, dibenzofuran, and hexamethylbenzene.
The electron-transfer quenching experiment with 4,4′-dimethylbiphenyl,
which yielded the biphenyl cation radical29 and the dichloroxyloquinone
anion radical, was also used to determine the extinction coefficient of CX•-
at 430 nm to be ꢀ430 ) 6800 ( 100 M-1 cm-1
.
III. The Decay of the Transient Intermediate with the
Concomitant Formation of Ion Radicals in Acetonitrile. The
absorption spectra described in the previous section decayed
rapidly on the nanosecond time scale. For example, the
absorption band (above 700 nm) observed upon the 10-ns laser
(29) (a) It is noteworthy that the absorption spectra of biphenyl and
naphthalene cation radicals were sufficiently red-shifted29b to allow
simultaneous observation with the chloranil anion radical in a roughly 1:1
molar ratio. In these cases, however, the NIR absorption of the encounter
complex was not sufficiently resolved to allow quantitative spectral
evaluation. (b) Shida, T. Electronic Absorption Spectra of Radical Ions;
Elsevier: Amsterdam, 1988.
(27) The temporal behavior of CA* (centered at 510 nm) decoupled from
the slower formation of the NIR band was obscured in acetonitrile by the
rapid followup reactions (i.e. ion-pair formation, Vide infra) that occurred
on the same time scale.
(30) The absorption maximum of chloranil anion radical in acetonitrile
is centered at 450 nm. See: Andre´, J. J.; Weill, G. Mol. Phys. 1968, 15,
97.

The Encounter Complex in Photoinduced Electron Transfers
J. Am. Chem. Soc., Vol. 119, No. 47, 1997 11471
Figure 2. Temporal evolution of the near-IR band at λ > 700 nm from (A) 0.002 M chloranil and 0.1 M mesitylene in acetonitrile and (B) 0.002
M CX and 0.1 M durene in chloroform following the application of the 25-ps laser pulse at 355 nm. The insets (×5) in part B establish the
invariance of the local band of CX* at 500 nm during the slower growth period of the CT absorption at λ > 700 nm in the time interval among
(a) 70 ps, (b) 570 ps, and (c) 1.07 ns.
Figure 3. Progressive bathochromic shift of the near-IR absorptions with decreasing energy gap (∆E°) between the oxidation potential of the
aromatic donor and the reduction potential of the quinone acceptor in chloroform.
the transient intermediates did not result in the formation of
the ion-radical pair Q•- and ArH•+. Instead, the protonated
quinone anion radical (Q-H•)18 was the species observed after
the decay of the transient intermediate. For example, im-
mediately upon 10-ns laser excitation of dichloroxyloquinone/
durene in chloroform, the spectrum of the transient intermediate
was observed with an absorption maximum at 500 nm and a
broad tail absorption extending beyond 900 nm (Figure 5). The
TI absorption decayed completely to baseline within 400 ns (kdec
) 1.6 × 107 s-1) and ultimately led to a new spectrum with a
very narrow absorption band at 420 nm (see Figure 5). The
420-nm absorption was readily assigned to the protonated form
of the dichloroxyloquinone anion radical (CX-H•) by spectral
comparison with that of the chloranil analogue (Vide infra).
A. Solvent Variation. Similar transient spectra were
obtained upon the quenching of CX* with hexamethylbenzene
and in other solvents of low polarity such as dichloromethane
or carbon tetrachloride. In the case of chloranil, the protonated
chloranil anion radical (CA-H• or hydrochloranil) was ob-
served.18 Hydrochloranil yields could be estimated on the basis
of its extinction coefficient at 435 nm (ꢀ435 ) 7700 M-1 cm-1).31
For example, the quenching of CA* by mesitylene led to the
CA-H• in a yield of Φ ) 0.7 ( 0.1, which was invariant in
dichloromethane, chloroform, and carbon tetrachloride. Simi-
larly, the yields of protonated anion radicals generated from
(31) Wong, S. K.; Fabes, L.; Green, W. J.; Wan, J. K. S. J. Chem. Soc.,
Faraday Trans. 1 1972, 68, 2211.

11472 J. Am. Chem. Soc., Vol. 119, No. 47, 1997
Rathore et al.
pairs in chloroform by the addition of salt, and ion radical yields
of 0.7 and 0.3 were obtained for the dichloroxyloquinone/
hexamethylbenzene and chloranil/mesitylene pairs, respectively.
V. Evaluation of the Electron-Transfer Kinetics of Q*.
For the kinetic studies, the spectral decays in Figures 4 and 5
were monitored on the nanosecond/microsecond time scale as
a function of the concentration of aromatic donor, solvent,
temperature, and added salt as follows.
1. Dependence on Donor Concentration. 2,5-Dichlorox-
yloquinone (0.002 M) was photoexcited in chloroform at 355
nm with a 10-ns laser pulse, and the first-order decay (kobs) of
the transient absorbance at 500 nm was monitored on the
nanosecond/microsecond time scale as a function of added
durene. For concentrations up to 0.005 M, the kinetics plots in
Figure 6A revealed a more or less linear dependence of kobs on
the durene concentration. However, at higher concentrations
(>0.005 M), the decay rates did not increase linearly with
increasing durene concentration, and a limiting plateau value
of kmax ) 1.6 × 107 s-1 was obtained for concentrations >0.2
M (see the triangles in Figure 6A). Similar (curved) plots were
also obtained in acetonitrile, dichloromethane, and carbon
tetrachloride (see Figure 6A). Most notably, the degree of
curvature and the absolute value of kmax for the plateau varied
dramatically with the solvent.
The limiting values of kobs shown in Figure 6A at high donor
concentrations were diagnostic of a preequilibrium intermedi-
ate32 between the excited quinone and the aromatic donor. As
such, the asymptotic behavior could be readily linearized in a
double-reciprocal presentation, from which the formation con-
stant KTI of the transient intermediate between quinone and
donor and its intrinsic decay constant (k) could be readily
extracted, i.e.
Figure 4. Transient absorption spectra from 0.002 M CX and 0.04 M
DUR in acetonitrile following the 10-ns laser excitation at 355 nm.
The absorption bands of CX* with λmax ) 500 nm and [CX*, DUR]
at λ > 700 nm (see inset) decayed with the same rate constant (k )
5.4 × 107 s-1) as the first-order appearance of CX•- at λmax ) 330 and
430 nm.
1
kobs
1
k
1
)
+
(2)
KTIk[D]
[Note that the slow natural decay of Q* (k0 in the absence of
donor)16 was irrelevant, since k0 , k.] The values for KTI and
k, as evaluated from the double-reciprocal plots such as those
in Figure 6B, are listed in Table 2 for various quinone/arene
combinations. Since the spectral decays clearly led to ion-
(32) For a general kinetics description, see: (a) Espenson, J. D. Chemical
Kinetics and Reaction mechanisms, 2nd ed.; McGraw-Hill: New York,
1995; p 89f. (b) Connors, K. A. Chemical Kinetics: The Study of Reaction
Rates in Solution; VCH: New York, 1990; p 101. (c) See also: Kobashi,
H.; Okada, T.; Mataga, N. in ref 15a. (d) The ground-state EDA complex
in Table 4 cannot contribute significantly to the curvature in Figure 6A
since the formation constant KEDA < 7 M-1 indicated that its concentration
was diminishingly small.
(33) Since the Coulombic work terms and solvation energies from all
quinone/arene combinations are taken to be invariant (and omitted from eq
4), the simple approximation for ∆GET underscores the dominant driving-
force dependence on the electrode potentials.
Figure 5. Transient absorption spectra from 0.002 M CX and 0.3 M
DUR in chloroform following the 10-ns laser excitation at 355 nm to
show the absorption bands of CX* and [CX*/DUR] decaying at
identical rates (k ) 1.4 × 107 s-1). After an interval of 200 ns, only
the narrow absorption band of the protonated anion radical (CX-H•
with λmax ) 420 nm) is observed.
dichloroxyloquinone did not vary significantly in dichlo-
romethane, chloroform, and carbon tetrachloride (Φ ) 0.5-
0.7).
(34) Iverson, D. J.; Hunter, G.; Blount, J. F.; Damewood, J. R., Jr.;
Mislow, K. J. Am. Chem. Soc. 1981, 103, 6073.
(35) Hubig, S. M.; Rathore, R.; Kochi, J. K. To be submitted for
publication.
B. Salt Effects. Upon the photoexcitation of dichlorox-
yloquinone in the presence of durene dissolved in chloroform
containing 0.1 M tetra-n-butylammonium hexafluorophosphate
(TBA+PF6-), we initially observed the formation of the transient
intermediate with its characteristic near-IR absorbance beyond
700 nm, as previously observed in the absence of added salt.
However, the decay of the transient intermediate in the presence
of added salt did not lead to protonated quinone anion radicals
(as observed in the absence of TBA+PF6-), but the quinone
(36) Benesi, H. A.; Hildebrand, J. H. J. Am. Chem. Soc. 1949, 71, 2703.
(37) (a) 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. J. Phys. Chem.
1952, 56, 801. (d) Mulliken, R. S.; Person, W. M. Molecular Complexes;
Wiley: New York, 1969. (e) In Mulliken theory, the charge-transfer
absorption band varies as hc/λCT ) IP - EA - ω, where IP is the ionization
potential of the donor, EA is the electron affinity of the acceptor, and ω is
the electrostatic work term of the ion pair [D•+,A•-]. This relationship can
be approximated as hc/λCT ) ∆E + constant, where ∆E is the energy gap
(in solution) for structurally related D/A pairs. Mulliken theory also predicts
KEDA to increase with increasing donor and acceptor strengths.
(38) A detailed analysis of these steric effects has been presented
separately. See: Rathore, R.; Lindeman, S. V.; Kochi, J. K. J. Am. Chem.
Soc. 1997, 119, 9393.
anion radical itself (CX•-) was detected with yields of Φion
=
0.7. A similar transformation from the transient intermediate
to the ion radical could be achieved with other donor/acceptor

The Encounter Complex in Photoinduced Electron Transfers
J. Am. Chem. Soc., Vol. 119, No. 47, 1997 11473
Figure 6. (A) Asymptotic behavior in various solvents of the rate constant (kobs) for electron transfer from durene to CX* with increasing donor
concentration and (B) its double-reciprocal evaluation according to eq 2. The solid lines in part A represent computer simulations of the data points
with the values of KTI and k from Table 2. The inset in part A magnifies (×10) the acetonitrile data to establish the nonlinear trend.
Table 2. Kinetic Data for Electron Transfer from Arene Donors to
Photoactivated Quinones
According to eq 3, the bimolecular rate constant (k2) could be
determined either empirically as the slope of the initial (linear)
part of the kinetics plots or arithmetically as the product of KTI
and kET, both being extracted from the double-reciprocal analysis
(eq 2) of the entire (curved) plots. Table 2 summarizes the
values of k2, KTI, and kET for four combinations of quinones
and arene donors in four solvents. In acetonitrile, significant
curvature of the kobs/[D] diagrams was only obtained in the case
of the dichloroxyloquinone/durene system, which allowed us
to extract values of KTI and kET according to eq 2. The other
quinone/donor combinations did not exhibit sufficient curvature
in the kobs Vs [D] plots to obtain reliable values of KTI and kET
in acetonitrile. The lack of sufficient curvature in these cases
arose from the plateau values of kobs which severely exceeded
the time resolution of the 10-ns laser pulse (kmax . 108 s-1).
solvents
quinone/arene
combinationa
kinetic
parametersb
CH3CN CH2Cl2 CHCl3 CCl4
CX/DUR
CX/HMBg
CA/MES
CA/XYL
k2 [108 M-1 s-1
]
]
]
]
25.6
(25.5)
15
17
51.4
3.0
(3.8)
14
2.7
55
(73)
67
11
12
(11)
30
3.6
12
(8.0)
10
2.3d
(2.9)
18e
1.6f
23
(37)
71
5.2
6.7
(11.3) (3.9)
54
2.1
6.6
(6.8)
19
1.1
(1.4)
24
0.6
23
(36)
202
1.8
2.5
K
k
TI [M-1
ET [107 s-1
k2 [108 M-1 s-1
]
]
K
k
TI [M-1
ET [107 s-1
k2 [108 M-1 s-1
]
c
c
39
]
K
k
TI [M-1
ET [107 s-1
k2 [108 M-1 s-1
]
c
c
54
49
]
0.8
1.8
(2.9)
21
TI [M-1
ET [107 s-1
]
c
c
2. Solvent and Driving-Force Dependence of the Kinetic
Parameters. As shown in Table 2, each kinetic parameter (Viz,
k2, KTI, and kET) exhibited its particular solvent dependence.
For example, the KTI values increased by a factor of 2-3 with
decreasing solvent polarity. On the other hand, the solvent
dependence of kET in Table 2 followed the opposite trend as
compared to that of KTI. Thus with increasing solvent polarity,
the values of kET showed significant increases. This effect was
best illustrated with the dichloroxyloquinone/durene pair for
which we could compare kET data in all four solvents. Thus,
K
k
]
8.0
3.6
1.4
a CX ) 2,5-dichloroxyloquinone, CA ) chloranil, HMB ) hexam-
ethylbenzene, DUR ) durene, MES ) mesitylene, XYL ) p-xylene.
b See text. The k2 values are extracted from the initial slope of the
kinetics plots. The values in parentheses are the calculated rate constants
(see text). c Not determined owing to insufficient curvature of kinetics
plots (see text). d In the presence of salt: 2.1 (3.6) × 108 M-1 s-1.e In
the presence of salt: 15 M-1
.
f In the presence of salt: 2.4 × 107 s-1
.
g The k2 values for the hindered analogue hexaethylbenzene are 9.7 ×
108, 1.6 × 108, 0.4 × 108, and 0.04 × 108 M-1 s-1 in acetonitrile,
dichloromethane, chloroform, and carbon tetrachloride, respectively.
k
ET increased by a factor of 28 going from carbon tetrachloride
to acetonitrile. The solvent effect on kET was very similar for
all four quinone/donor combinations in Table 2. The k2 values
in Table 2 followed the same solvent trend as described for
radical pairs (compare Figure 4), the kinetic process must be
derived from electron transfer, i.e. k ) kET
.
For low donor concentrations of [D] < 0.005 M, a linear
correlation between the rate constant (kobs) for first-order decay
and the donor concentration ([D]) was found, which also allowed
us to determine the bimolecular rate constant (k2) from the initial
slope of the kinetics plots. For this concentration range, eq 2
was simplified to a linear correlation between kobs and [D] to
reveal the relationship between k2, KTI, and kET as presented in
eq 3:
k
ET. The latter was not surprising if one considers the fact that
k2 represented the product of kET and KTI (see eq 3). In fact,
the empirical data for k2 in Table 2, which were obtained from
the initial slopes of the plots, matched the computed values of
KTIkET (given in parentheses) fairly well.
To analyze the driving-force dependence of kET, the free-
energy change (∆GET) on proceeding from the transient
intermediate to the ion radical was evaluated for each donor-
acceptor pair in Table 2. As a reasonable approximation for
kobs ) KTIkET[D] ) k2[D]
(3)

11474 J. Am. Chem. Soc., Vol. 119, No. 47, 1997
Rathore et al.
Table 3. Temperature Dependence of the Kinetic Parameters for
the Electron Transfer from Durene to Photoactivated
Dichloroxyloquinone in Acetonitrile
temp (K)
k2a (108 M-1 s-1
)
KTIb (M-1
)
kETc (108 s-1
)
295
277
253
25.6
26.2
26.0
15
65
90
1.7
0.6
0.5
a Bimolecular rate constant determined from the slope of the initial
(linear) portion of the kinetics plot. b TI formation constant. c First-
order rate constants of the decay of the transient intermediate by electron
transfer.
∆GET, we took the sum of the triplet energy (ET ) 2.13 eV39)
of the corresponding quinone and the potential difference
between the oxidation potential (E° ) of the arene donor and
ox
the reduction potential (Er°ed) of the quinone acceptor,33 i.e.
-∆GET ) ET - E° + E° + constant
(4)
ox
red
Figure 7. Salt effect on the rate constants (kobs) for electron transfer
Although there were only four data points per solvent, a
consistent pattern was noticeable in that kET increased by a factor
of 2-3 in going from the most endothermic pair (Viz, dichlor-
oxyloquinone/durene: ∆GET ) +0.21 eV) to the most exo-
thermic pair (Viz, chloranil/xylene: ∆GET ) -0.09 eV) in Table
2. We also noted a peculiar driving-force dependence of KTI.
Thus, the largest TI formation constants were found for the
transient intermediates derived from either chloranil/mesitylene
or dichloroxyloquinone/hexamethylbenzene (see Table 2). In-
terestingly, these were the two quinone/arene combinations for
which ∆GET was equal or very close to zero (see Discussion).
3. Temperature Dependence of the Electron-Transfer
Kinetics. Electron-transfer quenching of photoexcited dichlorox-
yloquinone by durene in acetonitrile was monitored at 4 and
-20 °C, and the data were compared with those obtained at
room temperature. The kobs Vs [D] plots at the lower temper-
atures were also curved in a manner similar to that obtained at
room temperature. The kinetic parameters KTI and kET were
extracted from the double-reciprocal fit of the kinetic data
according to eq 2. As shown in Table 3, the value of KTI
increased and that of kET decreased significantly at the lower
temperatures. However, the bimolecular rate constant [k2 as
determined from the slope of the initial (linear) portion of the
plot] showed no variation over the entire temperature range of
40 °C. The temperature dependence of the formation constants
(KTI) was evaluated from the van’t Hoff relationship, and a
formation enthalpy of ∆HTI ) -5.6 kcal mol-1 was extracted
from the plot of ln KTI Vs the reciprocal temperature. The free-
energy change at the medium temperature of 277 K was ∆GTI
) -RT ln KTI ) -2.3 kcal mol-1, and the entropy change of
TI formation was calculated to be ∆STI ) -12 cal mol-1 K-1
from the Gibbs equation.
4. Effects of Added Salt on Electron-Transfer Kinetics.
As described above, the addition of an inert salt to solvents of
low polarity led to quinone anion radicals (Q•-) as opposed to
protonated anion radicals (Q-H•) in the absence of salt.
However, the kinetics were remarkably unaffected by the
presence of salt. For example, the addition of 0.1 M tetra-n-
butylammonium hexafluorophosphate to a solution of dichlorox-
yloquinone and durene in chloroform raised the plateau value
of kobs only slightly, and the formation constant KTI was not
affected at all by the added salt. In Figure 7, the corresponding
acetonitrile data were replotted to demonstrate the remarkably
different effects of solvent polarity and addition of salt on the
electron-transfer kinetics.
-
from durene to CX* in (b) chloroform containing 0.1 M TBA+PF6
relative to that in either (O) chloroform alone or (9) acetonitrile alone.
The solid lines represent the computer simulations of the data points
according to eq 2 with the values of KTI and k from Table 2.
5. Steric Effects on Electron-Transfer Kinetics. To study
the structural effects of the donor on the electron-transfer
kinetics, we compared the kobs Vs [D] plots and the time-resolved
spectra of photoexcited quinones with hindered and nonhindered
polymethylbenzenes of similar oxidation potentials. Whereas
the dichloroxyloquinone/durene data showed curved kinetics
plots (see Figure 6A), the correlation between kobs and the donor
concentration in the case of the highly hindered hexaethylben-
zene34 or octamethyloctahydroanthracene35 remained remarkably
linear even to very high donor concentrations. Moreover, the
transient absorption spectra exhibited only the absorption bands
of CX*, and no transient (near IR) absorptions beyond 700 nm
were detected. Similar results were obtained in the comparison
of other pairs of hindered and nonhindered aromatic donors. In
all experiments with the sterically hindered donors, we did not
detect any of the diagnostic signs for the formation of the
transient intermediate, as deduced from the absence of transient
absorption bands at long wavelengths and the lack of curvature
in the kinetics plots. Thus, neither spectroscopic nor kinetic
evidence was observed for the formation of the transient
intermediate from hindered donors prior to electron transfer.
VI. Ground-State Formation of Electron Donor-Accep-
tor (EDA) Complexes of Quinones with Arene Donors.
Chloranil as well as dichloroxyloquinone both formed (ground-
state) electron donor-acceptor complexes21 with the various
polymethylbenzenes in Chart 1, as revealed by the typical
charge-transfer absorption bands observed immediately upon
mixing solutions of quinones and arenes. It is particularly
important to emphasize that the spectral changes in Figure 8
showed that the intensity of the CT absorption band at λCT
)
522 nm increased with donor concentrations under conditions
in which the local band of chloranil at λmax ) 374 nm was
singularly invariant. In other words, the ground-state EDA
complexes consisted of the superposition of two absorption
bandssone from the quinone itself and one from the Q/D charge
transfersas originally predicted by Mulliken.37 Table 4 lists
the spectral maxima (λCT) of the charge-transfer absorption
bands and the formation constants (KEDA) for the electron
donor-acceptor complexes of various quinone/arene combina-
tions in four solvents determined by the Benesi-Hildebrand
method36 (see Experimental Section). For all donor-acceptor
combinations, a significant decreasing trend of the formation
(39) Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of Photochem-
istry; Dekker: New York, 1993.

The Encounter Complex in Photoinduced Electron Transfers
J. Am. Chem. Soc., Vol. 119, No. 47, 1997 11475
leads to its efficient excitation to the triplet state Q* with a
sizable energy of ET = 50 kcal mol-1 39
For chloranil and a
.
series of related quinones, Q* is spontaneously generated with
unit efficiency,16 owing to ultrafast intersystem crossing (kISC
= 1011 s-1) from the singlet manifold.26 In other words, Q* is
quantitatively formed as a discrete electron acceptor within 50
ps, which allows sufficient time resolution for the fast electron-
transfer kinetics to be measured via the temporal evolution of
its characteristic absorption spectrum (Figure 1), as follows.
II. Direct Observation and Characterization of the
Encounter Complex. The transient absorption spectrum of Q*
is strongly affected by the presence of aromatic donors
(ArH)srapidly leading to broad unstructured absorptions of the
transient intermediate in the near-infrared (wavelength) region
beyond 700 nm shown in Figure 3. Most notably, the
bathochromic shifts of the near-IR bands that follow the trend
in ∆GET (as defined in eq 4) point to the charge-transfer (CT)
character of the new electronic transition, in accord with
Mulliken theory.37 Indeed, similar NIR absorptions were
previously ascribed to exciplexes and spectrally assigned to
charge-transfer transitions by Tsubomura, Mataga, and co-
workers,15 and confirmed by time-resolved resonance Raman
spectroscopy.20 In other words, the transient intermediate (TI)
identified in this study is identical to these types of exciplexes,
and both are best defined in this context as intermolecular
complexes of excited quinones with aromatic donors which
exhibit spectrally observable charge-transfer character. Ac-
cordingly for the purpose of a clear definition,40 we refer
hereinafter to the transient intermediate as the encounter complex
(EC) derived from the excited quinone and the aromatic donor,
i.e.
Figure 8. Charge-transfer absorption band at λmax ) 522 nm of the
EDA complex of 0.004 M chloranil with increasing concentrations (10-
50 mM) of hexamethylbenzene in dichloromethane from ref 38. (The
dashed curve is for 0.004 M chloranil alone.)
Table 4: Formation Constants (KEDA), Products KEDAꢀCT, and the
Charge-Transfer Absorption Maxima (λCT) of Various Electron
Donor-Acceptor (EDA) Complexes of Quinones and
(Poly)methylbenzenes in Different Solventsa
arene/quinone
solvent
combinationb
EDA complex
(λCT, nm)
parametersc CH3CN CH2Cl2 CHCl3 CCl4
CA/HMB (522)
CA/DUR (476)
CA/MES (432)
CA/XYL (470)
CX/HMB (476)
CA/HEB (e)
KEDA
0.8
5300
0.6
1200
0.3
530
0.5
350
0.3
350
0
2.8
7700
1.3
2200
0.8
1200
0.5
460
0.5
460
2.4
6100
1.3
2300
0.7
1200
d
6.9
18000
3.9
5000
1.9
2500
d
K
EDAꢀCT
KEDA
K
EDAꢀCT
KEDA
K
EDAꢀCT
KEDA
K
EDAꢀCT
d
d
KEDA
0.6
490
d
1.7
1400
d
K
EDAꢀCT
KEDA
<0.008
<10
kEC
K
EDAꢀCT
0
d
d
Q* + ArH 8 [Q*, ArH]
encounter
(5)
a The values of KEDA and KEDAꢀCT were extracted from Benesi-
Hildebrand plots, see Experimental Section. b HMB ) hexamethyl-
benzene, HEB ) hexaethylbenzene, DUR ) durene, MES ) mesity-
lene, XYL ) xylene, CA ) chloranil, CX ) 2,5-dichloroxyloquinone.
c KEDA ) formation constant of the EDA complex in M-1. For the
product KEDAꢀCT, see Experimental Section. d Not determined. e No new
absorption band observed.
complex
As such, kEC for the formation of the encounter complex in eq
5 coincides with the experimentally measured value of kTI = 1
× 1010 M-1 s-1 for the diffusional rate constant of the transient
intermediate from CA* and mesitylene (Vide supra). Most
importantly, the absorption spectrum which is assigned to this
encounter complex with its near-IR band at λCT > 800 nm
includes the local absorption band of Q* in the visible region
(∼500 nm), as established in Figures 2B, 4, and 5. Such spectral
features are highly reminiscent of those present in the ground-
state EDA complex (see Figure 8). Indeed, the striking spectral
comparison of the encounter complex in Figures 2B and 4 with
the ground-state EDA complex in Figure 8 demonstrates that
the local bands of both quinones (Q* and Q) remain essentially
unperturbed by complex formation. Moreover, the spectrum
of the encounter complex (consisting of the local quinone and
the NIR CT band) is clearly distinguished from the spectrum
of the corresponding ion-radical pairs (consisting of the
composite absorptions of Q•- and D•+ in Figure 4). Most
notably, the fast diffusion-controlled formation of the encounter
complex is independent of added salt to also distinguish it from
the appearance of the ion-radical pair (in chloroform).
constants KEDA was noted with increasing solvent polarity. We
also found the strongest complex to be formed between chloranil
and hexamethylbenzene, in accord with the dependence of KEDA
on the oxidation and reduction potentials of the electron donor
and acceptor moieties.37
To study structural effects on the formation of EDA com-
plexes, we compared the formation of the ground-state com-
plexes from aromatic donors of similar oxidation potentials, but
different steric requirements. For example, hexamethylbenzene
and hexaethylbenzene have almost identical oxidation poten-
tials.22 However, hexamethylbenzene readily formed the various
EDA complexes with aromatic acceptors in Table 4, whereas
hexaethylbenzene showed no charge-transfer absorption when
mixed with chloranil in dichloromethane, even at concentrations
close to saturation (1.8 M). Similar steric effects were observed
when other pairs of hindered and nonhindered donors were
compared.38
III. Charge-Transfer Character of the (Excited-State)
Encounter Complex in Comparison with the Electron
Discussion
(40) The term “exciplex” is frequently used in a wider, but rather
ambiguous, way to include both charge-transfer complexes in the excited
state (as defined here) as well as ion-radical pairs.19,20 Since we can clearly
distinguish excited CT complexes from contact ion pairs spectroscopically,
we prefer encounter complex to establish the more general connection of
this work to other (adiabatic) electron-transfer processes. For further
definitions of encounter complexes, exciplexes, collision complexes, and
ion-radical pairs, see also: Kavarnos, G. J.; Turro, N. J. Chem. ReV. 1986,
86, 401.
The use of quinones (Q) and aromatic donors (ArH) in Chart
1 allows the temporal course of electron transfer to yield Q•-
and ArH•+ in eq 1 to be precisely monitored and the sequence
of reactive intermediates to be directly observed step by step:
I. Spontaneous Generation of the Electron Acceptor Q*.
The application of a 25-ps laser pulse to a quinone (in solution)

11476 J. Am. Chem. Soc., Vol. 119, No. 47, 1997
Rathore et al.
42
44
Chart 2. Solvent Trends in KEDA and KEC
Chart 3. Driving-Force Dependence of KEC and KEDA
ground-state
solvent
(polarity) kcal mol-1
KEDA, M-1
KEC, M-1
excited-state
CH3CN
CH2Cl2
CHCl3
CCl4
(46)
(41)
(30)
(32)
0.3
0.5
0.6
1.7
<15
67
71
Q/ArH
∆Ggs, eV
KEDA, M-1
∆Ges, eV
KEC, M-1
CA/HMB
CA/DUR
CA/XYL
CA/MES
CX/HMB
CX/DUR
+1.60
+1.81
+2.04
+2.09
+2.13
+2.34
2.8
1.3
0.5
0.8
0.5
0.3
-0.53
-0.32
-0.09
-0.04
0
<5
<5
10
30
67
14
202
Donor-Acceptor (Ground-State) Complex. The charge-
transfer absorption of the encounter complex is far red-shifted
beyond λCT(EC) > 900 nm,41 as compared to the CT absorption
in the corresponding ground-state EDA complex with λCT(EDA)
= 480 nm in Table 4. Furthermore, the formation constants of
the encounter complex KEC ()KTI) in Table 2 are several orders
of magnitude larger than the corresponding values KEDA in Table
4. Such differences between the excited-state and ground-state
charge-transfer interactions derive naturally from Mulliken
theory,37 if one considers that quinones are much stronger
electron acceptors in the excited state (Q*) as compared to the
ground state (Q). Indeed, the smaller gap between the donor
and acceptor potentials in [Q*, ArH], relative to that in [Q, ArH]
as evaluated by the quinone triplet energy of roughly 50 kcal
+0.21
acetonitrile competes with the aromatic donor in complex
formation, and leads to lower values of both KEC and KEDA
43
.
B. Energy-Gap Dependence of the Formation Constant.
The charge-transfer character of the (excited-state) encounter
complex vis-a`-vis the ground-state EDA complex is also
manifested in the comparative driving-force dependence of KEC
and KEDA summarized in Chart 3.44 Typically, Mulliken
describes weak charge-transfer complexes as the following: ΨN-
(AD) ) aψ0(A,D) + bψ1(A- D+) + ..., where ψ0 represents
the “no bond” function and ψ1 is the dative (ion pair) or charge-
transfer contribution; and thus the charge-transfer stabilization
increases in measure with the ratio of b/a.37 In the ground-
state EDA complex of quinone and arene, the charge-transfer
contribution of ∼b/a is diminishingly small, and it is reflected
in rather limited values of KEDA that decrease with ∆Ggs in Chart
3. By comparison, the larger magnitudes of KEC in the excited-
state encounter complex derive from the significantly lower
values of ∆Ges in Chart 3 that arise from quinone promotion to
Q* by ET = 50 kcal mol-1. The latter results in a closer
matching of the energy levels of the neutral (“local” excited)
state with the ion-pair (charge-transfer excited) state, i.e. {Q*,
ArH} T {Q•-, ArH•+}. Indeed, the largest value of KEC in
Chart 3 obtains for the CX*/HMB pair with ∆Ges ) 0 to allow
optimal mixing45 of {CX*, HMB} for which ET ) 2.13 eV39
with the isoenergetic ion-pair state {CX•-, HMB•+} with an
energy of E° - E° ) 2.13 eV.46
mol-1 39
, predicts a corresponding red shift in the CT absorption
band to longer wavelength and a concomitant enhancement in
the formation constant of the excited-state complex.15 The
further confirmation of the charge-transfer character of these
absorptions is shown in Figure 3 by the progressive bathochro-
mic shift of the NIR absorptions with decreasing energy gap
(∆E°) between the oxidation potential of the aromatic donor
and the reduction potential of the quinone acceptor according
to the expectations of Mulliken theory.37e Accordingly, let us
consider other ways in which the excited-state and ground-state
complexes can be compared.
A. Solvent Dependence of the Formation Constants. The
less polar solvents like carbon tetrachloride favor the formation
of both the encounter complex (EC) as well as the EDA
complex, as shown by the parallel trends in KEC and KEDA in
Chart 2.42 Since the solvent dependence for complex formation
is a measure of the difference between the solvation energy of
the donor-acceptor complex and of its individual components,
the sharply increased trend of KEC above can be taken as a
reflection of the more charge-polarized character of the (excited-
state) encounter complex relative to KEDA of the ground-state
complex. In addition, a solvent of high donicity such as
HMB
CX
C. Formation Enthalpies of CT Complexes. The energet-
ics for the formation of ground-state EDA complexes and
excited-state encounter complexes of quinones and aromatic
donors are evaluated by the enthalpy of formation ∆HEDA and
∆HEC from the temperature dependence of the formation
constants KEDA and KEC, respectively. The value of ∆HEC for
the [CX*, DUR] encounter complex in acetonitrile (as evaluated
in this study) is comparable to that previously reported for the
[CA*, MES] exciplex in carbon tetrachloride.19c Furthermore,
these enthalpies of formation fall in essentially the same energy
range as those of the ground-state EDA complexes of the related
fluoranil acceptor (see FA in Chart 4).47 In all cases, complex
formation is only slightly exothermic, and ∆HEC and ∆HEDA
are comparable to formation enthalpies arising from mostly van
der Waals forces.48 This finding may seem somewhat anoma-
lous if one considers the significantly different charge-transfer
character of the (excited-state) encounter complex relative to
the (ground-state) EDA complex as evaluated by the (b/a)2 ratio
for the degree of charge transfer in the Mulliken formulation
(Vide supra). For example, the latter can be evaluated directly
(41) (a) The broad charge-transfer absorptions in the near IR are
reminiscent of those found in other donor/acceptor pairs with an energy
gap between the oxidation potential of the donor and the reduction potential
of the acceptor close to zero.41b (b) Badger, B.; Brocklehurst, B. Trans.
Faraday Soc. 1969, 65, 2576, 2582, 2588. For charge-resonance absorptions
in mixed-valence metal complexes, see: Curtis, J. C.; Meyer, T. J. Inorg.
Chem. 1982, 21, 1562. Curtis, J. C.; Meyer, T. J. J. Am. Chem. Soc. 1978,
100, 6284.
(42) The data in Table 2 are for the CX/HMB complex; the solvent
polarity is the transition energy of solvatochromic absorption bands. See:
(a) Dimroth, K.; Reichardt, C.; Siepmann, T.; Bohlmann, F. Liebigs Ann.
Chem. 1963, 661, 1. (b) Reichardt, C. SolVents and SolVent Effects in
Organic Chemistry; VCH: New York, 1988.
(43) Foster, R. Organic Charge-Transfer Complexes; Academic Press:
New York, 1969; p 182.
(44) Formation constants obtained in dichloromethane. ∆Ggs represents
the energy gap between the oxidation potential of the donor and the reduction
potential of the acceptor, i.e. ∆Ggs ) Eo°x - Er°ed. ∆Ges represents the
energy gap between excited quinone (ET) and the ion-pair state {A-, D+},
and is thus equivalent to the driving force (∆GET) defined in eq 4. Thus,
the excited and ground state are related by ∆Ges ) ∆Ggs - 2.13 eV.
(45) See: Weller, A. in ref 7, p 23.
(46) The electrostatic work term is taken to be constant.
(47) Foster, R. Organic Charge-Transfer Complexes; Academic Press:
New York, 1969; p 210.
from the enthalpy and charge-transfer data as (b/a)2 = -∆HEDA
/
hνCT (evaluated in Chart 4 and Table 4);49 and the “degree of
Chart 4. Formation Enthalpies of Charge-Transfer Complexes19c,47
Q/ArH
solvent
∆HEC,EDA, kcal mol-1
CX*/DUR
CA*/MES
FA/MES
FA/DUR
FA/HMB
CH3CN
CCl4
CCl4
CCl4
CCl4
-5.6
-7.7
-3.0
-3.9
-5.4
(48) Ben-Naim, A. J. Chem. Phys. 1971, 54, 1387.
(49) (a) Ketelaar, J. A. A. J. Phys. Radium 1954, 15, 197. (b) Rabie, U.
M.; Patel, B. P.; Crabtree R. H.; Mahmoud, M. R. Inorg. Chem. 1997, 36,
2236.

The Encounter Complex in Photoinduced Electron Transfers
J. Am. Chem. Soc., Vol. 119, No. 47, 1997 11477
Chart 5
charge transfer” is found to be 0.3 in the encounter complex,
but only 0.05 in the EDA complex. Thus, the formation
enthalpies are rather insensitive to the charge-transfer character
of the excited encounter complex relative to the ground-state
complex, especially at the limited levels of the degree of charge
transfer shown by quinone/aromatic donor-acceptor pairs.
D. Steric Requirements for Complex Formation. The
structural parallel between the EDA (ground-state) complex and
the encounter (excited-state) complex is further underscored by
the sharp differentiation of electron-rich aromatic donors of
different steric sizes. For example, hexamethylbenzene readily
forms the 1:1 EDA complex with chloranil in Table 4, in which
the charge-transfer absorption derives from the π-π interaction
of the donor-acceptor pair at an intermolecular distance of d
) 3.5 Å in structure A, established by X-ray crystallography.50
acceptor pair moves closer together in the excited state
(compared to d ) 3.5 Å in the ground-state EDA complex) to
form a rather tight complex with a strong overlap of the π -
networks and enhanced degree of charge transfer, as evaluated
by the (b/a)2 values (Vide supra).52
IV. Electron Transfer in the Direct Evolution of the
Encounter Complex to Ion-Radical Pairs. The temporal
evolution of the encounter complex (EC) by electron transfer
directly to the ion-radical pair [Q•-, ArH•+] is established by
the spectral changes in Figure 4. Thus, the first-order decays
of the twin (near-IR and 500-nm) bands of [Q*, ArH] occur
simultaneously with the first-order growth of the diagnostic 430-
nm band of Q•- in acetonitrile, i.e.53
kET
[Q*, ArH] 8 [Q•-, ArH•+]
(6)
both within 10 ns, as described by the common values of the
first-order rate constant kET in Table 2. However, the formation
of the ion radicals as a contact ion pair could not be rigorously
distinguished from a pair of solvated ion radicals, owing to their
rapid dissociation, i.e.
kdiss
[Q•-, ArH•+]
contact
Q , ArH
•-
•+
8
(7)
CH3CN
solvated
ion radicals
The inability of either hexaethylbenzene (HEB) or octameth-
yloctahydroanthracene (OMA) to show any signs of complex
formation (despite oxidation potentials comparable to those of
hexamethylbenzene or durene) can be ascribed to steric interfer-
ence by the pendant methyl groups, as shown by comparison
of the PLUTO structures in Chart 5, that discourage the close
cofacial approach of the quinone acceptor to the benzenoid
chromophore.50
The same steric discrimination is shown in the encounter
(excited-state) complex. Thus, hexamethylbenzene forms the
strongest encounter (excited-state) complexes, as judged by the
intense near-IR bands (Figure 3) and by the large values of KEC
in all solvents (Table 2). By contrast, neither HEB nor OMA
shows any indication of the encounter complex with either
chloranil or dichloroxyloquinone, as deduced by the absence
of near-IR absorption bands and by the remarkably linear kobs
Vs [D] plots, even to high donor concentrations.51 Finally, the
larger formation constants in Table 2 indicate that the donor-
ion pair
which is known to occur with first-order rate constants kdiss
>
109 s-1 in polar solvents such as acetonitrile.54 Although the
spectral indistinguishability of contact and solvated ion-radical
pairs55 precludes the kinetic separation of kET and kdiss, we infer
from the observed time scales that they arise from sequential
(stepwise) and not simultaneous events, i.e. rate-limiting electron
transfer followed by very fast ion separation.
In the less polar solvents (chloroform, dichloromethane, etc.),
the first-order decay of the encounter complex takes up to 160
ns, as shown by the rate constants kET in the range of 107-108
s-1 in Table 2. The roughly 10-fold factor in rate difference
(52) (a) Kim, E. K.; Kochi, J. K. J. Am. Chem. Soc. 1991, 113, 4962.
(b) Friedrich, H. B.; Person, W. B. J. Chem. Phys. 1966, 44, 2161. (c)
Kampar, V. E.; Valtere, S. P.; Neilands, O. Y. Theor. Exp. Chem. 1978,
14, 288.
(53) The stoichiometry of the electron-transfer reaction demands that
ArH•+ is cogenerated with the quinone anion-radical Q•-. However, the
absorption bands of ArH•+ are weak and, in most cases, obscured by the
strong absorption of Q•-. See: refs 28 and 29.
(54) (a) Mattes, S. L.; Farid, S. J. Am. Chem. Soc. 1983, 105, 1396. (b)
Knibbe, H.; Rehm, D.; Weller, A. Ber. Bunsenger. 1968, 72, 257. (c) Ojima,
S.; Miyasaka, H.; Mataga, N. J. Phys. Chem. 1990, 94, 7534. (d) Hubig, S.
M. J. Phys. Chem. 1992, 96, 2903.
(55) Arnold, B. R.; Atherton, S. J.; Farid, S.; Goodman, J. L.; Gould, I.
R. Photochem. Photobiol. 1997, 65, 15. See also: Mataga et al. in ref 54c
and Hubig in ref 54d.
(50) (a) Harding, T. T.; Wallwork, S. C. Acta Crystallogr. 1955, 8, 787.
(b) The optimum intermolecular distance in EDA complexes is the van der
Waals separation of 3.5 Å.38 Thus, the van der Waals radii of HEB and
OMA of r ) 3.2 Å exceed that for sufficient overlap of their π-chro-
mophores with that of the quinone acceptors.
(51) For the effects of steric hindrance on exciplex formation, see:
Jacques, P.; Allonas, X.; Suppan, P.; Von Raumer, M. J. Photochem.
Photobiol. A: Chemistry 1996, 101, 183. See also: Zacchariasse, K. in ref
7, p 286.

11478 J. Am. Chem. Soc., Vol. 119, No. 47, 1997
Chart 6. Driving-Force Dependence of Electron Transfer58
Rathore et al.
Q* + ArH {K } [Q*, ArH] 8 [Q•-, ArH•+]
(10)
kET
EC
Q/ArH
∆GET (kcal mol-1
)
kET (107 s-1
)
CA/XYL
CA/MES
CX/HMB
CX/DUR
-2.1
-0.9
0
8.0
3.6
11
The direct intervention of the encounter complex in the electron
transfer from aromatic donors (ArH) to quinones (Q*) according
to eq 10 is based on three independent lines of experimental
evidence involving (a) the spectroscopic observation of distinc-
tive far red-shifted bands in the near-IR region in Figures 2
and 3, (b) the curved kobs vs [D] plots in Figure 6, and (c) the
lack of temperature dependence of the electron-transfer rate
constants of k2 in Table 3.61
V. Mechanistic Significance of the Encounter Complex
to Bimolecular Electron Transfers. The use of quinone as
the acceptor to examine the dynamics of electron transfer is a
particularly felicitous choice since the preassociation is spectrally
observable both in its excited state Q* (by the near-IR bands in
Figures 2 and 3) and in its ground state Q (by the visible
spectrum in Figure 8). Indeed, the parallel pathways for electron
transfer to Q* and to Q from the graded series of aromatic
donors (D) can be directly compared by emphasizing the overall
thermodynamic relationships among the various species, as
diagrammatically presented in Scheme 1. [Note that the relative
energy levels are somewhat arbitrary, and the spin multiplicities
are not included.]
4.8
2.7
between acetonitrile and chloroform is consistent with the
increased stabilization of ion radicals in the more polar medium.
Even more striking is the spectral evolution of the encounter
complex to yield the protonated quinone anion radical (Q-H•),
as the electron-transfer product first observed in chloroform (see
Figure 5). In such a medium of low polarity, we expect ion
dissociation (kdiss) in eq 7 to be substantially slower than that
in acetonitrile, which allows the fast proton transfer (kH) within
the ion-radical pair to compete effectively, i.e.56
kH
CHCl 8
Q-H•, Ar•
radical pair
[Q•-, ArH•+]
contact
(8)
3
ion pair
Such a stepwise process from the contact ion pair is confirmed
by the pronounced salt effect induced by the added inert salt
-
TBA+ PF6 which effects ion-radical (pair) separation,57 i.e.
Scheme 1
[Q•-, ArH•+] + TBA+ PF6- a
-
Q•-, TBA+ + ArH•+, PF6 (9)
The effectiveness of the ion-pair exchange in eq 9 is demon-
strated by the spectral decay of the transient intermediate (in
chloroform containing added salt) to afford the quinone anion
•
•-
radical (ΦQ ) 0.7) and not Q-H , as observed in the absence
of salt (Figure 5). The exchange in eq 9 efficiently separates
the ion radicals and thus effectively precludes the proton transfer
within the contact ion pair according to eq 8. Importantly, the
kinetics study in Figure 7 demonstrates that the salt effect is
limited to the behavior of the contact ion pair following the
rate-limiting electron transfer, since all the kinetic parameters
associated with the encounter complex (k2, KEC, and kET) are
singularly unaffected by the added salt in Table 2.
The electron-transfer rates within the encounter complex in
eq 6 are experimentally represented by the value of kET in Chart
6 measured at different driving forces ∆GET, as defined in eq
4.58 Although we note a slight increase in kET with increasing
exergonicity of the electron transfer from the aromatic donor
to Q*, the magnitude of the trend is clearly insufficient to
quantitate. Indeed, such a mild ∆GET dependence is typically
observed for back-electron-transfer rates upon the photoexci-
tation of charge-transfer complexes.8,59,60
The electron-transfer mechanism as depicted in the lower
(ground-state) manifold in Scheme 1 proceeds via the preequi-
librium formation of the electron donor-acceptor complex
[D,Q]EDA which has been elaborated in earlier reports.62 In this
study, laser photoactivation spontaneously generates Q*, which
resides at an energy level of ET ) 50 kcal mol-1 above that of
Q, and this necessitates the use of fast (time-resolved) spec-
troscopy to directly measure its fast temporal evolution via the
encounter complex [D,Q*]EC, as presented in the upper (excited-
state) manifold in Scheme 1. Otherwise, Q and Q* follow the
same overall electron-transfer (kinetics) behavior; and the basic
difference is in the energeticssthe encounter complex lying at
substantially higher energy levels than the EDA complex. As
a result, the driving force for electron transfer from [D,Q*]EC
is highly exergonic under conditions in which that from
[D,Q]EDA is endergonic.
Time-resolved spectroscopy thus demonstrates that the en-
counter complex (EC) plays a pivotal role in effecting the
electron transfer from aromatic donors to Q*, i.e.
The encounter complex and the EDA complex both represent
the preorganization of the reactants prior to electron transfer,63
as underscored by the intimate cofacial juxtaposition of the
donor and acceptor for optimal overlap of their π-orbitals, as
seen in the side and top perspectives AS and AT below.50
In such complexes, the electronic coupling of the donor/acceptor
moieties is given roughly by the degree of charge-transfer (b/
a)2, which reflects the contribution from the ionic or [D•+, Q•-]
form.37 Thus, the reasonable values of (b/a)2 = 0.3 in [D,Q*]EC
(56) Fast proton transfer from polymethylbenzene cation-radicals is
known to occur with rate constants of about 108 s-1. See: (a) Masnovi, J.
M.; Sankararaman, S.; Kochi, J. K. J. Am. Chem. Soc. 1989, 111, 2263. (b)
Schlesener, C. J.; Amatore, C.; Kochi, J. K. J. Am. Chem. Soc.1984, 106,
7472. (c) See also: Jones et al. in ref. 18.
(57) Yabe, T.; Kochi, J. K. J. Am. Chem. Soc. 1992, 114, 4491. See
also: Bockman, T. M.; Kochi, J. K. in ref 17c.
(58) In the gas phase, the driving force to move an electron from the
HOMO of the donor to the LUMO of the acceptor is equal to the difference
of the ionization potential and the electron affinity of the donor and the
acceptor, respectively. In solution, solvation terms and coulombic work terms
must also be considered.22.23
(59) Asahi, T.; Ohkohchi, M.; Mataga, N. J. Phys. Chem. 1993, 97,
13132.
(61) For the mechanistic deductions of this unusual temperature effect,
see: Baggott in ref 13c.
(62) (a) For a review, see: Kochi, J. K. Acta Chem. Scand. 1990, 44,
409. (b) We take the electronic coupling of the donor and acceptor in the
encounter complex to be symptomatic of an inner-sphere complex, since
such an interaction is minimal in an outer-sphere complex.3 The charac-
teristics of inner-sphere Vs outer-sphere behavior in bimolecular electron-
transfer reactions will be discussed in a subsequent paper.35
(60) Hubig, S. M.; Bockman, T. M.; Kochi, J. K. J. Am. Chem. Soc.
1996, 118, 3842.

The Encounter Complex in Photoinduced Electron Transfers
J. Am. Chem. Soc., Vol. 119, No. 47, 1997 11479
distillation. Tetrachloro-p-benzoquinone (chloranil, Aldrich) was sub-
limed in Vacuo and recrystallized from benzene.
Synthesis of 2,5-Dichloro-3,6-dimethyl-1,4-benzoquinone (CX).
Chlorination of 2,5-dimethyl-1,4-dimethoxybenzene64 with sulfuryl
chloride in the presence of a catalytic amount of iodine in dichlo-
romethane afforded 2,5-dichloro-3,6-dimethylhydroquinone dimethyl
ether quantitatively, which was readily oxidized to the corresponding
quinone with excess nitric acid (63%) in an excellent yield of 88%.
Mp 55 °C; 1H NMR δ 2.24 (s, 6H); 13C NMR (CDCl3) 14.45, 140.95,
142.16, 177.45. GC-MS m/z 205 (M+), calcd for C8H6O2Cl2. Anal.
Calcd for C8H6O2Cl2: C, 46.85; H, 2.95; Cl, 34.58. Found: C, 46.67;
H, 2.90; Cl, 34.43. Dichloromethane, acetonitrile, chloroform, and
carbon tetrachloride were purified according to standard procedures.65
point to an encounter complex that is already predisposed
significantly toward electron transfer.63b As such, the charge-
transfer character of the encounter complex provides the
rationale for its intermediacy in electron transfer,62b especially
since the prior association step occurs at a diffusion-controlled
rate.
Instrumentation. The UV-vis absorption spectra were recorded
on a Hewlett-Packard 8453 diode-array spectrometer. For the nano-
second-laser experiments, the third-harmonic (355 nm) output of a
Q-switched Nd:YAG laser (10-ns fwhm, 22 mJ)66 was used to generate
the triplet quinones. The picosecond time-resolved pump-probe
experiments were carried out with the third-harmonic (355 nm) output
of a mode-locked Nd:YAG laser (25 ps fwhm, 10 mJ).66 The one-
electron oxidation potentials of the arene donor were determined by
cyclic voltammetry with use of a BAS 100A electrochemical analyzer.
All potentials in Chart 1 were measured in dichloromethane containing
10% trifluoroacetic acid and 0.1 M tetra-n-butylammonium hexafluo-
rophosphate.22 The standard calomel electrode was used as reference.
Summary and Conclusions
Time-resolved (ps f µs) spectroscopy reveals three distinct
and sequential events (together with the attendant intermediates)
in the bimolecular electron transfer from aromatic donors (D)
to photoactivated quinones (Q*). Initially, the redox-active
acceptor Q* is formed within the 25-ps laser pulse with unit
quantum yield. Subsequently, on the early nanosecond time
scale, the diffusion-controlled association of Q* with the
Determination of the Formation Constants of the EDA Com-
plexes. A 0.001 M solution of quinone was placed in a 1-cm quartz
cuvette equipped with a side arm and a Schlenk adapter under an argon
atmosphere. Known amounts of arene donor were added in small
increments, and the growth of the charge-transfer absorption band was
monitored at several wavelengths. The absorbance data were then
evaluated with the aid of the Benesi-Hildebrand correlation,36 i.e.
aromatic donors leads to the encounter complex [D,Q*]EC
.
Finally, on the ns/µs time scale, the internal electron transfer
(kET) within the encounter complex results in the formation of
the ion-radical pair [D•+, Q•-] with unit efficiency in acetonitrile.
Thus, the encounter complex and the ion-radical pair are neither
the same nor alternative (solvent-dependent) intermediates in
bimolecular electron transfer; but their formation and decay can
be individually monitored on distinct time scales in polar
solvents. Moreover, the observation of ion radicals (Q•-) even
in solvents of low polarity (upon the addition of salt) confirms
the generality of this sequence of ET intermediatessindependent
of the polarity of the reaction medium.
The encounter complex is characterized by strong charge-
transfer interactions between the donor/acceptor moieties that
are revealed spectroscopically by broad absorptions in the near-
IR region and thermodynamically by large formation constants
KEC as compared to those of (ground-state) EDA complexes.
Apart from differences in energy content, however, photoacti-
vated encounter complexes and EDA complexes exhibit quite
similar structural and charge-transfer properties, and they play
analogous roles as preassociation intermediates in photoactivated
and ground-state electron-transfer processes, respectively, as
pictorially depicted in Scheme 1. Both complexes thus represent
experimentally observable examples of precursor complexes of
the type theoretically postulated for bimolecular (adiabatic)
electron transfer.2,3
-1
[Q]/ACT ) ꢀCT + (KEDAꢀCT[D])-1, where [Q] was the quinone
concentration, ACT was the charge-transfer absorbance (monitored at a
wavelength (λCT) within the CT absorption band), ꢀCT was the extinction
coefficient of the EDA complex at the monitoring wavelength λCT, KEDA
was the formation constant of the EDA complex, and [D] was the donor
concentration. Thus, the linear plot of [Q]/ACT Versus the reciprocal
-1
donor concentration exhibited a slope of (KEDAꢀCT
)
and an intercept
-1
of ꢀCT from which the values of KEDA and KEDAꢀCT in Table 4 were
extracted with correlation coefficients R > 0.999. Each Benesi-
Hildebrand experiment was carried out twice to ensure reproducibility
within the least-squares error bars.
Determination of the Maximum Extinction Coefficient and the
Quantum Yield of Triplet 2,5-Dichloroxyloquinone. The absorption
spectrum of the triplet state of 2,5-dichloroxyloquinone (CX) exhibits
a maximum at 500 nm (see Figure 1B). To determine the maximum
triplet extinction coefficient (ꢀT, 500), we carried out an energy-transfer
experiment to generate triplet CX from triplet benzophenone67 as
follows: Benzophenone (13.8 mg, 0.015 M) and CX (1.1 mg, 0.001
M) were dissolved together in acetonitrile (5 mL) under an argon
atmosphere. Upon laser excitation (Nd:YAG, 10 ns, 355 nm), the
transient absorption spectra recorded between 0 and 400 ns showed
the decay of triplet benzophenone in the wavelength region between
500 and 600 nm and the growth of triplet CX in the region between
400 and 500 nm. The transient absorbance monitored at 500 nm did
not change over the same time period. This isosbestic point at 500
nm indicated that triplet benzophenone and triplet CX exhibit the same
extinction coefficient at this wavelength (ꢀT, 500). We determined the
extinction coefficient of triplet benzophenone in acetonitrile at 500 nm
to be 5300 M-1 cm-1 by quantitative comparison of triplet benzophe-
none spectra in acetonitrile and benzene.25 On the basis of the isosbestic
point at 500 nm in the benzophenone/CX transient spectra, we conclude
that the maximum extinction coefficient of 2,5-dichloroxyloquinone
in acetonitrile at 500 nm amounts to ꢀT, 500 ) 5300 M-1 cm-1. The
Experimental Section
Materials. Hexaethylbenzene (Acros), hexamethylbenzene, and
durene (Aldrich) were recrystallized from ethanol and heptane. Mesi-
tylene (Aldrich) and p-xylene (Aldrich) were purified by fractional
(63) (a) It has been long recognized in enzymology that an analogous
preorganization in the form of reversible (enzyme/substrate) complexation
greatly facilitates processes that are otherwise unfavorable (and slow), and
it is thus not unexpected that eq 2 has essentially the same form as the
Michaelis-Menten equation (see: Stryer, L. Biochemistry, Freeman: San
Francisco, 1985; pp 103-185). (b) The enhanced degree of charge transfer
in [D,Q*]EC in lowering the energy of the transition state for the exergonic
electron transfer in the excited-state manifold (relative to that in [D,Q]EDA
for the ground-state manifold compared in Scheme 1) accords with the
Hammond postulate [Hammond, G. S. J. Am. Chem. Soc. 1995, 77, 334].
(64) Rathore, R.; Bosch, E.; Kochi, J. K. Tetrahedron 1994, 50, 6727.
(65) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals, 3rd ed.; Pergamon: Oxford, 1988.
(66) Bockman, T. M.; Karpinski, Z. J.; Sankararaman, S.; Kochi, J. K.
J. Am. Chem. Soc. 1992, 114, 1970.
(67) Bensasson, R.; Land, E. J. Trans. Faraday Soc. 1971, 67, 1904.

11480 J. Am. Chem. Soc., Vol. 119, No. 47, 1997
Rathore et al.
triplet quantum yield of 2,5-dichloroxyloquinone was determined by
transient actinometry.25 Thus, a solution of benzophenone in benzene
and a solution of dichloroxyloquinone in acetonitrile, both with matched
absorbance at 355 nm, were excited with the Q-switched Nd:YAG laser
(10 ns, 355 nm), and the transient absorbances extrapolated to time t
) 0 were measured at 525 and 500 nm, respectively. The triplet
quantum yield of dichloroxyloquinone ΦT,CX was then determined
Determination of Free-Ion Yields. The free-ion yields in aceto-
nitrile and chloroform (with added salt) were determined by the same
method as described above in detail for the determination of the triplet
quantum yield. Thus, benzophenone in benzene was used as the
transient actinometer25 and samples of the quinone/donor combinations
and of benzophenone with matching absorbances at 355 nm were
photoexcited with the 10-ns Nd:YAG laser. The free-ion yields were
according to the following: ΦT,CX ) (A500,CX/A525,BP)(ꢀ525,BP/ꢀ500,CX)ΦT,BP
,
then determined by quantitative comparison of the maximum absorbance
25
where A500,CX and A525,BP were the transient absorbances of CX and
benzophenone at 500 and 525 nm, respectively, ꢀ525,BP and ꢀ500,CX were
the triplet extinction coefficients of benzophenone in benzene and CX
in acetonitrile at 525 and 500 nm, respectively, and ΦT,BP was the triplet
quantum yield of benzophenone in benzene. Taking ꢀ525,BP ) 7220
of triplet benzophenone at 525 nm (ꢀ ) 7220 M-1 cm-1
)
and the
maximum absorbance of chloranil anion radical at 450 nm (ꢀ ) 9700
M-1 cm-1
)
or CX anion radical at 430 nm (ꢀ ) 6800 M-1 cm-1).28
30
Acknowledgment. We thank T. M. Bockman for many
helpful discussions and the National Science Foundation and
the R. A. Welch Foundation for financial support.
M-1 cm-1 25 ꢀ500,CX ) 5300 M-1 cm-1, and ΦT,BP ) 1.0,68 we obtained
,
a triplet quantum yield of ΦT,CX ) 1.0 ( 0.05.
(68) Moore, W. M.; Hammond, G. S.; Foss, R. P. J. Am. Chem. Soc.
1961, 83, 2789.
JA971188Y