Quenching processes of aromatic hydrocarbons in the higher triplet excited states-energy transfer vs. electron transfer

Article abstract of DOI:10.1039/b400128a

Quenching processes of several aromatic hydrocarbons (AH) such as naphthalene (NAP), dibenz[a,h]anthracene (DBA), and chrysene (CHR) in the higher triplet excited states (T₂) by different quenchers (Q) such as p-dichlorobenzene, o-dicyanobenzene aromatic compounds, and chloroalkanes (RCl), have been investigated by the two-color two-laser excitation method. AH in the higher triplet excited states (AH(Tn, n ≥ 2)) initially generated by the excitation of AH(T₁) at the wavelength tuned to the absorption of AH(T₁). AH(T_n) decays to a AH(T₂) with the longest lifetime among AH(T_n) through the fast internal conversion. In the presence of Q, the competition of triplet energy transfer (TENT) and electron transfer (ELT) reactions between AH(T₂) and Q are expected. However, no AH radical cation was observed, especially when the quenchers were chloroalkanes such as carbon tetrachloride (CCl₄), methylene dichloride (CH₂Cl₂), 1,2-dichloroethane, which are good electron acceptors. It is suggested that the TENT is important during the quenching of AH(T_n) by Q. The lifetimes of NAP(T₂), DBA(T₂), and CHR(T₂) were calculated from the TENT quenching experiments. It was found that the lifetimes of AH(T₂) increase in the order of NAP(T₂) (4.5 ps) < DBA(T₂) (16 ps) < CHR(T₂) (60 ps), which is consistent very well with the energy gap law for the transition from AH(T₂) to AH(T₁).

Full text of DOI:10.1039/b400128a

View Article Online / Journal Homepage / Table of Contents for this issue
R E S E A R C H P A P E R
Quenching processes of aromatic hydrocarbons in the higher triplet
excited states-energy transfer vs. electron transfery
Xichen Cai, Masanori Sakamoto, Michihiro Hara, Sachiko Tojo, Kiyohiko Kawai,
Masayuki Endo, Mamoru Fujitsuka and Tetsuro Majima*
Institute of Scientific and Industrial Research (SANKEN), Osaka University,
Mihogaoka 8-1, Ibaraki, Osaka 567-0047, Japan. E-mail: majima@sanken.osaka-u.ac.jp;
Fax: þ81-6-6879-8499; Tel: þ81-6-6879-8495
Received 6th January 2004, Accepted10th February 2004
F|rst published as an AdvanceArticle on the web 3rd March 2004
Quenching processes of several aromatic hydrocarbons (AH) such as naphthalene (NAP),
dibenz[a,h]anthracene (DBA), and chrysene (CHR) in the higher triplet excited states (T ) by different
2
quenchers (Q) such as p-dichlorobenzene, o-dicyanobenzene aromatic compounds, and chloroalkanes (RCl),
have been investigated by the two-color two-laser excitation method. AH in the higher triplet excited states
(
AH(T
AH(T , n ꢀ2)) initially generated by the excitation of AH(T ) at the wavelength tuned to the absorption of
n
1
1
). AH(T
n
) decays to a AH(T
2
n
) with the longest lifetime among AH(T ) through the fast internal
conversion. In the presence of Q, the competition of triplet energy transfer (TENT) and electron transfer (ELT)
reactions between AH(T ) and Q are expected. However, no AH radical cation was observed, especially when
the quenchers were chloroalkanes such as carbon tetrachloride (CCl ), methylene dichloride (CH Cl ),
2
4
2
2
1,2-dichloroethane, which are good electron acceptors. It is suggested that the TENT is important during the
quenching of AH(T ) by Q. The lifetimes of NAP(T ), DBA(T ), and CHR(T ) were calculated from the TENT
n
2
2
2
quenching experiments. It was found that the lifetimes of AH(T
(
2
2
) increase in the order of NAP(T )
4.5 ps) < DBA(T ) (16 ps) < CHR(T ) (60 ps), which is consistent very well with the energy gap law for the
2 2
2 1
transition from AH(T ) to AH(T ).
the increase of ꢁDG in the Marcus normal region, while k
Introduction
ELT
decreases with an increase of ꢁDG in the Marcus inverted
Quenching processes of molecules in the excited states by
quenchers (Q) have been investigated as an interesting subject
from experimental and theoretical points of view. The occur-
rence of two reactions, energy and electron transfer (ENT
and ELT, respectively), has been well established. The ENT
and ELT are equilibrium controlled processes, and depend
region. k
indicates the maximum value at ꢁDG ¼l.
ELT
Experimental evidence for ELT in the Marcus normal region
has been well established in the intramolecular and intermole-
cular electron donor/acceptor systems, while ELT in the Mar-
1
cus inverted region has been limited to the intramolecular
12–15
donor/acceptor systems with a fixed distance.
tal evidence for ELT in the Marcus inverted region in the inter-
Experimen-
1
,2
mainly on the overall free energy changes (DG). Both ENT
and ELT are permitted to occur competitively at ꢁDG > 0.
Many studies have been carried out to establish the parameters
determining the quenching efficiencies and rates in energy and
electron donor–acceptor systems in which either ENT or ELT
16,17
molecular electron donor/acceptor systems is scarce.
Weller measured kELT and DG for a series of ELT and
observed that kELT becomes constant at large ꢁDG in the Mar-
18
cus inverted region. Therefore, he proposed an empirical
3
,4
occurs. The determining factors for the occurrence of ENT
or ELT are still unclear. Although there is no simple relation-
6
2
6
2 1/2
6¼
expression, DG ¼DG/2 þ[(DG/2) þ(DG (0)) ] , DG (0)
is the value at which DG is zero, to fit the experimental results.
The Weller’s theory has been widely used for explaining kELT
at large ꢁDG in the Marcus inverted region.
5
ship between the reaction rate constant (k) and DG in ENT, it
is believed that ENT occurs at the diffusion controlled rate
ꢁ
1 6,7
constant (kENT ꢂk ) at ꢁDG > 13 kJ mol
diff
.
Intermolecular ELT usually occurs between the lowest
On the other hand, more complex relationship exists
between kELT and DG in ELT. Usually k expressed as
1 1 1
excited states (S , T , or D ) and Q as electron donor or
ELT
acceptor, and the ꢁDG value can be altered by changing the
6
6¼
k
activation energy.
ELT ¼Aexp(ꢁDG /RT), where A is a constant, DG is the
19
redox potential or solvent polarity. Because of considerably
1
According to the Marcus theory on
ELT reactions, a parabolic relationship exists between the
high excitation energy of the higher excited states (S
), various electron donor or acceptor molecules can be used
as Q, and the ꢁDG value can be changed over a wide range.
The higher triplet excited states (T ) is one of the candidates
n n
, T , or
D
n
6
6
2
driving force, DG and DG , DG ¼l/4 (1 þDG/l) , where l
8
–11
is the total reorganization energy.
Therefore, kELT
¼
2
n
Aexp[ꢁ(l/4 (1 þDG/l) )/RT], two kinetic domains are
to study ELT with the ꢁDG value in a wide range, because
defined, the Marcus normal region with ꢁDG < l and the
the lifetimes of anthracene (ANT), 9,10-dibromoanthracene,
Marcus inverted region with ꢁDG > l. k
increases with
ELT
naphthalene (NAP), and chrysene (CHR) in the T
found to be sufficiently long for the occurrence of intermolecu-
n
states were
7
,20–22
lar reactions.
of several aromatic hydrocarbons (AH) such as NAP, CHR,
We now report on the quenching processes
yElectronic supplementary information (ESI) available: The quench-
ing of DBA(T ) by CCl , CHR(T ) by NAP, the evidences of no DBA
n
4
n
and dibenz[a,h]anthracene (DBA) in the T states by several
n
and CHR ions produced during two-color two-laser flash photolysis,
and the evidence of formation of benzene/Cl complex. See http://
www.rsc.org/suppdata/cp/b4/b400128a/
Qs (Scheme 1) using the two-color two-laser flash photolysis
method in which AH(T ) is firstly generated by the excitation
1
T h i s j o u r n a l i s Q T h e O w n e r S o c i e t i e s 2 0 0 4
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 1 7 3 5 – 1 7 4 1
1735

View Article Online
transient absorption spectra. The total system was controlled
with a personal computer via GP-IB interface. To avoid any
damage of the sample solution by the probe light, a suitable
cut-off filter was used in front of the sample.
Experiments and results
Quenching of NAP(T
polar solvents
n n n
), DBA(T ), and CHR(T ) by Q in
In order to generate NAP(T
NAP(S ), benzophenone was used as the triplet sensitizer.
1
) without fluorescence of
2
3
1
NAP(T
through triplet energy transfer (TENT) from benzophen-
1
) with an absorption peak at 415 nm was generated
ꢁ
3
one(T
benzophenone (6.0 ꢃ10
solution was irradiated by a 355-nm laser flash.
was excited using an OPO laser at 425 nm with a delay time
of 100 ns after the first 355-nm laser to give NAP(T ). No
1
) to NAP when a mixture of NAP (6.0 ꢃ10 M) and
ꢁ3
M) in Ar-saturated acetonitrile
2
3,24
1
NAP(T )
Scheme 1 Structures of aromatic hydrocarbons (AH) and quenchers
Q) used in this study.
(
n
change of the transient absorption of NAP(T ) was observed
1
of AH and subsequently excited to AH(T
excitation.
n
) by the second
during and after the irradiation with the second 425-nm laser.
It is suggested that the fast internal conversion (IC) of T
occurred within the second 425-nm laser flash duration (5 ns)
and no other process including photoionization occurred
!T
n
1
from NAP(T
in the acetonitrile solution, the bleaching of the transient
absorption of NAP(T ) at 415 nm (DDOD415) was also
observed immediately after the second 425-nm irradiation
and increased with an increase of [CCl ] or [CH Cl ], and no
) at 415 nm
n 4 2 2
). However, in the presence of CCl or CH Cl
Experimental section
Materials
1
Dibenz[a,h]anthracene (DBA) and chrysene (CHR) were pur-
chased from Aldrich Chem. Co. (purities > 99%) and used
without further purification. Naphthalene (NAP), anthracene
(ANT), biphenyl (BIP), benzophenone, p-dichlorobenzene,
and o-dicyanobenzene were purchased from Nacalai Tesque
4
2
2
recovery of the transient absorption of NAP(T
was observed (Fig. 1).
1
However, when 1,2-dichloroethane was used as Q in place of
CCl or CH Cl , no DDOD415 was observed. On the other
hand, the bleaching and recovery of the transient absorption
of NAP(T ) at 415 nm were observed in the presence of p-
Inc. and recrystallized two times from ethanol before use. Car-
bon tetrachloride (CCl
4
2
2
4 2 2
), methylene dichloride (CH Cl ), and
1
,2-dichloroethane were purchased from Wako Chem. Co.
1
Acetonitrile and benzene were purchased from Nacalai Tesque
Inc. All liquid compounds were spectral grade. Sample solu-
tions were freshly prepared in a 1 cm ꢃ1 cm ꢃ4 cm rectangu-
lar Suprasil cell and deoxygenated by bubbling with argon (Ar)
gas for 20 min before irradiation. All experiments were carried
out at room temperature.
dichlorobenzene or o-dicyanobenzene having no absorption
in the wavelength longer 350 nm in the ground states. The
bleaching increased with an increase of [Q] lower than 1.0
M. The recovery was accomplished in 100% yield and the
recovery time of approximately 100 ns was almost constant
at various [Q] (Fig. 2).
The control experiments performed with the 355- or 425-nm
laser irradiation indicate that Q does not react with NAP(S
benzophenone(S ), NAP(T ), and benzophenone(T ). The
0
),
Two-color two-laser flash photolysis
0
1
1
The laser flashes at 355 and 532 nm were obtained from two
sets of Nd : YAG lasers. The diameter of laser flash was 0.5
cm and duration was 5 ns. The laser energy was adjusted to
experimental results observed during the two-color two-laser
flash photolysis were certainly resulted from quenching pro-
cesses of NAP(T
Similar experiments to those of NAP(T
out for DBA(T ) and CHR(T ) in acetonitrile. First 355-nm
) by Q.
n
ꢁ1
be 5 and 50 mJ pulse at 355 and 532 nm, respectively. The
n
) were carried
ꢁ
1
laser flash at 425 nm (9 mJ pulse ) was obtained from an
OPO laser (Continuum, Surelite OPO) which was pumped
n
n
ꢁ
1
by a Nd : YAG laser (355 nm, 150 mJ pulse ). The delay time
of the second laser flash (532 nm or 425 nm) after the first laser
flash (355 nm) was set in the range of 10 ns–10 ms by a four
channel digital delay/pulse generator (Stanford Research Sys-
tems, Inc., Model DG 535). Two laser beams were parallel to
each other and overlapped through the sample. The 308-nm
ꢁ
1
laser (25 ns, 130 mJ pulse ) was obtained from a XeCl exci-
mer laser (Lambda Physik, COMPex). The probe light was
obtained from a 450 W Xe-lamp (Osram XBO-450). The probe
beam was passed through an iris with a diameter of 0.3 cm and
sent into the sample with a perpendicular intersection of the
laser beams. Additionally, the probe beam was closely
arranged to the entrance side of the laser beam. The probe
beam was then focused on a monochromator (Nikon G250).
The output of the monochromator was monitored using a
PMT (photomultiplier tube; Hamamatsu Photonics R928).
The signal from the PMT was recorded on a transient digitizer
Fig. 1 Time profiles of DOD415 during the two-color two-laser flash
photolysis of a mixture of NAP and benzophenone (the triplet sensiti-
zer) in the absence and presence of CCl (0.25, 0.35, 0.50, and 0.60 M)
4
(
TDS 580D four channel digital phosphor oscilloscope, 1
in Ar-saturated acetonitrile. The initial growth of the transient absorp-
tion of NAP(T ) at 415 nm in the time scale of few tens ns corresponds
GHz, Tektronix). A Hamamatsu Photonics multi-channel
analyzer (C5967) system was used for measurement of the
1
1 1
to TENT from benzophenone(T ) to NAP to give NAP(T ).
1
736
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 1 7 3 5 – 1 7 4 1
T h i s j o u r n a l i s Q T h e O w n e r S o c i e t i e s 2 0 0 4

View Article Online
Fig. 2 Time profiles of DOD415 during the two-color two-laser flash
Fig. 4 Time profiles of the transient absorption at 570 and 415 nm
assigned to DBA(T ) and NAP(T ), respectively, during two-color
two-laser flash photolysis of DBA (3.6 ꢃ10 M) in the absence (a)
and presence of NAP at 0.7 (b) and 0.9 (c) M in Ar-saturated acetoni-
trile solution. Inset shows the amplification of the growth and decay of
ꢁ3
photolysis of a mixture of NAP (6.0 ꢃ10 M) and benzophenone (the
1
1
ꢁ3
ꢁ4
triplet sensitizer) (6.0 ꢃ10 M) in the absence and presence of p-
dichlorobenzene (0.5, 0.7, and 0.9 M) in Ar-saturated acetonitrile.
The inset shows the amplification of the bleaching and recovery of
DOD415 immediately after the second 425-nm laser irradiation. The
NAP(T ) at 415 nm (d) together with the bleaching and recovery of
1
DBA(T ) at 570 nm (e) which were obtained by subtraction of (a) from
1
initial growth of the transient absorption of NAP(T
the time scale of few tens ns corresponds to TENT from benzophen-
one(T ) to NAP to give NAP(T ).
1
) at 415 nm in
(c). The trace lines were obtained from fits to the pseudo-first-order
rate reactions. The initial growth of the transient absorption of
1
1
DBA(T
to the intersystem crossing from DBA(S
minus signal of DOD at 415 nm in the range of 100 ns is due to the
fluorescence of DBA(S ) with the lifetime of 30 ns.
1
) at 570 nm in the time scale of a few tens of ns corresponds
1
) to DBA(T ). The large
1
ꢁ
1
ꢁ1
(
5 mJ pulse ) and second 532-nm (50 mJ pulse ) lasers were
used for two-color two-laser flash photolysis of DBA and
CHR. In the presence of CCl in acetonitrile solution, the
bleaching of the transient absorptions of DBA(T ) at 570 nm
1
4
1
It is clearly shown that TENT from AH(T
to give Q(T ) and AH(S ) based on the experimental results.
n
) to Q occurred
(
DDOD570) and CHR(T
immediately after the second 532-nm laser irradiation and
increased with an increase of [CCl ], and no recovery of the
transient absorptions of DBA(T ) at 570 nm and CHR(T )
1
) at 580 nm (DDOD580) were observed
1
0
n
In the presence of NAP and BIP as Q of DBA(T ) or
4
CHR(T ), the sequential TENT from Q(T ) to AH occurred
n
1
1
1
to give AH(T ). For example, in the presence of NAP as Q
of DBA(T ), DDOD
1
at 580 nm were observed (see electronic supplementary
increased with an increase of [NAP]
570
n
information).y
and the recovery of DOD570 was accomplished in 100% yield
Fig. 4). The growth and decay of the transient absorption at
15 nm assigned to NAP(T ) were observed together with
the bleaching and recovery of the transient absorption of
DBA(T ) at 570 nm (Fig. 4, inset). The growth of the transient
absorption of NAP(T ) increased with an increase of [NAP].
The transient absorption of DBA(T ) was quenched by NAP
in approximately 100% yield calculated from the absorption
When CH
2
Cl
2
or 1,2-dichloroethane was used as Q in place
) and CHR(T ) (DDOD570
and DDOD580 , respectively) was observed. When NAP or
BIP was used in place of CCl , the bleaching and recovery
of DOD570 for DBA(T ) and of DOD580 for CHR(T ) were
(
of CCl
4
, no bleaching for DBA(T
1
1
4
1
4
1
1
1
1
observed. The bleaching increased with an increase of [Q] less
than 1.0 M, and the recovery was accomplished in 100% yield.
Together with DDOD570 for DBA(T
1
1
) or DDOD580 for
coefficients of DBA(T ) and NAP(T ) at 570 and 415 nm of
the peak wavelengths (e570 ¼25 100 and e415 ¼24 500 M
1
1
CHR(T ), a new peak at 415 nm appeared in the presence of
1
ꢁ1
NAP as Q. The experimental results of DBA were shown in
Figs. 3 and 4, the experimental results of CHR were given in
the electronic supplementary information.y
ꢁ1
23
cm , respectively). Because [NAP(T
ꢁ6
1
)] ¼1.0 ꢃ10 M, esti-
ꢁ4
mated from DDOD570 and the growth of NAP(T
1
) at 415 nm,
was much lower than [DBA] ¼3.6 ꢃ10 M, the decay and
growth can be analyzed according to the pseudo-first-order
rate equation (Fig. 4, inset). The rate constants of the growth
1
0
(k
g
) of (e) and decay (k
d
) of (d) were estimated to be 2.0 ꢃ10
ꢁ1 ꢁ1
diff
ꢁ
1
ꢁ1
10
M
acetonitrile.
s
, which is similar to the k ¼1.9 ꢃ10
M
s
in
2
3
Possibility of ELT quenching of AH(T ) by Q
n
In order to elucidate whether ELT from AH(T ) to Q occurs or
n
not, the reference experiments were carried out where radical
cations of NAP, DBA, and CHR (NAP , DBA , and
ꢄ
ꢄ
þ
þ
ꢄ
þ
CHR ) are generated from acceptor-sensitized photoinduced
electron transfer of AH in acetonitrile. The experimental
results for NAP are described below, while those for DBA
and CHR are shown in the electronic supplementary
information.y
It is well known that ELT from NAP(S
1
) to o-dicyanoben-
ꢄ
Fig. 3 Transient absorption spectra observed at 30 ns after the sec-
ond 532-nm laser irradiation during the two-color two-laser flash
photolysis of DBA in the absence (a) (broken line) and presence (b)
(solid line) of NAP (0.9 M) in Ar-saturated acetonitrile solution. The
spectra of (c) (broken line) and (d) (solid line) were observed at 2.5
þ
zene occurs to give NAP and o-dicyanobenzene radical anion
(
lysis of a mixture of NAP and o-dicyanobenzene in acetoni-
ꢄ
o-dicyanobenzene ) during the 308-nm excimer laser photo-
ꢁ
2
5
trile. The experimental results of the 308-nm flash
photolysis and the two-color two-laser flash photolysis were
ms after the second 532-nm laser irradiation.
T h i s j o u r n a l i s Q T h e O w n e r S o c i e t i e s 2 0 0 4
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 1 7 3 5 – 1 7 4 1
1737

View Article Online
between DDOD and [Q]. The inverse of DDOD is represented
by [Q] as the following equation,
shown in Fig. 5. The absorption peaks at 625 and 570 nm
assigned to NAP and NAP dimer radical cation, (NAP ) ,
ꢄ
ꢄ
þ
þ
27,28
2
respectively, were observed during the 308-nm excimer laser
photolysis (Fig. 5A). On the other hand, it was revealed that
jDDODjꢁ ¼b þbðkTENTt½QꢅÞꢁ1
1
;
ð1Þ
ꢄ
þ
no NAP was produced by the ELT from NAP(T
n
) to
energy of NAP(T
T2 ¼385 kJ mol ) is larger than the S energy of NAP
) has a sufficiently long
where b is a constant depending on the experimental condition,
TENT is the rate constant of TENT from AH(T ) to Q, and
t is the lifetime of AH(T ). According to eqn. (1), the Stern–
Volmer plots of |DDOD| vs. [Q] gave a linear line with
an intercept of b and slope of b(kTENTt) . For example, in
the case of quenching DBA(T
was obtained as show in Fig. 6.
Because the T energy (E ) of AH(T ) is expected to be
n
o-dicyanobenzene (Fig. 5B), although the T
2
2
)
ꢁ1
k
n
(
(
E
E
1
ꢁ
1 23
n
ꢁ
S1 ¼384 kJ mol
)
and NAP(T
2
1
ꢁ1
lifetime for the occurrence of bimolecular reactions, as discussed
below.
Consequently, it is clearly evidenced that no ELT occurred
ꢁ1
n 4
) by CCl , a linear line
from NAP(T
n n n
), DBA(T ), or CHR(T ) to Q.
T
n
n
much higher than the T
assumed that TENT from AH(T ) to Q occurs at k . Since
n
the lifetime-dependent quenching should be considered in the
case of quenching of short-lived species such as AH(Tn),
1
energy (E 1
T
1
) of Q(T ), it is reasonably
AH(T
TENT
n
)-sensitized cleavage of C–Cl bond of CCl through
4
diff
2
9,30
When CCl4 was used as a Q for NAP(T ), DBA(T ), and
n
n
k
TENT can be expressed as eqn. (2),
CHR(T
n
), the bleaching and no recovery of the transient
absorptions of AH(T ) were observed. Since neither AH
ꢄ
þ
1
1=2
ꢄ
nor AH was observed during the two-color two-laser flash
ꢁ
kTENT ¼kdiff þkdiff s=ðpDtÞ
;
ð2Þ
photolysis of AH in the presence of CCl , the decomposition
4
ꢄ
where kdiff ¼4pNsD, N is the Avogadro number, s is the reac-
4 1 3
of C–Cl bond of CCl (T ) occurred via TENT to give CCl
ꢄ
tion distance, D is the sum of the diffusion coefficients for the
)
2
6
and Cl radicals. This cleavage mechanism was confirmed
by formation of a benzene–Cl complex during the two-color
two-laser excitation experiments of NAP in the presence of
benzene with various [benzene] (0–2.0 M) in Ar-saturated
excited molecule and quencher molecule. Thus, t of AH(T
n
2
ꢁ7
can be calculated at s ¼0.6 nm and D ¼2.0 ꢃ10
dm
and was equivalent for the Stern–Volmer plots using
ꢁ1 29
s
,
several Qs. The
CHR(T ) were obtained to be 4.5, 16, 60 ps, respectively.
Since neither unimolecular reaction nor luminescence of
AH(T ) was observed, AH(T ) is suggested to be deactivated
through IC to AH(T ). Therefore, the rate constant of the
n n
t values of NAP(T ), DBA(T ), and
CCl (supplementary information).y
4
n
Lifetimes of AH(T
n
) depending on the energy gap law
) occurred within the
n
n
1
Since the formation and decay of AH(T
n
IC (kIC) of AH(T ) is defined to be the inverse of t. Generally
n
duration of the second laser flash (5 ns), the transient phenom-
ena of AH(T ) could not be monitored directly. However, the
lifetime of AH(T ) can be calculated from the relationship
kIC depends on the energy gap between two states, following
the energy gap law for IC, eqn. (3),
n
6
n
ꢁ
1
13
ðsꢁ1Þ¼kIC ꢂ10 expðꢁa DEÞ;
t
ð3Þ
ꢁ
1
where a (eV ) is a constant, DE (eV) is the energy gap
between the zero point vibrational levels of the states under-
ꢁ1
going IC. The a value is usually smaller than 5 eV and does
not change so much for rigid AHs.
eV was calculated from the data of ANT(T ) which decays
6
,29,31
For example, a ¼3.3
ꢁ
1
2
through radiationless processes with t ¼11 ps and DET –T1
¼
2
2
9
ꢁ1
1
.39 eV. With a ¼3.3 eV and t of AH(T
n
) calculated from
the experimental results, the DE values between the T
states (DETn–T1) of NAP, DBA, and CHR were calculated
to be 1.15, 1.54, and 1.94 eV, respectively. The DET2–T1
values between the T and T states for NAP and CHR
1
and
T
n
1
2
are theoretically calculated to be 1.17 and 1.75 eV, while
DET3–T1 values of NAP and CHR are 1.29 and 2.03 eV,
3
2–34
respectively.
assigned to NAP(T
Therefore, NAP(T
n n
) and CHR(T ) are
2
) and CHR(T ), respectively. Since no
2
Fig. 5 The transient absorption spectrum observed at 100 ns after the
308-nm laser irradiation during the 308-nm laser flash photolysis of a
ꢁ3
mixture of NAP (6.0 ꢃ10 M) and o-dicyanobenzene (0.7 M) in Ar-
saturated acetonitrile solution (A) (inset shows the time profiles at 570
nm and 625 nm). The transient absorption spectra observed at 200
after the first 355-nm and 100 ns after the second 425-nm laser irradia-
tion during the two-color two-laser flash photolysis of a mixture of
ꢁ
3
ꢁ3
NAP (6.0 ꢃ10 M) and benzophenone (6.0 ꢃ10 M), in the absence
broken line) and presence of o-dicyanobenzene (0.7 M) (solid line) in
Ar-saturated acetonitrile solution (B) (inset shows the time profiles at
70 nm and 625 nm). The peaks around 415 nm in (A) and (B) are due
to the transient absorption of NAP(T ).
(
ꢁ
1
ꢁ1
4
vs. [CCl ]
5
Fig. 6 Plots of |DDOD570
|
in the two-color two-laser
4
.
1
photolysis of DBA in the presence of CCl
1
738
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 1 7 3 5 – 1 7 4 1
T h i s j o u r n a l i s Q T h e O w n e r S o c i e t i e s 2 0 0 4

View Article Online
excitation energies of the excited molecules and Rehm–Weller
theoretical study has been reported on DBA(T
manifolds of DBA(T ) can not be determined. However,
n
), the triplet
1
8,23
equation, respectively.
DGELT were sufficiently negative for the occurrence of both
TENT and ELT from most of AH(T ) to Q (Table 2).
Occurrence of the TENT quenching of AH(T ) by Q was
It was found that both DGTENT and
n
from the results of NAP, ANT, and CHR, DBA(T
tatively assigned to be DBA(T ). It is suggested that AH(T
initially generated decays through fast IC to a lower triplet
excited state, T state, which has the longest lifetime among
the higher triplet excited states. NAP(T ) having a large
n
) is ten-
2
n
)
n
n
ꢄ
ꢄ
þ
ꢁ
2
observed, while no formation of AH or AH was observed
with the transient absorption measurements. This may be due
n
ꢄ
ꢄ
þ
ꢁ
transition dipole moment was found to be NAP(T10) accord-
ing to theoretical calculation using the QCFF/PI þCISD
to a small formation yield of AH or AH . However, this
possibility is excluded because AH(T ) was excited to give
AH(T ) in 65–100% yield within a laser flash under the experi-
mental conditions. The concentration of AH or AH pro-
duced through ELT from AH(T ) to Q or from Q to AH(T
must be sufficiently high to be detected. Because the transient
1
3
method. All other T
2
n
T
1
(2 < n < 10) transitions are for-
n
ꢄ
ꢄ
þ
ꢁ
bidden or have very small oscillator strengths. Therefore, the
T–T absorption spectrum is assigned to the T10 transi-
tion. In other words, NAP(T10) is generated from the 425-
nm laser excitation of NAP(T ) and decays through fast
IC within ps to NAP(T ) with t of 4.5 ps.
T
1
n
n
)
ꢄ
ꢄ
þ
ꢁ
1
absorption spectra of AH and AH are entirely different
from that of AH(T ) and molar absorption coefficients of the
2
1
ꢄ
absorption peaks of AH and AH are sufficiently high,
ꢄ
þ
ꢁ
The t values of NAP(T
lated from the TENT quenching experiments increased in the
order of NAP(T ) (4.5 ps) < DBA(T ) (16 ps) < CHR(T ) (60
ps). This order is consistent very well with the energy gap
law for the transition from AH(T ) to AH(T ). Therefore,
AH(T ) with the longest lifetime among AH(T ) is responsible
2
), DBA(T
2
), and CHR(T
2
) calcu-
3
9
those can easily be detected. Therefore, the results show that
ELT from AH(T ) to Q does not occur.
2
2
2
n
2
1
2
n
to the species for the TENT quenching. The results were
shown in Table 1 in which the reported lifetime of ANT(T )
No occurrence of ELT quenching of AH(T
n
) by Q
2
2
9
11 ps) was added between NAP(T
The competitive TENT and ELT quenching processes of
(
2
) and DBA(T
2
).
2,5
AH(T
and Q form an encounter complex AH(T
DGTENT > 0 and ꢁDGELT > 0, both TENT and ELT possi-
bly occur. The k value of ELT from AH(T ) to Q can be
n
) by Q are represented in Scheme 2, Firstly, AH(T
n
)
n
)|Q at kdiff . Since
ꢁ
Discussion
ELT
n
estimated from the ꢁDGELT , l, and the electronic coupling
Mechanism of bimolecular quenching processes of AH(T
n
) by Q
8–11,17,40
(V) values according to the Marcus theory,
(eqn. (4)),
Occurrence of TENT and ELT between AH(T ) and Q can be
n
ꢁ
1
2
1=2
2
1
,2,5
kELT ¼kdiff ꢁ1 þ½ðp=lkBTðh=2pÞÞ ꢃjVj
considered in terms of DG.
The DG values for TENT
2
ꢁ1
;
(
DGTENT) and ELT (DGELT) can be estimated from the
ꢃexpðꢁðl þDG Þ=4lk TÞꢅ
ð4Þ
ELT
B
1
0
ꢁ1 ꢁ1
23
in which kdiff ¼1.9 ꢃ10
M
s
in acetonitrile at 298 K,
k is the Boltzmann constant, and T is the temperature, respec-
B
Table 1 Lifetimes (t) of AH(T
2
) and the calculated energy gap
between T
1 2 states (DET2–T1
and T )
tively. The kELT value is the maximum and equivalent with kdiff
.
at ꢁDGELT ¼l. At ꢁDGELT
6
AH
t of T
2
/ps
DET –T1/eV
2
The l value is usually represented by l ¼l þl in which l
v
s
v
s
and l are the inner-sphere reorganization energy and outer-
NAP
ANT
DBA
CHR
4.5
11
1.17
1.39
1.54
1.75
2
sphere reorganization energy, respectively. The lv refers to
the energy change accompanying changes in bond lengths
and bond angles during ELT. The l refers to the solvent effect.
16
60
s
We can discuss the possibility of occurrence of intermolecular
a
n
Table 2 The DGTENT and DGELT values calculated for the TENT and ELT from AH(T ) to Q
ox
1=2
red
E
for
E
Q in DMF or
for
1=2
E 1
T
for
AH/kJ mol
E n
T
for
AH/kJ mol
AH in ACN
(V/SCE)
E
T1 for
DGTENT
/
DGELT
kJ mol
/
Behaviors of
b
AH(T )
1
ꢁ1
ꢁ1
ꢁ1
ꢁ1
ꢁ1
No.
Q
AH
Q/kJ mol
ACN (V/SCE) kJ mol
c
d
1
2
3
4
5
6
7
8
9
CCl
CH Cl
2
4
NAP 255
DBA 218
CHR 239
385
þ1.70
þ1.19
þ1.35
—
—
ꢁ0.6 ꢆꢁ0.8
ꢁ1.3 ꢆꢁ1.5
No data
ꢁ1.60
<0
<0
ꢁ164 ꢆꢁ145 Ble.
d
2
2
2
ꢁ96 ꢆꢁ77
Ble.
No Ble.
DCE
DCB
DCN
—
> 0
–
335
305
—
ꢁ50
ꢁ80
<0
ꢁ67
ꢁ41
Ble. Rec.
Ble. Rec.
ꢁ1.87
e
CCl
CH
4
367
ꢁ0.6ꢆꢁ0.80
ꢁ1.3 ꢆꢁ1.6
No data
ꢁ2.55
ꢁ194 ꢆꢁ175 Ble.
2
Cl
—
—
> 0
> 0
ꢁ127 ꢆꢁ98
No Ble.
No Ble.
DCE
BIP
NAP
—
274
255
—
ꢁ93
ꢁ112
<0
ꢁ6
ꢁ12
Ble. Rec.
Ble. Rec.
10
11
12
13
14
15
ꢁ2.49
f
CCl
CH
4
408
ꢁ0.6 ꢆꢁ0.80
ꢁ1.3 ꢆꢁ1.6
No data
ꢁ2.55
ꢁ220 ꢆꢁ201 Ble.
2
Cl
—
—
> 0
> 0
ꢁ152 ꢆꢁ123 No Ble.
DCE
BIP
NAP
—
ꢁ32
No Ble.
Ble. Rec.
Ble. Rec.
274
255
ꢁ134
ꢁ2.49
ꢁ153
ꢁ37
a
DGTENT ¼ꢁETn(AH) þET1(Q); DGELT ¼96.488 ꢃ[E (AH) ꢁE (Q)] ꢁETn(AH), the values of ET1(AH), Eox (AH), ET1(Q), and Ered (Q) from
ox
1=2
red
1=2
1=2
1=2
b
ref. 23. DCE ¼1,2-dichloroethane, DCB ¼p-dichlorobenzene, DCN ¼o-dicyanobenzene, and ACN ¼acetonitrile. Behavior of the transient
c
1 1
absorption of AH(T ) after the second laser irradiation, bleaching (Ble.) and recovery (Rec.) of the transient absorption of AH(T ). Ref. 32.
d
e
f
n
Ref. 35–38. Calculated from the energy gap law and t of DBA(T ). Refs. 33 and 34.
T h i s j o u r n a l i s Q T h e O w n e r S o c i e t i e s 2 0 0 4
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 1 7 3 5 – 1 7 4 1
1739

View Article Online
Scheme 2 Competitive TENT and ELT quenching processes of
0
AH(T
n
) by Q. k sep and ksep are rate constants of dissociation of the
pair, AH|Q(T ) and AH |Q , to free components, respectively.
ꢄ
ꢄ
þ
ꢁ
1
ELT during the quenching of AH(T
and l for two cases of AH(T )–Q.
In the case of quenching of NAP(T
or o-dicyanobenzene and DBA(T ) and CHR(T
n
) by Q depending on V
n
n
) by p-dichlorobenzene
n
n
) by BIP or
ꢁ
1
ꢁ1
NAP in acetonitrile, V ¼11.5 cm and l ¼181 kJ mol
can be used for analysis of the ELT quenching of AH(T
n
) by
Q in the similar way reported. The V and l values have been
4
1
ꢁ1
Fig. 7 Plots of log(k ) vs. ꢁDG
ELT
ELT
at l ¼181 kJ mol (a) and
ꢁ1
ꢁ1
ꢁ1
reported to be 8.0, 8.5, 11.5 cm and 159, 167, 181 kJ mol
(
l ¼352 kJ mol
n
(b) during the quenching of AH(T ) by Q at
ꢁ1
V ¼11.5 cm . The circles show the aromatic compounds as Q of
1.65, 1.73, 1.88 eV), respectively, for radical ion pairs of radi-
AH(T
n
). The full squares and empty squares show RCl as Q of AH(T
n
)
cal anions of 9,10-dicyanoanthracene and 2,6,9,10-tetracya-
noanthracene with radical cations of aromatic compounds
having three, two, and one benzene rings, respectively, in aceto-
ꢁ1
at l ¼181 and 352 kJ mol , respectively. The curve lines (a) and (b)
ꢁ1
were obtained from eqn. (4) with V ¼11.5 cm and l ¼181 and 352
ꢁ1
kJ mol , respectively. The large l value changes are due to the contri-
4
1
nitrile. The k
1
values were calculated to be less than ca.
bution of C–Cl bond dissociation energy of RCl on the l values.
ELT
v
8
0 M
ꢁ1 ꢁ1
s
. The intermolecular TENT should occur at kdiff
,
values are much larger
Therefore, the intermolecular TENT
kTENT ¼k , because the ꢁDG
diff
TENT
ꢁ
1 6,7
.
than 13 kJ mol
The experimental results provided an evidence for the occur-
rence of only intermolecular TENT during the quenching of
occurs at kTENT ¼k > kELT as the main process.
diff
n n
In the case of quenching of NAP(T ), DBA(T ), and
AH(T
TENT and ELT, as shown in Fig. 8.
Before AH(T ) decays into AH(S ), AH(T
the second laser (hn ) to give AH(T ) Since the t value of
n
) by Q. Potential energy diagrams are suggested for
CHR(T ) by RCl such as CCl , CH Cl , and 1,2-dichloro-
n
ethane in acetonitrile, l
4
2
2
ꢁ
1
s
¼128 kJ mol (1.33 eV) can be
36,37
1
0
1
) is excited with
0
used similar to the reports in which RCl quenching 2,2 -and
0
2
n
4
,4 -biphenyldiols in the S
1
states were studied.
Usually,
n
AH(T ) is sufficiently long for bimolecular reactions to occur,
bimolecular TENT or ELT can occur in the presence of Q.
However, because ELT occurs at kELT which is smaller than
1
,2
l value is quite small or negligible. According to eqn. (4)
v
ꢁ
1
ꢁ1
with V ¼11.5 cm and l ¼128 kJ mol , kELT was calcu-
1
0
ꢁ1 ꢁ1
lated to be ca. 10
(
M s almost equivalent to k
). It is suggested that triplet radical ion pair
diff
kdiff , ELT can not compete with TENT which occurs at kdiff
Consequently, TENT occurs as the main pathway for the
quenching of AH(T ) by Q. Since E 1 values of Q(T ) such
as BIP(T ), NAP(T ), p-dichlorobenzene(T ), and o-dicyano-
benzene(T ) have longer lifetimes than 100 ms, and are higher
than those of AH(T ), the sequential TENT from Q(T ) to AH
occurs to reproduce AH(T ). In the case of RCl as Q, the
.
1
0
ꢁ1 ꢁ1
ꢁ
1.9 ꢃ10
M
AH |RCl ) would be produced to dissociate into AH and
s
ꢄ
ꢄ
ꢄ
3
þ
þ
(
RCl , because the charge recombination between AH and
ꢄ
ꢄ
n
T
1
ꢁ
ꢁ
þ
ꢄ
ꢄ
ꢄ
RCl in (AH |RCl ) giving AH|RCl is a spin forbidden
1
1
1
3
þ
ꢁ
23
ꢄ
process. However, no AH was observed, indicating no
1
4
2
þ
1
1
occurrence of the ELT quenching of AH(T ) by RCl. There-
n
1
ꢁ1
ꢁ1
fore, V ¼11.5 cm and l ¼128 kJ mol can not be used
1
decomposition of RCl(T ) occurs through the C–Cl bond clea-
in the quenching of AH(T ) by RCl. A dissociative electron
n
vage depending on the DE between AH(T ) and RCl(T ) and
n
1
transfer mechanism was applied in the case of RCl quenching
BCEC–Cl . No sequential TENT from RCl(T
to give AH(T1).
1
) to AH occurs
0
0
2
,2 -and 4,4 -biphenyldiols in the S
bond dissociation energy of RCl has serious effect on the l
value. l became unusually high compare to the normal
organic donor–acceptor systems.
eV) was selected. Similarly, the simultaneous C–Cl bond clea-
vage of RCl accompanied TENT or ELT from AH(T ) should
also be considered in our system. Therefore, V ¼11.5 cm
1
states, in which the C–Cl
v
The occurrence of TENT and ELT has been discussed in
terms of the ꢁDG, V, and l values during the quenching of
v
3
6,37
ꢁ1
l ¼352 kJ mol (3.65
n
AH(T ) by Q. However, other factors such as effects of the
structural changes of AH and Q and solvent effects may be
considered during TENT and ELT.
n
ꢁ
1
ꢁ1
and l ¼352 kJ mol can be used. The kELT values were cal-
8
culated to be less than ca. 10 M
ꢁ1 ꢁ1
s . Fig. 7 shows the plots
of log(kELT) vs. ꢁDG
with the different l values (l ¼181
ELT
ꢁ
1
ꢁ1
and 352 kJ mol ) and V ¼11.5 cm mentioned above.
Because different experimental results were observed in the
case of CCl
AH(T ), which can not explained by only ELT mechanism.
4 2 2
, CH Cl , and 1,2-dichloroethane as Q of
n
n n
For example, in the case of quenching DBA(T ) or CHR(T )
by CCl and CH Cl , if only ELT from DBA(T ) or CHR(T )
n
4
2
2
n
to CCl
observed in both CCl and CH Cl as Q. On the other hand,
4
2 2
and CH Cl occurred, the bleaching should be
4
2
2
n
in the case 1,2-dichloroethane as Q of AH(T ), no bleaching
was observed, although 1,2-dichloroethane is good electron
2
7,43
acceptor and should usually be used as electron scavenger.
Therefore, the competition between TENT and ELT are
expected during the quenching of AH(T ) by RCl. Whether
Fig. 8 Potential energy diagrams for AH and Q as the energy and
electron donor and accepter, respectively. ꢁDGTENT > 0 and
ꢁ
n
DGELT > 0 for TENT and ELT, respectively, from AH(T ) to Q.
n
TENT occurs with the formation of Q(T
position of Q(T ) through C–Cl bond cleavage in the case of RCl as
Q.DG denotes the activation energy of ELT. The small arrows show
1
) and/or sequential decom-
the bleaching occurs or not is due to the difference of the
C–Cl bond dissociation energies (BDEC–Cl), 340, > 314, and
1
6¼
ꢁ
1
44
45
> 306 kJ mol
CCl
for 1,2-dichloroethane, CH Cl , and
2 2
n
the pathways of the quenching of AH(T ) by Q through TENT or
4
6
4
,
respectively.
ELT.
1
740
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 1 7 3 5 – 1 7 4 1
T h i s j o u r n a l i s Q T h e O w n e r S o c i e t i e s 2 0 0 4

View Article Online
1
5
G. L. Closs, L. T. Calcaterra, N. J. Green, K. W. Penfield and
J. R. Miller, J. Phys. Chem., 1986, 90(16), 3673.
Conclusions
Quenching processes of several AH(T
n
) such as NAP(T
n
),
16 S. Fukuzumi, O. Inada and T. Suenobu, J. Am. Chem. Soc., 2003,
125(16), 4808.
DBA(T
two-color two-laser excitation method. It was found that
AH(T
, n ꢀ2) initially generated from the second laser excita-
tion of AH(T ) decays to the AH(T ) with the longest lifetime
among various AH(T ) through fast IC. The t values of
NAP(T ), DBA(T ), and CHR(T ) were calculated from the
TENT quenching of AH(Tꢄ ) by Q to be 4.5, 16, and 60 ps,
n n
), and CHR(T ) by several Qs have been studied using
1
7
S. Fukuzumi, K. Ohkubo, H. Imahori and D. M. Guldi, Chem.
Eur. J., 2003, 9(7), 1585.
n
1
1
8
9
D. Rehm and A. Weller, Isr. J. Chem., 1970, 8(2), 259.
D. M. Guldi, C. Luo, M. Prato, E. Dietel and A. Hirsch, Chem.
Commun., 2000, 5, 373.
1
2
n
2
2
2
20 W. G. McGimpsey and J. C. Scaiano, J. Am. Chem. Soc., 1989,
111(1), 335.
n
þ
2
1
X. Cai, M. Hara, K. Kawai, S. Tojo and T. Majima, Chem. Com-
mun., 2003, 2, 222.
respectively. No free AH was observed during two-color
two-laser flash photolysis suggested that ELT was not the
2
2
X. Cai, M. Hara, K. Kawai, S. Tojo and T. Majima, Chem. Phys.
Lett., 2002, 368(3/4), 365.
main decay pathway during the quenching of AH(T
n
) by Q,
although both ꢁDG
and ꢁDG
values were sufficiently
TENT
ELT
23 S. L. Murov, I. Carmichael and G. L. Hug, Handbook of Photo-
chemistry, Marcel Dekker, Inc., New York, 1993.
24 I. Carmichael and G. L. Hug, J. Phys. Chem. Ref. Data, 1986,
large for competitive quenching by TENT and ELT. Studying
the quenching processes of molecules in the Tn states by Q
using the two-color two-laser flash photolysis technique pro-
vide a novel method to investigate the nature of energy and
1
5(1), 1.
25
26
27
A. Tsuchida, Y. Tsujii, S. Ito, M. Yamamoto and Y. Wada,
J. Phys. Chem., 1989, 93(4), 1244.
T. Gannon and W. G. McGimpsey, J. Org. Chem., 1993, 58(21),
5639.
A. Ishida, M. Fukui, H. Ogawa, S. Tojo, T. Majima and S.
n
electron transfer processes from the T states.
Takamuku, J. Phys. Chem., 1995, 99(27), 10 808.
28 T. Majima, M. Fukui, A. Ishida and S. Takamuku, J. Phys.
Chem., 1996, 100(21), 8913.
Acknowledgements
This work has been partly supported by a Grant-in-Aid for
Scientific Research on Priority Area (417), 21 Century COE
Research, and others from the Ministry of Education, Culture,
Sports, Science and Technology (MEXT) of the Japanese
Government.
29 C. Bohne, S. R. Kennedy, R. Boch, F. Negri, G. Orlandi, W.
Siebrand and J. C. Scaiano, J. Phys. Chem., 1991, 95(25), 10 300.
st
3
0
J. C. Andre, M. Niclause and W. R. Ware, Chem. Phys., 1978,
8(3), 371.
2
3
1
P. Brodard, A. Sarbach, J. C. Gumy, T. Bally and E. Vauthey,
J. Phys. Chem. A, 2001, 105(27), 6594.
3
3
2
3
M. Z. Zgierski, J. Chem. Phys., 1997, 107(19), 7685.
S. Koshihara and T. Kobayashi, J. Chem. Phys., 1986, 85(3),
1
211.
34 S. Koshihara, T. Kobayashi and S. Iwashima, Chem. Phys. Lett.,
986, 124(4), 331.
References
1
1
2
G. J. Kavarnos and N. J. Turro, Chem. Rev., 1986, 86(2), 401.
G. J. Kavarnos, Fundamentals of Photoinduced Electron Transfer,
VCH Publisher, Inc., New York, 1993.
35 L. Meites and P. Zuman, CRC Handbook Series in Organic Elec-
trochemistry, CRC Press, Inc., Cleveland, 1976, Vol. I.
36 J. Mohanty, H. Pal, S. K. Nayak, S. Chattopadhyay and A. V.
Sapre, J. Chem. Phys., 2002, 117(23), 10 744.
37 J. Mohanty, H. Pal and A. V. Sapre, J. Chem. Phys., 2002,
116(18), 8006.
38 C. Costentin, M. Robert and J.-M. Saveant, J. Am. Chem. Soc.,
2003, 125(35), 10 729.
39 T. Shida, Electronic Absorption Spectra of Radical Ions, Elsevier
Science Publishers Inc., Tokyo, 1988.
3
4
5
F. Wilkinson and C. Tsiamis, J. Am. Chem. Soc., 1983,
1
05(4), 767.
K. Gollnick and A. Schnatterer, Photochem. Photobiol., 1986,
3(4), 365.
4
M. A. Fox and M. Chanon, Photoinduced Electron Transfer Part
A. Conceptual Basis, Elsevier Science Publishing Inc., Tokyo,
1988.
6
7
N. J. Turro, Modern Molecular Photochemistry, The Benjamin/
Cummings Publishing Co., Inc., Menlo Park, California, 1978.
R. S. H. Liu and J. R. Edman, J. Am. Chem. Soc., 1969,
91(6), 1492.
R. A. Marcus, J. Chem. Phys., 1956, 24, 966.
R. A. Marcus, J. Phys. Chem., 1963, 67, 853.
R. A. Marcus, J. Chem. Phys., 1965, 43(2), 679.
R. A. Marcus and N. Sutin, Biochim. Biophys. Acta, 1985, 811(3),
40 S. Fukuzumi, K. Ohkubo, T. Suenobu, K. Kato, M. Fujitsuka
and O. Ito, J. Am. Chem. Soc., 2001, 123(35), 8459.
41 I. R. Gould, D. Ege, J. E. Moser and S. Farid, J. Am. Chem. Soc.,
1990, 112(11), 4290.
42 O. Nicolet and E. Vauthey, J. Phys. Chem. A, 2003, 107(31),
5894.
43 N. Ichinose and T. Majima, Chem. Phys. Lett., 2000, 322(1–2), 15.
44 J. Cioslowski, G. Liu and D. Moncrieff, J. Am. Chem. Soc., 1997,
119(47), 11 452.
45 R. Greenhorn, M. A. Jamieson and N. Serpone, J. Photochem.,
1984, 27(3), 287.
8
9
0
1
1
1
2
65.
J. R. Miller, L. T. Calcaterra and G. L. Closs, J. Am. Chem. Soc.,
984, 106(10), 3047.
1
2
1
1
1
3
4
G. McLendon and R. Hake, Chem. Rev., 1992, 92(3), 481.
I. R. Gould and S. Farid, Acc. Chem. Res., 1996, 29(11), 522.
46 D. R. Lide, CRC Handbook of Chemistry and Physics, CRC Press
INC., Tokyo, 1992–1993.
T h i s j o u r n a l i s Q T h e O w n e r S o c i e t i e s 2 0 0 4
P h y s . C h e m . C h e m . P h y s . , 2 0 0 4 , 6 , 1 7 3 5 – 1 7 4 1
1741