Laser photolysis investigation of induced quenching in photoreduction of benzophenone by alkylbenzenes and anisoles

Article abstract of DOI:10.1039/a707900a

The quenching processes of triplet benzophenone (JBP) by alkylbenzenes (AB) and anisole derivatives (AD) in benzene (Bz) and a mixture of acetonitriie (ACT-,) and water (4 :1 v/V; have been studied on the basis of rate constants and efficiencies determined by nanosecond laser flash photolysis a; 355 n m at 295 K. It was found that (1) the deactivation of 3BPby ADs in ACN H2O (4 :1 v/v) was governed by electron transfer (ET) to produce the benzophenone anion (BP'~) and corresponding cation (AD' + ) radicals wiih efficiencies, atj < 1 whereas no chemical species were formed in Bz; and 2) photoreduction of 3BPby ABs resulted in benzophenone ketyl radical (BPK) formation by benzylic hydrogen abstraction (HA) with efficiencies XHA < 1 in 3z and ACN-H2O (4 :1 v/v). The residual efficiency (a: 1 -ET or ! -aH/1) was attributed to a birnolecular process with no photochemical product, which was named 'induced-quenching (IQf. The quenching rate constants (Jcq) of ;'BPby ADs and ABs were less than the diffusion limits of both Bz and AC1~H2O (4 :1 v/v). The net bimolecular rate constants for the ET, HA and IQ processes were estimated from the k values and efficiencies. The rate constants (%T and k,Q) of ET and IQ with AD versus the oxidation potential (￡) of AD followed Rchm-Weller behaviour while logarithmic rate constants {/CHA and ki(j) of HA and IQ by ABs increased linearly with a decrease in the Em of AB. It was suggested, for the deactivation mechanism of 3BPby ABs and ADs (RH), that ;1) the IQ process was intersystem crossing (ISC) enhanced by the partial charge transfer (CT) character of the triplet excipiexes, 3(BP"~- A-RHa + )a,e; (2) radical ion formation by ET might be accomplished in a polar solvent by further CT interaction in the excipiex; (3) the process of BPK formation was inferred to be H-atom transfer in the exciplex, where the more protic H-atom was readily mobile, rather than ET followed by proton transfer and (4) the loss of efficiencies of photochemicalproduct formation was derived not from back ET but from the IQ process, inherent to photoreactions, via triplet excipiexes. The deactivation processes of 3BPby RH are illustrated in Scheme 1. I ET BP'- + RH'(3BP' + RHJcoj -3(BPO- RHg,.-BPK 4 R' BP + RH Scheme 1.

Full text of DOI:10.1039/a707900a

View Online / Journal Homepage / Table of Contents for this issue
Laser photolysis investigation of induced quenching in photoreduction
of benzophenone by alkylbenzenes and anisoles¤
Katsuji Okada, Minoru Yamaji and Haruo Shizuka*
Department of Chemistry, Gunma University Kiryu, Gunma 376 8515, Japan
The quenching processes of triplet benzophenone (3BP*) by alkylbenzenes (AB) and anisole derivatives (AD) in benzene (Bz) and
a mixture of acetonitrile (ACN) and water (4 : 1 v/v) have been studied on the basis of rate constants and efficiencies determined
by nanosecond laser Ñash photolysis at 355 nm at 295 K. It was found that (1) the deactivation of 3BP* by ADs in ACNÈH O
2
(4 : 1 v/v) was governed by electron transfer (ET) to produce the benzophenone anion (BP~~) and corresponding cation (AD~`)
radicals with efficiencies, a \ 1 whereas no chemical species were formed in Bz; and (2) photoreduction of 3BP* by ABs resulted
ET
in benzophenone ketyl radical (BPK) formation by benzylic hydrogen abstraction (HA) with efficiencies a \ 1 in Bz and
HA
ACNÈH O (4 : 1 v/v). The residual efficiency (a : 1 [ a or 1 [ a ) was attributed to a bimolecular process with no photo-
2
IQ
ET
HA
chemical product, which was named “induced-quenching (IQ)Ï. The quenching rate constants (k ) of 3BP* by ADs and ABs were
q
less than the di†usion limits of both Bz and ACNÈH O (4 : 1 v/v). The net bimolecular rate constants for the ET, HA and IQ
2
processes were estimated from the k values and efficiencies. The rate constants (k and k ) of ET and IQ with AD versus the
q
ET
IQ
oxidation potential (E ) of AD followed RehmÈWeller behaviour while logarithmic rate constants (k and k ) of HA and IQ by
ABs increased linearly with a decrease in the E of AB. It was suggested, for the deactivation mechanism of 3BP* by ABs and
ADs (RH), that (1) the IQ process was intersystem crossing (ISC) enhanced by the partial chargeÈtransfer (CT) character of the
ox
HA
IQ
ox
triplet exciplexes, 3(BPd~É É ÉRHd`)* ; (2) radical ion formation by ET might be accomplished in a polar solvent by further CT
cage
interaction in the exciplex; (3) the process of BPK formation was inferred to be H-atom transfer in the exciplex, where the more
protic H-atom was readily mobile, rather than ET followed by proton transfer and (4) the loss of efficiencies of photochemical-
product formation was derived not from back ET but from the IQ process, inherent to photoreactions, via triplet exciplexes. The
deactivation processes of 3BP* by RH are illustrated in Scheme 1.
Scheme 1
Photoreduction of triplet ketones by alkanes, alcohols,
amines, etc. has been studied experimentally and theoretically
since it is preceded by elemental chemical reactions, H-atom
and electron transfer (HT and ET).1h14 The mechanism of the
primary photoreduction of benzophenone (BP) has been
extensively studied by means of product analysis and laser
photolysis techniques.15h45 The development of pico-, and
femto-second laser instruments has enabled it to be
shown36h42 that photoreduction of triplet BP (3BP*) by
amines (AH) proceeds via ET followed by proton transfer (PT)
in the ion pair or exciplex with strong CT character, resulting
in the production of the benzophenone ketyl radical (BPK)
(the ETPT mechanism):2,11,17,33b,36h42
unity.34,39 The loss in product yield was considered to be due
to back ET in the ion pair or CT complex prior to PT leading
to BPK formation.11,17,37,39 In the formation of BPK from
3BP* by amines, a strong CT exciplex or ion pair plays an
important role in the primary step of photoreduction.
When the quenching of the excited molecules is governed
by CT interaction, the quenching rate constant depends on
the redox potentials of the reactants.46 Plots of the quenching
rate constant of 3BP* versus the ionization (or oxidation)
potentials of quenchers, not only amines but also alkyl-
benzenes, have well documented CT interaction.43,44 Wagner
et al. showed by the product-analysis method that triplet ace-
tophenone derivatives were reduced in a CT process, even by
alkylbenzenes.28 Such processes may involve triplet exciplexes
with weak CT interaction, while no formation of the BP anion
radical nor CT exciplex was reported for reduction of 3BP* by
alcohols.23 There have been many reports of rate constants
for HA of 3BP* by alkylbenzenes to produce BPK.1a,4,28b
However, the rate constant of BPK formation has been
regarded to be equal to that of the quenching process. In
other words, the efficiency of BPK formation has been
assumed to be unity. In order to understand the HA mecha-
nism by alkylbenzenes where exciplex formation may be
ET
PT
3BP* ] AH ÈÈÈ 3(BP~É É ÉAH`) ÈÈÈ BPK ] A~
where A~ represents an amine radical. The quantum yields of
BPK formation by some amines were found to be less than
¤ A preliminary discussion of their work was given at the 54th
Okazaki Conference: Dynamic Studies on Hydrogen Atom Transfer
Reactions, January, 1996, Okazaki, Japan and it was presented at the
XVIth IUPAC Symposium on Photochemistry, July, 1996, Helsinki,
Finland.
J. Chem. Soc., Faraday T rans., 1998, 94(7), 861È866
861

View Online
involved, the efficiency and net rate constant for BPK forma-
tion should be determined precisely.
In our previous work on photoreduction of BP by toluene
3BP*.1a,4,28b Fig. 1(a) shows the time-resolved transient
absorption spectra obtained by laser Ñash photolysis at 355
nm in the BPÈTol (2.8 mol dm~3) system in Bz. The absorp-
tion band at 530 nm observed at 300 ns is ascribed to the
tripletÈtriplet (TÈT) absorption of BP in a non-polar
solvent.52 With time, the TÈT absorption of BP decreases in
intensity with a Ðrst-order decay rate (k ) to give the
derivatives substituted with halogen atom(s) (XPhCH ), the
3
deactivation pathways of 3BP* were investigated in terms of
heavy atom e†ects by nanosecond laser Ñash photolysis tech-
niques.45 It was found that the efficiency of BPK formation
from 3BP* was less than unity. The residual efficiency implied
another deactivation pathway of 3BP* as well as HA; this was
called “induced quenching (IQ)Ï.
obsd
absorption spectrum of the BPK radical having a peak at 545
nm53 at 4.0 ls. Thus, Fig. 1(a) demonstrates that the deactiva-
tion of 3BP* in the presence of Tol involves HA from Tol to
give BPK and the benzyl radical which has no absorption in
the wavelength range used 380È650 nm.54 The experimentally
obtained k
value increased linearly with increase in [Tol].
value may be expressed by:
obsd
Thus, the k
obsd
k
\ k ] k [Tol]
(1)
obsd
0
q
where k and k represent the rate constants for the quenching
Scheme 2
q
0
of 3BP* by Tol and the decay of 3BP* in the absence of Tol,
where XPhCH ~ stands for the corresponding benzyl-type
2
respectively. From eqn. (1), k (3.1 ] 105 dm3 mol~1 s~1) and
radicals. In the XPhCH case, the rate constant of HA of
q
3
k
(3.3 ] 105 s~1) values in the BP-Tol system in Bz were
3BP* from XPhCH was almost independent of the spinÈorbit
0
3
obtained.
coupling constants (m) of XPhCH , whereas those of IQ
3
The efficiency (a ) for HA of 3BP* from Tol was deter-
mined on the basis of the quantum yield (U ) for the forma-
value was determined from the net
absorbance (*A ) at 545 nm due to formation of BPK after
increased with an increase in the m2 value. These results
HA
implied that IQ would have an intersystem crossing
pathway.45 However, the total mechanism of deactivation,
including the IQ process, has not yet been made clear.
BPK
tion of BPK. The U
BPK
545
depletion of 3BP* [see e.g. Fig. 1(b)] by:
In the present study, nanosecond laser photolysis was used
to investigate the photoreduction of BP by alkylbenzenes with
various oxidation potentials. For comparison, anisole deriv-
atives were employed, as they had no benzylic H-atom to be
abstracted and lower oxidation potentials than alkylbenzenes.
The deactivation mechanism of 3BP* by benzene derivatives
was discussed in terms of triplet exciplexes with CT character
originating ET, HA and IQ processes.
U
\ *A eBPK~1I~1 N
545 545 abs
(2)
BPK
A
where eBPK , I and N are the molar absorption coefficient of
545 abs
A
BPK at 545 nm (3700 dm3 mol~1 cm~1),53 the absorbed
photon Ñux of an incident laser pulse and AvogadroÏs number,
respectively. I was evaluated by using the TÈT absorption
of BP in ACN as an actinometer. After laser pulsing in an
ACN solution of BP in a 10 mm path length cell, the initial
abs
absorbance change (*A ) due to 3BP* at 520 nm is related
Experimental
520
to I by:
abs
BP was puriÐed by recrystallization from ethanol. Anisole, 4-
methoxytoluene (4MeOTol), mesitylene (135MeB), 1,2,4-tri-
methoxybenzene (124MeOB), p-xylene (14MeB), m-xylene
(13MeB), 1,2,4-trimethylbenzene (124MeB) and 1,2,3,4-tetra-
methybenzene (1234MeB) were distilled in vacuo. 1,3,5-Tri-
*A \ UBP eBP I N~1
520 ISC 520 abs
(3)
A
where UBP and eBP represent the triplet yield of BP (1.0)55
ISC 520
and the molar absorption coefficient of 3BP* in ACN at 520
nm (6500 dm3 mol~1 cm~1),56 respectively. The plots of U
,
BPK
methoxybenzene
(135MeOB),
1,2,3-trimethoxybenzene
obtained experimentally in the BPÈTol system, in Bz, as a
(123MeOB), 1,4-dimethoxybenzene (14MeOB), pentamethyl-
benzene (PMeB), hexamethylbenzene (HMeB) and durene
(1245MeB) were puriÐed by vacuum sublimation. These
materials were commercially available. Toluene (Tol, InÐnity
Pure from Wako) was used as supplied. 1,2,4,5-Tetramethoxy-
benzene (1245MeOB) was synthesized according to the liter-
ature.47
function of [Tol] are shown in Fig. 2.
With an increase in [Tol], the U
BPK
showing a negative curve. On the other hand, the U
value increases,
value
BPK
The oxidation potentials (E ) of AB and AD were obtained
ox
from the literatures48,49 or determined from the CT absorp-
tion band with tetracyanoethylene in acetonitrile (ACN) at
295 K.50 Benzene (Bz, Dojin Spectrosol) was used as supplied.
ACN and deionized water were distilled. Bz and a mixture of
ACN and H O (4 : 1 v/v) were used as solvents.
2
The concentration of BP was 6.7 ] 10~3 mol dm~3
throughout this work. All samples, in a quartz cell of a 10 mm
path length, were thoroughly degassed by freezeÈpumpÈthaw
cycles on a high vacuum line.
Laser Ñash photolysis was carried out at 295 K by using the
third harmonics (355 nm) of an Nd3` : YAG laser from J. K.
Lasers (HY-500). The pulse duration, energy and diameter of
the laser beam were 8 ns, 40 mJ pulse~1, and 12 mm, respec-
tively. The detection system for transient absorption spectra
has been reported elsewhere.51
Results
Fig. 1 (a) Transient absorption spectra at 300 ns (L) and 4.0 ls (…)
obtained by 355 nm laser Ñash photolysis in the BPÈTol (2.8 mol
dm~3) system in Bz. (b) The time proÐle of the transient absorbance
at 545 nm for the corresponding system.
Quenching of 3BP* by ABs
It is well known that toluene behaves as an H-donor for
862
J. Chem. Soc., Faraday T rans., 1998, V ol. 94

View Online
Fig. 2
U
vs. [Tol] for the BPÈTol system in Bz. The solid curve
BPK
was calculated by eqn. (4). See text for details.
can be formulated by:
Fig. 3 (a) Transient absorption spectra observed at (1) 100, (2) 180,
k a [Tol]
(3) 300 and (4) 800 ns after 355 nm laser pulsing for BPÈ1245MeOB
q HA
U
\
UBP
(4)
(6.0 ] 10~4 mol dm~3) in ACNÈH O (4 : 1 v/v). (b) Reference absorp-
BPK
ISC
k ] k [Tol]
2
0
q
tion spectra for BP~~ (ÈÈ) and 1245MeOB~` (È È È).
Using a best-Ðt technique, the a
value was found to be
HA
0.77 ^ 0.04. The solid curve in Fig. 2 was simulated with the
best-Ðt parameters. The residual efficiency (a \ 1 [ a ) was
3BP* by AD. Fig. 3(a) shows the time-resolved transient
absorption spectra obtained by laser Ñash photolysis at 355
nm in the BPÈ1245MeOB (6.0 ] 10~4 mol dm~3) system in
IQ
HA
considered to be attributed to IQ of 3BP* by Tol, since no
reaction except for HA occurred during the decay of 3BP*, as
seen from the transient absorption spectral changes. Hence,
ACNÈH O (4 : 1 v/v). The transient absorption spectrum at
2
520 nm observed at 100 ns may be ascribed to 3BP* in a polar
the a value was determined to be 0.23 ^ 0.04 for the BPÈTol
IQ
solvent.52 With time, the TÈT absorption of BP decreases in
intensity according to Ðrst-order kinetics to give residual
absorption bands with peaks at 630 and 450 nm at 800 ns
with isosbestic points at 555 and 465 nm. This residual
absorption spectrum was reproduced by superposing those of
the benzophenone anion (BP~~)57 and 1245MeOB cation
(1245MeB~`)58 radicals [see Fig. 3(b)] in a 1 : 1 concentration
ratio. Hence, the spectral changes in Fig. 3(a) demonstrate ET
from 1245MeOB to 3BP* to produce BP~~ and 1245MeOB~`.
system in Bz.
The rate constants (k and k ) for HA and IQ are given
by:
HA
IQ
k
\ a
\ a
k
(5)
(6)
HA
HA q
k
k
IQ
IQ q
where k \ k ] k . The k and k values were 2.4 ] 105
q
HA
IQ
HA
IQ
and 7.1 ] 104 dm3 mol~1 s~1, respectively, for the BPÈTol
The quantum yield (U ) for formation of BP~~ was
system in Bz.
ani
obtained from:
Since formation of BPK was observed in other BPÈAB
systems in Bz as well as in ACNÈH O (4 : 1 v/v), the AB used
seemed to behave as an H-atom donor. The efficiencies and
rate constants for other BPÈAB systems were determined by
the same procedures as for Tol, and are listed in Table 1.
U
\ *A eani~1I~1
(7)
2
ani
630 630 abs
where *A
and eani represent the net absorbance at 630 nm
630
630
due to formation of BP~~ and the molar absorption coefficient
of BP~~ at 630 nm (5700 dm3 mol~1 cm~1)” respectively.°
Quenching of 3BP* by anisole derivatives
On the other hand, the U value may be calculated from
ani
eqn. (8) with the use of the efficiency (a ) of ET between 3BP*
Upon 355 nm laser excitation in the BPÈAD systems in Bz,
the TÈT absorption of BP decreased according to Ðrst-order
kinetics, giving no residual absorption spectrum. Hence, the
deactivation processes of 3BP* by ADs in Bz were due com-
pletely to IQ, i.e. a \ 1.0. The k [\k , see eqn. (6)] values
ET
and 1245MeOB.
” By taking the e value of BPK as 3700 dm3 mol~1 cm~1 at 545
nm,53 e for BP~~ was re-estimated to be 5700 dm3 mol~1 cm~1 at 630
nm by the method given in ref. 57(a).
IQ
q
IQ
were obtained by the same procedure as for the BPÈAB
° The AD cation radicals, except for 135MeOB~~, produced in this
system.
work had no absorption at 630 nm. The e135 value of 135MeOB~` at
630
630 nm was 500 dm3 mol~1 cm~1 in ACNÈH O (4 : 1 v/v). In analysis
2
On changing the solvent from Bz to ACNÈH O (4 : 1 v/v), a
2
surprising change was observed in the quenching proÐle of
for the 135MeOB case, eani in eqn. (7) was replaced by eani ] e135 .
630
630
630
Table 1
E
of AB and efficiencies and rate constants obtained for the BPÈAB system in Bz and ACNÈH O (4 : 1 v/v)a
ox
2
Bz
ACNÈH O (4 : 1 v/v)
2
AB
E b/V
k c
a
d
a
e
k
f
k
g
k c
a
d
a
e
k
f
k g
ox
q
HA
IO
HA
IO
q
HA
IO
HA
IO
Tol
13MeB
14MeB
135MeB
124MeB
1245MeB
1234MeB
PMeB
HMeB
4MeOTolj
2.25
2.11
2.06i
2.02
1.89i
1.79
1.76
1.71
1.58
1.58k
3.1 ] 105
1.5 ] 106
1.7 ] 106
9.9 ] 106
7.2 ] 106
3.2 ] 107
3.1 ] 107
1.8 ] 108
1.1 ] 109
2.4 ] 107
0.77
0.50
0.73
0.21
0.44
0.36
0.31
0.16
0.08
0.17
0.23
0.50
0.27
0.79
0.56
0.64
0.69
0.84
0.92
0.83
2.4 ] 105
7.5 ] 105
1.2 ] 106
2.1 ] 106
3.2 ] 106
1.2 ] 107
9.6 ] 106
2.9 ] 107
8.8 ] 107
4.1 ] 106
7.1 ] 104
7.5 ] 105
4.6 ] 105
7.8 ] 106
4.0 ] 106
2.0 ] 107
2.1 ] 107
1.5 ] 108
1.0 ] 109
2.0 ] 107
n.d.h
n.d.h
n.d.h
4.7 ] 107
3.5 ] 107
1.7 ] 108
1.7 ] 108
8.7 ] 108
4.2 ] 109
1.9 ] 108
n.d.h
n.d.h
n.d.h
0.11
0.25
0.27
0.21
0.25
0.24
0.26
n.d.h
n.d.h
n.d.h
0.89
0.75
0.73
0.79
0.75
0.76
0.58
n.d.h
n.d.h
n.d.h
5.2 ] 106
8.8 ] 106
4.6 ] 107
3.6 ] 107
2.2 ] 108
1.0 ] 109
4.9 ] 107
n.d.h
n.d.h
n.d.h
4.2 ] 107
2.6 ] 107
1.2 ] 108
1.3 ] 108
6.5 ] 108
3.2 ] 109
1.1 ] 108
a Errors within 5% for rate constants and efficiencies. b vs. SCE in ACN. Data from ref. 48. c Quenching rate constant of 3BP* by AB in dm3
mol~1 s~1. d Efficiencies for BPK formation. e Efficiencies for IQ of 3BP* by AB. f Rate constants for HA of 3BP* from AB in dm3 mol~1 s~1.
g Rate constants for IQ of 3BP* by AB in dm3 mol~1 s~1. h Not determined due to low solubility. i vs. SCE in triÑuoroacetic acid. j See text for
details. k Obtained by the CT method. See text.
J. Chem. Soc., Faraday T rans., 1998, V ol. 94
863

View Online
a
k [1245MeOB]
ET q
U
\
UBP
(8)
ani
ISC
k ] k [1245MeOB]
0
q
Using a best-Ðt technique for the experimental U values and
ani
the k (7.6 ] 109 dm3 mol~1 s~1) and k (1.5 ] 105 s~1)
q
0
values for the BPÈ1245MeOB system in ACNÈH O (4 : 1 v/v),
2
as for Tol, the a value was found to be 0.96 ^ 0.05. The
ET
residual efficiency due to IQ (a \ 1 [ a ) of 3BP* by
IQ
ET
1245MeOB was found to be 0.04 ^ 0.05.
The rate constants (k and k ) for the BPÈ1245MeOB
ET
IQ
Fig. 4 Logarithmic plots of k (…) and k (L) in ACNÈH O (4 : 1
system in ACNÈH O (4 : 1 v/v) were determined to be
7.3 ] 109 and 3.0 ] 108 dm3 mol~1 s~1, using eqn. (9) and (6),
ET
IQ
2
v/v) and k (|) in Bz for the BPÈAD systems, as a function of the E
2
IQ
ox
of AD. The solid curve was calculated for k by eqn. (12). See text for
ET
respectively.
details.
k
\ a
k
(9)
ET
ET q
The formation of radicals was observed in other BPÈAD
a function of the oxidation potential (E ) of AD. With a
ox
systems in ACNÈH O (4 : 1 v/v), except for the anisole case,
where the deactivation of 3BP* involved only IQ. The rate
constants and efficiencies for ADs are summarized in Table 2.
decrease in the E value of AD, the k value increases and
2
ox
ET
then levels o†. The k
value can be expressed by an
ET
Arrhenius-type expression:46
k
\ A exp([*Gt /RT )
(12)
Quenching of 3BP* by 4MeOTol
ET
ET
where *Gt is the activation energy which is correlated by the
RehmÈWeller equation with the Gibbs energy change (*G
of ET.46
In Bz, 4MeOTol quenched 3BP* efficiently to yield BPK as
well as other ABs. The same procedures for determination of
the rate constants and efficiencies as used for ABs were
ET
)
ET
(13)
employed. In ACNÈH O (4 : 1 v/v) however, simultaneous for-
*Gt \ *G /2 ] [(*G /2)2 ] (*Gt (0))2]1@2
2
ET
ET
ET
ET
mation of both BPK and BP~~ was found. The U
and U
BPK
ani
values were evaluated from the transient absorbances at 545
and 630 nm by eqn. (10) and (7), respectively, since BPK had
no absorption at 630 nm.
where *Gt (0) (\ 2.4 kcal mol~1)59 denotes the Gibbs energy
of activation at *G \ 0. The *G value was evaluated
ET
ET
ET
from:46
U
\ (*A [ *A eani eani~1)eBPK~1I~1 N
(10)
*G \ 23.06(E [ E ) [ E [ d (in kcal mol~1) (14)
ET ox red
BPK
545
630 545 630 545
abs
A
T
where eani is the molar absorption coefficient of BP~~ at 545
where E , E and d represent the reduction potential of BP
545
red
T
nm (2850 dm3 mol~1 cm~1).57 The efficiency of IQ in
([1.83 V vs. SCE in ACN),55 the triplet energy of BP (68.1
kcal mol~1 at 295 K)45 and the calibrating term, respectively.
The solid curve in Fig. 4 was calculated using eqn. (12) with
the best-Ðt parameters, A \ 1.4 ] 1010 dm3 mol~1 s~1 and
d \ 8.0 kcal mol~1. Hence, the ET in the BPÈAD system may
be interpreted by the RehmÈWeller relation. The radical for-
mation process via ET can be analysed into two steps as
follows: according to the conservation rule of spin-
ACNÈH O (4 : 1 v/v) was calculated from eqn. (11).
2
a
\ 1 [ a [ a
(11)
IQ
HA
ET
Discussion
The bimolecular rate constants of quenching (k ) were found
q
to be less than the di†usion limits of Bz (1.1 ] 1010 dm3
multiplicity, a triplet radical pair, 3(BP~~ ] AD~`)
is pro-
mol~1 s~1)55 and ACNÈH O (4 : 1 v/v) (1.4 ] 1010 dm3
cage
2
duced in a solvent cage after ET in a collision complex (3BP*
mol~1 s~1).Ò This indicates that the deactivation of 3BP* by
] AD) formed between 3BP* and AD.
AB and AD proceeds via collision processes. It is reasonable
col
to assume formation of collision complexes immediately after
ET
(3BP* ] AD) ÈÈÈ 3(BP~~ ] AD~`)
(I)
encounter of 3BP* with AB or AD. In ACNÈH O (4 : 1 v/v),
col
cage
2
ET between 3BP* and AD was observed with bimolecular
The rate of ISC of the triplet radical pair would be less than
that of escaping from the solvent cage, owing to its relatively
long triplet lifetime. That is, the triplet radical pair can escape
readily from the solvent cage before ISC to the singlet state
where back ET becomes possible.
rate constants (k ). Such outer-sphere ET in the triplet state
ET
is often explained by the RehmÈWeller mechanism.46 Fig. 4
shows the plots of k for the BPÈAD systems as well as k as
ET
IQ
Ò Data from T. Tanaka, M. Yamaji and H. Shizuka, J. Chem. Soc.,
Faraday T rans., in press.
escape
3(BP~~ ] AD~`)
ÈÈÈ BP~~ ] AD~`
(II)
cage
Table 2
E of AD and efficiencies and rate constants and rate constants obtained in the BPÈAD systems in Bz and ACNÈH O (4 : 1 v/v)a
ox
2
ACNÈH O (4 : 1 v/v)
2
Bz
AD
E b/V
k c
a
d
a
e
k
f
k
g
k g
ox
q
ET
IO
1.0
ET
IO
IO
Anisole
1.76
1.58h
1.49
1.42
1.34
1.12
0.81
1.2 ] 107
1.9 ] 108
5.1 ] 108
6.6 ] 108
3.8 ] 109
5.9 ] 109
7.6 ] 109
0
0
1.2 ] 107
1.1 ] 108
3.7 ] 108
3.9 ] 108
1.5 ] 108
4.7 ] 108
3.0 ] 108
2.1 ] 106
2.0 ] 107
9.1 ] 107
1.3 ] 108
5.1 ] 108
2.0 ] 109
4.5 ] 109
4MeOTol
135MeOB
123MeOB
14MeOB
124MeOB
1245MeOB
0.16
0.27
0.41
0.96
0.92
0.96
0.58
0.73
0.59
0.04
0.08
0.04
3.0 ] 107
1.4 ] 108
2.7 ] 108
3.6 ] 109
5.4 ] 109
7.3 ] 109
a Errors within 5% for rate constants and efficiencies. b vs. SCE in ACN. Data from ref. 49. c Quenching rate constants of 3BP* by AD in dm3
mol~1 s~1. d Efficiency for the formation of BP~~. e Efficiency for IQ of 3BP* by AD. f Rate constant of ET of 3BP* from AD in dm3 mol~1 s~1.
g Rate constant of IQ of 3BP* by AD in dm3 mol~1 s~1. h Obtained by the CT method. See text.
864
J. Chem. Soc., Faraday T rans., 1998, V ol. 94

View Online
Hence, back ET is no longer the major process for reduction
in the efficiency of radical formation. The precursor of the IQ
process should be e†ective before formation of the triplet
radical pair in the solvent cage. Considering the k values in
IQ
both Bz and ACNÈH O (4 : 1 v/v), they follow RehmÈWeller
2
behaviour versus the E . These results imply that the inter-
mediate for the ET process would be common with that for
ox
IQ. In order to describe the mechanism including both ET
and IQ, we propose a triplet exciplex with CT character,
3(BPd~É É ÉADd`)* as a common precursor.
cage
Fig. 5 Logarithmic plots of k in Bz (…) and ACNÈH O (4 : 1 v/v)
(3BP* ] AD) ] 3(BPd~É É ÉADd`)*
(III)
IQ
2
col
cage
(>), and of k
in Bz (L) and ACNÈH O (4 : 1 v/v) (|) for the
HA
2
BPÈAB systems, as a function of E of AB
After collisions between 3BP* and AD, electronic interaction
starts to form a partial CT exciplex [eqn. (III)]. Since little
change in the absorption spectrum of 3BP* was observed
throughout the present study, CT interaction must be weak. It
has been reported that partial CT enhances ISC from the
excited singlet to the triplet state.60h62 Similarly, it may
enhance crossing from the triplet to the ground state [eqn.
(IV)].
ox
not the ETPT process, considering the pK values of ABs (e.g.
a
42 for toluene),63 but HT. We have reported on HT in triplet
exciplexes of benzophenone, that the more protic H-atom is
the more reactive in HT.64,65 In the present case, the E can
ox
be a measure of the degree of CT interaction. As E decreases,
ox
the degree of partial CT may increase. In other words, AB
ISC
with the smaller E has the more protic H-atom(s) to be
3(BPd~É É ÉADd`)* ÈÈÈ BP ] AD
(IV)
ox
cage
transferred to the electron-sufficient carbonyl site in the CT
Considering the change in spin-multiplicity for IQ, the corre-
sponding process is ISC. We have shown that IQ undergoes
the heavy atom e†ect, which is characteristic of ISC.45 Thus,
we tentatively conclude that IQ is enhanced by the CT char-
acter of the triplet exciplexes. In the present work, no radical
formation was observed in Bz whose relative permittivity
e \ 2.3.55 In the CT exciplex in Bz, the rate of radical-ion for-
mation may be less than that of ISC induced by the CT char-
acter. On the other hand, in the more polar solvent, the
partial CT readily proceeds to full charge separation, i.e. ET,
since the radical ions produced are then stabilized by solvent
molecules.
exciplex.
From the viewpoint of bimolecular quenching processes of
3BP* by ABs and ADs (RH), IQ was an inherent deactivation
pathway originating from the CT character in the exciplexes
involved in a di†usion process. 4MeOTol employed in the
present study, anticipated as a hybrid molecule for electron
and H-atom donors, showed both reactions in ACNÈH O
(4 : 1 v/v). Photochemical reactions (ET and HA) concomitant
with IQ depended on the E value of RH, the solvent polarity
2
ox
and the presence of an H-atom to be abstracted by 3BP*.
Consequently, the total deactivation scheme of 3BP* by RH
can be illustrated in Scheme 1.
It was revealed in the present study that IQ competing with
ET and HA was ISC-enhanced by the CT character of the
triplet exciplexes. One should realize that the efficiencies of
photoreactions of triplet ketones are not always unity because
of the inherent IQ process.
ET
3(BPd~É É ÉADd`)* ÈÈÈ 3(BP~~ ] AD~`)
(V)
cage
cage
Owing to the large polarity of ACNÈH O (4 : 1 v/v) (i \
44.5),** the rate of charge separation may increase sufficiently
2
to compete with ISC.
When AD was replaced by AB as a quencher, HA occurred
in both Bz and ACNÈH O (4 : 1 v/v) to yield BPK together
with the corresponding benzyl radicals (BzyR). Considering
Concluding remarks
2
Laser Ñash photolysis studies on IQ in the photoreduction of
BP by ABs and ADs were carried out in both benzene and a
mixture of acetonitrile (ACN) and water (4 : 1 v/v) at 295 K.
We may conclude that (1) the IQ process of 3BP* by AB and
AD was ISC-enhanced by the partial CT character of the
the formation process of these radicals step by step, a triplet
radical pair might be produced via the collision complex,
according to the spin-conservation rule.
HA
(3BP* ] AB) ÈÈÈ 3(BPK ] BzyR)
(VI)
col
cage
triplet exciplexes, 3(BPd~É É ÉRHd`)* . (2) Radical ion forma-
cage
Fig. 5 shows plots of k and k for the BPÈAB systems in
both Bz and ACNÈH O (4 : 1 v/v). Both logarithmic rate con-
stants show a linear correlation with the E of AB. These
linear plots imply CT interaction of precursors for HA and
tion by ET might be accomplished in a polar solvent by
HA
IQ
strong CT interaction in the exciplex. (3) Radical formation
was inferred to be HT in the exciplex, where the more protic
H-atom was readily mobile, rather than ET followed by PT.
(4) The loss in efficiency of photochemical product formation
arose not from back ET, but from the IQ process inherent to
photoreactions via the triplet exciplexes.
2
ox
IQ. By analogy with the BPÈAD system, even in the BPÈAB
system, we assume a CT exciplex, 3(BPd~É É ÉABd`)*
to
cage
induce ISC as a common precursor with HA.
The deactivation processes of 3BP* by AB and AD (RH)
(3BP* ] AB) ÈÈÈ 3(BPd~É É ÉABd`)*
(VII)
(VIII)
(IX)
col
cage
are illustrated in Scheme 1.
ISC
3(BPd~É É ÉABd`)* ÈÈÈ BP ] AB
cage
This work was supported by a ScientiÐc Research Grant-in-
Aid for Priority-Area-Research on Photoreaction (No.
06239101) from the Ministry of Education, Science, Sports
and Culture of Japan.
HA
3(BPd~É É ÉABd`)* ÈÈÈ 3(BPK ] BzyR)
cage
cage
As mentioned for the IQ mechanism in the BPÈAD system,
ISC may be enhanced by the CT character of the triplet exci-
plex [eqn. (VIII)]. As for HA in the CT exciplex [eqn. (IX)], it
can be safely said that the mechanism for BPK formation is
References
1
(a) J. C. Scaiano, J. Photochem., 1973/74, 2, 81; (b) C. M. Previtali
and J. C. Scaiano, J. Chem. Soc. Perkin T rans. 2, 1972, 1667;
1972, 1672; M. V. Encinas, E. A. Lissi and J. C. Scaiano, J. Phys.
Chem., 1980, 84, 948; J. C. Scaiano, G. W. McGimpsey, W. J.
Leigh and S. Jakobs, J. Org. Chem., 1987, 52, 4540.
** The e value of ACNÈH O (4 : 1 v/v) was estimated to be pro-
2
portional to the molar fraction of those of ACN (37.5)55 and H O
2
(80.2).55
J. Chem. Soc., Faraday T rans., 1998, V ol. 94
865

View Online
33 (a) M. Hoshino and H. Shizuka, J. Phys. Chem., 1987, 91, 714; (b)
2
3
4
P. J. Wagner, T op. Curr. Chem., 1976, 66, 1.
J. Bell and H. Linschitz, J. Am. Chem. Soc., 1963, 85, 528.
L. Giering, M. Berger and C. Steel, J. Am. Chem. Soc., 1974, 96,
953.
M. Hoshino and M. Kogure, J. Phys. Chem., 1989, 93, 728.
34 S. Inber, H. Linschitz and S. G. Cohen, J. Am. Chem. Soc., 1980,
102, 1419; 1981, 103, 1048.
5
F. D. Lewis, R. W. Johnson and D. R. Kory, J. Am. Chem. Soc.,
1974, 96, 6100.
35 W. J. Leigh, E. C. Lathioor and M. J. St. Pierre, J. Am. Chem.
Soc., 1996, 118, 12339.
6
7
8
9
H. Murai and K. Obi, J. Phys. Chem., 1975, 79, 2446.
R. W. Yip and W. Siebrand, Chem. Phys. L ett., 1972, 13, 209.
P. Suppan, J. Chem. Soc., Faraday T rans. 2, 1986, 82, 2167.
S. J. Formosinho, J. Chem. Soc., Faraday T rans. 2, 1976, 72,
1313; L. G. Arnaut, S. J. Formosinho and A. M. Da Silva, J.
Photochem., 1984, 27, 185; S. J. Formosinho and L. G. Arnaut,
Adv. Photochem., 1991, 16, 67.
36 C. G. Shaefer and K. S. Peters, J. Am. Chem. Soc., 1980, 102,
7566; J. D. Simon and K. S. Peters, J. Am. Chem. Soc., 1981, 103,
6403; 1982, 104, 6542; 1983, 105, 4875; K. S. Peters and J. Lee, J.
Phys. Chem., 1993, 108, 3761.
37 L. E. Manring and K. S. Peters, J. Am. Chem. Soc., 1985, 107,
6452.
38 H. Ohtani, T. Kobayashi, K. Suzuki and S. Nagakura, Nippon
10 N. J. Turro, in Molecular Photochemistry, Benjamin/Cummings,
Menlo Park, CA, 1978.
11 M. Hoshino and H. Shizuka, in New Aspects of Radiation Curing
in Polymer Science and T echnology, ed. J. P. Foussier and J. F.
Rabek, Elsevier, London, 1993, vol. 2, p. 637.
Kagaku Kaishi, 1984, 00, 1479.
39 C. Devadoss and R. W. Fessenden, J. Phys. Chem., 1990, 94,
4540.
40 C. Devadoss and R. W. Fessenden, J. Phys. Chem., 1991, 95,
7253.
12 M. Hoshino and H. Shizuka, in Photo-induced Electron T ransfer,
ed. M. A. Fox and N. Chanon, Elsevier, Amsterdam, 1988, part
C, p. 313.
13 J. N. Pitts Jr., R. L. Letsinger, R. P. Taylor, J. M. Patterson, G.
Recktenwald and R. B. Martin, J. Am. Chem. Soc., 1959, 81, 1063.
14 W. M. Moore, G. S. Hammond and R. P. Foss, J. Am. Chem.
Soc., 1961, 83, 2789; G. S. Hammond, W. P. Baker and W. M.
Moore, J. Am. Chem. Soc., 1961, 83, 2795; G. S. Hammond and
N. J. Turro, Science, 1963, 142, 1541.
15 C. Walling and M. J. Gibian, J. Am. Chem. Soc., 1965, 87, 3361.
16 S. G. Cohen, D. A. Laufer and W. V. Sherman, J. Am. Chem. Soc.,
1964, 86, 3060; S. G. Cohen and R. J. Baugarten, J. Am. Chem.
Soc., 1967, 89, 3471; S. G. Cohen, N. Stein and H. M. Chao, J.
Am. Chem. Soc., 1968, 90, 521; S. G. Cohen and A. D. Litt, T etra-
hedron L ett., 1970, 11, 837; A. H. Parola, A. W. Rose and S. G.
Cohen, J. Am. Chem. Soc., 1975, 97, 6202; J. Khan, D. Casey, H.
Linschitz and S. G. Cohen, J. Org. Chem., 1991, 56, 6080.
17 S. G. Cohen, A. Parola and G. H. Parsons Jr., Chem. Rev., 1973,
73, 141.
41 E. Haselbach, P. Jacques, D. Pilloud, P. Suppan and E. Vauthey,
J. Phys. Chem., 1991, 95, 7115.
42 H. Miyasaka and N. Mataga, Bull. Chem. Soc. Jpn., 1990, 63, 131;
H. Miyasaka, K. Morita, K. Kamada and N. Mataga, Bull.
Chem. Soc. Jpn., 1990, 163, 3385; H. Miyasaka, K. Morita, K.
Kamada, T. Nagata, M. Kiri and N. Mataga, Bull. Chem. Soc.
Jpn., 1991, 64, 3229; H. Miyasaka, K. Morita, K. Kamada and N.
Mataga, Chem. Phys. L ett., 1991, 178, 504; H. Miyasaka, T.
Nagata, M. Kiri and N. Mataga, J. Phys. Chem., 1992, 96, 8060.
43 J. B. Guttenplan and S. G. Cohen, J. Am. Chem. Soc., 1972, 94,
4040.
44 P. Jacques, J. Photochem. Photobiol. A, 1991, 56, 159.
45 K. Okada, M. Yamaji and H. Shizuka, Chem. Phys. L ett., 1996,
254, 79.
46 D. Rehm and A. Weller, Ber. Bunsen-Ges. Phys. Chem., 1969, 73,
834; Z. Phys. Chem. N.F., 1970, 69, 183; H. Knibbe, D. Rehm
and A. Weller, Ber. Bunsen-Ges. Phys. Chem., 1969, 73, 839.
47 F. Benington, R. D. Morin and L. C. Clark Jr., J. Org. Chem.,
1955, 20, 102.
18 D. E. Pearson and M. Y. Moss, T etrahedron L ett., 1967, 39, 3791.
19 N. Filipescu and F. L. Minn, J. Am. Chem. Soc., 1968, 90, 1544.
20 J. Saltiel, H. C. Curtis and B. Jones, Mol. Photochem., 1970, 2,
331.
48 J. O. Howell, J. M. Goncalves, C. Amatore, L. Klasinc, R. M.
Wightman and J. K. Kochi, J. Am. Chem. Soc., 1984, 106, 3968.
49 A. Zeweig, W. G. Hodgson and W. Jura, J. Am. Chem. Soc., 1964,
86, 4124.
21 D. R. G. Brimage, R. S. Davidson and P. F. Lambeth, J. Chem.
50 E. C. M. Chen and W. E. Wentworth, J. Chem. Phys., 1975, 63,
3183.
Soc. C, 1971, 1241.
22 J. Dedians, J. Phys. Chem., 1971, 75, 181; A. V. Buettner and J.
Dedinas, J. Phys. Chem., 1971, 75, 187.
51 M. Yamaji, Y. Aihara, T. Itoh, S. Tobita and H. Shizuka, J. Phys.
Chem., 1994, 98, 7014.
23 M. R. Topp, Chem. Phys. L ett., 1975, 32, 144.
52 M. Yamaji, T. Tanaka and H. Shizuka, Chem. Phys. L ett., 1993,
202, 191.
24 T. Yamada and L. A. Singer, J. Am. Chem. Soc., 1990, 112, 3926.
25 S. Inber, H. Linschitz and S. G. Cohen, J. Am. Chem. Soc., 1982,
104, 1679; Y. M. A. Naguib, C. Steel, S. G. Cohen and M. A.
Young, J. Phys. Chem., 1987, 91, 3033.
53 R. Bensasson and E. J. Land, T rans. Faraday Soc., 1971, 67, 1904.
54 H. Hiratsuka, T. Okamura, I. Tanaka and Y. Tanizaki, J. Phys.
Chem., 1980, 84, 285.
26 D. Griller, J. A. Howard, P. R. Marriott and J. C. Scaiano, J. Am.
Chem. Soc., 1981, 103, 619; P. K. Das, M. V. Encinas and J. C.
Scaiano, J. Am. Chem. Soc., 1981, 103, 4154; P. K. Das, M. V.
Encinas, S. Steeken and J. C. Scaiano, J. Am. Chem. Soc., 1981,
103, 4162; M. V. Encinas and J. C. Scaiano, J. Am. Chem. Soc.,
1981, 103, 6393.
55 S. L. Murov, I. Carmichael and G. L. Hug, in Handbook of
Photochemistry, Marcel Dekker, New York, 2nd edn., 1993.
56 R. V. Bensasson and J.-C. Gamain, J. Chem. Soc., Faraday T rans.
1, 1980, 76, 1801.
57 (a) M. Yamaji, T. Kiyota and H. Shizuka, Chem. Phys. L ett.,
1994, 226, 199; (b) T. Kiyota, M. Yamaji and H. Shizuka, J. Phys.
Chem., 1996, 100, 672.
58 I. Amada, M. Yamaji, M. Sase, H. Shizuka, T. Shimokage and S.
Tero-Kubota, Res. Chem. Intermed., 1998, 24, 1.
59 D. Rehm and A. Weller, Isr. J. Chem., 1970, 8, 259.
60 S. Iwata, J. Tanaka and S. Nagakura, J. Chem. Phys., 1967, 47,
2203.
27 J. C. Ronfarad-Hart, R. V. Bensasson and J. C. Gramain, Chem.
Phys. L ett., 1983, 96, 31.
28 (a) P. J. Wagner and R. A. Leavitt, J. Am. Chem. Soc., 1970, 92,
5806; 1973, 95, 3669; P. J. Wagner and A. E. Puchalski, J. Am.
Chem. Soc., 1978, 100, 5948; 1980, 102, 6177; P. J. Wagner and H.
M. H. Lam, J. Am. Chem. Soc., 1980, 102, 4167; (b) P. J. Wagner,
R. J. Truman, A. E. Puchalski and R. Wake, J. Am. Chem. Soc.,
1986, 108, 7727.
29 M. Okamoto and H. Teranishi, J. Am. Chem. Soc., 1986, 108,
6378; M. Okamoto, J. Phys. Chem., 1990, 94, 8182.
30 S. Arimitsu and H. Tsubomura, Bull. Chem. Soc. Jpn., 1972, 45,
1357; S. Arimitsu and H. Masuhara, Chem. Phys. L ett., 1973, 22,
543; S. Arimitsu, H. Masuhara, N. Mataga and H. Tsubomura, J.
Phys. Chem., 1975, 79, 1255.
31 H. Masuhara, Y. Maeda, N. Mataga, K. Tomita, H. Tatemitsu,
Y. Sakata and S. Misumi, Chem. Phys. L ett., 1980, 69, 182.
32 S. N. Battacharyya and P. K. Das, J. Chem. Soc., Faraday T rans.
2, 1984, 80, 1107; S. N. Battacharyya and P. K. Das, J. Phys.
Chem., 1986, 90, 3987.
61 C. R. Goldschmit, R. Potashnik and M. Ottolenghi, J. Phys.
Chem., 1971, 75, 1025.
62 B. T. Lim, S. Okajima, A. K. Chandra and E. C. Lim, Chem.
Phys. L ett., 1981, 79, 22.
63 S. H. Pain, in Organic Chemistry, McGraw-Hill Inc., 5th edn.,
1987.
64 M. Yamaji, K. Tamura and H. Shizuka, J. Chem. Soc., Faraday
T rans., 1994, 90, 533.
65 T. Yoshihara, M. Yamaji and H. Shizuka, Chem. Phys. L ett.,
1996, 261, 431.
Paper 7/07900A; Received 3rd November, 1997
866
J. Chem. Soc., Faraday T rans., 1998, V ol. 94