Photoalkylation of 10-alkylacridinium ion via a charge-shift type of photoinduced electron transfer controlled by solvent polarity

Article abstract of DOI:10.1021/ja004311l

A charge-shift type of photoinduced electron-transfer reactions from various electron donors to the singlet excited state of 10-decylacridinium cation (DeAcrH⁺) in a nonpolar solvent (benzene) is found to be as efficient as those of 10-methylac

Full text of DOI:10.1021/ja004311l

J. Am. Chem. Soc. 2001, 123, 8459-8467
8459
Photoalkylation of 10-Alkylacridinium Ion via a Charge-Shift Type of
Photoinduced Electron Transfer Controlled by Solvent Polarity
,
†
†
†
†
Shunichi Fukuzumi,* Kei Ohkubo, Tomoyoshi Suenobu, Kouta Kato,
‡
‡
Mamoru Fujitsuka, and Osamu Ito
Contribution from the Department of Material and Life Science, Graduate School of Engineering, Osaka
UniVersity, CREST, Japan Science and Technology Corporation (JST), Suita, Osaka 565-0871, Japan, and
Institute of Multidisciplinary Research for AdVanced Materials, Tohoku UniVersity, CREST, Japan Science
and Technology Corporation (JST), Sendai, Miyagi 980-8577, Japan
ReceiVed December 20, 2000. ReVised Manuscript ReceiVed June 14, 2001
Abstract: A charge-shift type of photoinduced electron-transfer reactions from various electron donors to the
+
singlet excited state of 10-decylacridinium cation (DeAcrH ) in a nonpolar solvent (benzene) is found to be
+
+
as efficient as those of 10-methylacridinium cation (MeAcrH ) and DeAcrH in a polar solvent (acetonitrile).
+
Irradiation of the absorption bands of MeAcrH in acetonitrile solution containing tetraalkyltin compounds
R4Sn) results in the efficient and selective reduction of MeAcrH to yield the 10-methyl-9-alkyl-9,10-
dihydroacridine (AcrHR). The same type of reaction proceeds in benzene when MeAcrH is replaced by
DeAcrH which is soluble in benzene. The photoalkylation of R′AcrH (R′ ) Me and De) also proceeds in
acetonitrile and benzene using 4-tert-butyl-1-benzyl-1,4-dihydronicotinamide (Bu BNAH) instead of R4Sn,
yielding MeAcrHBu . The quantum yield determinations, the fluorescence quenching of R′AcrH by electron
+
(
+
+
+
t
t
+
donors, and direct detection of the reaction intermediates by means of laser flash photolysis experiments indicate
+
that the photoalkylation of R′AcrH in benzene as well as in acetonitrile proceeds via photoinduced electron
t
+
transfer from the alkylating agents (R4Sn and Bu BNAH) to the singlet excited states of R′AcrH . The limiting
quantum yields are determined by the competition between the back electron-transfer process and the bond-
cleavage process in the radical pair produced by the photoinduced electron transfer. The rates of back electron
transfer have been shown to be controlled by the solvent polarity which affects the solvent reorganization
0
energy of the back electron transfer. When the free energy change of the back electron transfer (∆G bet) in a
polar solvent is in the Marcus inverted region, the rate of back electron transfer decreases with decreasing the
solvent polarity, leading to the larger limiting quantum yield for the photoalkylation reaction. In contrast, the
0
opposite trend is obtained when the ∆G bet value is in the normal region: the limiting quantum yield decreases
with decreasing the solvent polarity.
Introduction
electron transfer may be largely canceled out when the free
energy change of electron transfer is expected to be rather
independent of the solvent polarity. On the other hand, the
Electron transfer reactions are normally performed in polar
solvents such as acetonitrile, in which the product ions of the
electron transfer are stabilized by the strong solvation. When
8
1-5
solvent reorganization energy for the electron-transfer reaction
9
+
is expected to decrease with decreasing the solvent polarity. It
a cationic electron acceptor (A ) is employed in electron-transfer
has been reported that a decrease in the solvent reorganization
energy with decreasing the solvent polarity results in an increase
in the quantum yield for formation of biphenyl radical cation
produced in the photoinduced electron transfer from biphenyl
reactions with a neutral electron donor (D), the electron transfer
+
•+
from D to A produces a radical cation (D ) and a neutral
•
6,7
radical (A ). In such a case, the solvation before and after the
*
To whom correspondence should be addressed. E-mail: fukuzumi@
+ 8
to 10-methylacridinium cation (MeAcrH ). This result indicates
that separation within the initially formed radical cation/radical
ap.chem.eng.osaka-u.ac.jp.
†
Osaka University.
Tohoku University.
‡
pair can compete more effectively with the back electron transfer
(
1) Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-VCH:
Weinheim, 2001; Vol. 1-5.
2) Photoinduced Electron Transfer; Fox, M. A., Chanon, M., Eds.;
Elsevier: Amsterdam, 1988.
3) (a) Kochi, J. K. Angew. Chem., Int. Ed. Engl. 1988, 27, 1227. (b)
in nonpolar solvents as compared to that in polar solvents.8
Although the utility of photoinduced electron transfer reactions
is expected to be considerably extended if charge-shift type of
electron-transfer reactions is developed in nonpolar solvents,
there have so far been very few studies on photoinduced
(
(
Eberson, L. Electron-Transfer Reactions in Organic Chemistry; ReactiVity
and Structure; Springer: Heidelberg, 1987; Vol. 25.
(
4) (a) M u¨ ller, F.; Mattay, J. Chem. ReV. 1993, 93, 99. (b) Mella, M.;
Fagnoni, M.; Freccero, M.; Fasani, E.; Albini, A. Chem. Soc. ReV. 1998,
7, 81.
(6) Fukuzumi, S. In AdVances in Electron-Transfer Chemistry; Mariano,
P. S., Ed.; JAI Press: Greenwich, 1992; Vol. 2, p 65.
(7) Fukuzumi, S.; Tanaka, T. In Photoinduced Electron Transfer; Fox,
M. A., Chanon, M., Eds.; Elsevier: Amsterdam, 1988; Part C, p 578.
(8) Todd, W. P.; Dinnocenzo, J. P.: Farid, S.; Goodman, J. L.; Gould,
I. R. J. Am. Chem. Soc. 1991, 113, 3601.
(9) (a) Marcus, R. A. Annu. ReV. Phys. Chem. 1964, 15, 155. (b) Marcus,
R. A. Angew. Chem., Int. Ed. Engl. 1993, 32, 1111. (c) Eberson, L. AdV.
Phys. Org. Chem. 1982, 18, 79.
2
(
5) (a) Julliard, M.; Chanon, M.; Chem. ReV. 1983, 83, 425. (b) Lewis,
F. D. Acc. Chem. Res. 1986, 19, 401. (c) Yoon, U. C.; Mariano, P. S. Acc.
Chem. Res. 1992, 25, 233. (d) Kavarnos, G. J.; Turro, N. J. Chem. ReV.
986, 86, 401. (e) Gaillard, E. R.; Whitten, D. G. Acc. Chem. Res. 1996,
1
2
1
9, 292. (f) Rathore, R.; Kochi, J. K. AdV. Phys. Org. Chem. 2000, 35,
93.
1
0.1021/ja004311l CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/10/2001

8460 J. Am. Chem. Soc., Vol. 123, No. 35, 2001
Fukuzumi et al.
electron-transfer reactions leading to the stable products in
nonpolar solvents in comparison with those in polar solvents.
Electron donors (benzene, toluene, ethylbenzene, cumene, o-xylene,
m-xylene, p-xylene, 1,2,3-trimethylbenzene, 1,2,4-trimethylbenzene,
8,10
1
,3,5-trimethylbenzene 1,2,3,4-tetramethylbenzene, 1,2,3,5-tetrameth-
We report herein the first systematic study on photoinduced
electron-transfer reactions from a variety of neutral electron
donors to the singlet excited states of organic cations, 10-
ylbenzene, and 1,2,4,5-tetramethylbenzene, pentamethylbenzene, hex-
amethylbenzene, triphenylamine, N,N-dimethylaniline, ferrocene, deca-
methylferrocene) were obtained commercially and purified by the
+
methylacridinium cation (MeAcrH ), and 10-decylacridinium
18
standard method. Tetraalkyltin compounds [tetramethyltin (Me
4
Sn),
+
cation (DeAcrH ) in polar and nonpolar solvents, respectively.
i
tetraethyltin (Et
4
Sn), and tetra-iso-propyltin (Pr
commercially from Aldrich, and di-tert-butyldimethyltin (Bu Me Sn)
was prepared by the literature procedure. The tert-butylated BNAH
(Bu BNAH) was prepared by the Grignard reaction with BNA Cl .
9-Alkyl-9,10-dihydro-10-methylacridine (AcrHR) was prepared as
reported previously.20 Acetonitrile, dichloromethane, chloroform, and
4
Sn)] were obtained
t
It is found that the photoinduced electron-transfer reactions in
nonpolar solvents proceed as efficiently as those in polar
solvents. It is also found that photoinduced electron-transfer
2
2
12
t
+
- 11,19
+
+
reactions of MeAcrH and DeAcrH with organometallic
alkylating agents such as tetraalkyltin compounds (R4Sn) and
t
benzene used as solvents were purified and dried by the standard
4
-tert-butyl-1-benzyl-1,4-dihydronicotinamide (Bu BNAH) which
procedure.18 [ H
2
CN) was obtained from EURI SO-
3
]acetonitrile (CD
3
1
1
are known as novel organic alkylating agents lead to the
reductive alkylation to yield the corresponding alkylated dihy-
droacridine (MeAcrHR or DeAcrHR) as a stable product. The
efficiency for the product formation determined as the limiting
quantum yield is highly dependent on the solvent polarity and
the electron donor properties of the alkylating agents. The choice
of a series of R4Sn as electron donors is expected to benefit
from an additional advantage other than the ability to act as
alkylating reagents, that is the large inner-sphere reorganization
TOP, France. Tetrabutylammonium perchlorate (TBAP), obtained from
Fluka Fine Chemical, was recrystallized from ethanol and dried in vacuo
prior to use. Tetrahexylammonium perchlorate (THAP) was prepared
by the addition of NaClO
4
to the tetrahexylammonium bromide in
acetone, followed by recrystallization from acetone.
2
Reaction Procedure. Typically, an [ H
3
3
]acetonitrile (CD CN)
3
+
-2
solution (0.8 cm ) containing MeAcrH (1.0 × 10 M) in an NMR
tube sealed with a rubber septum was deaerated by bubbling with argon
gas through a stainless steel needle for 5 min. After an alkylating agent
t
-1
[R
4
Sn (2 µL) or Bu BNAH (2.4 mg)] was added to the solution, the
energy (λ ) 41 kcal mol ) associated with the electron-transfer
solution was irradiated with a xenon lamp through a UV-cut filter (λ
310 nm) at room temperature. After the reaction was complete, when
the solution became colorless, the reaction solution was analyzed by
12,13
oxidation.
The large reorganization energy for the electron-
<
transfer oxidation of R4Sn leads to extend the boundary of the
Marcus inverted region to the highly exergonic region (-1.8
1
1
H NMR spectroscopy. The H NMR measurements were performed
0
eV). Since the one-electron oxidation potentials (E ox) of R4Sn
using a Japan Electron Optics JNM-GSX-400 (400 MHz) NMR
+
can be finely tuned by the choice of alkyl groups (R), the free
energy change of back electron transfer can be so widely altered
by changing R and the solvent polarity as to lie in the Marcus
normal or inverted region. The reactivities of R4Sn can also be
spectrometer. The products for the photochemical reaction of MeAcrH
1
with R
spectra with those of the authentic samples.
Fluorescence Quenching. Quenching experiments of the fluores-
4
Sn were identified as MeAcrHR by comparing the H NMR
20
+
+
t
cence of MeAcrH and DeAcrH by electron donors were performed
using a Shimadzu RF-5000 fluorescence spectrophotometer. The
compared with the organic electron donor (Bu BNAH) which
has a much smaller reorganization energy (λ ) 22.0 kcal
+
+
excitation wavelength was 398 nm for MeAcrH and DeAcrH . The
monitoring wavelengths were those corresponding to the maxima of
the emission bands at λ ) 488 and 498 nm, respectively. The solutions
were deoxygenated by argon purging for 10 min prior to the measure-
ments. Relative emission intensities were measured for acetonitrile
-
1 14
mol ) Thus, the present study provides an excellent op-
portunity to clarify how the reactivity of charge-shift type of
photoinduced electron transfer reactions which lead to stable
products can be finely controlled by the solvent polarity.
+
+
-5
(
MeCN) solution containing MeAcrH or DeAcrH (5.0 × 10 M)
-3
Experimental Section
with electron donors at various concentrations (3.0 × 10 - 2.5 M).
There was no change in the shape but there was a change in the intensity
of the fluorescence spectrum by the addition of an electron donor. The
Stern-Volmer relationship (eq 1)
Materials. 10-Methylacridinium iodide was prepared by the reaction
of acridine with methyl iodide in acetone, and it was converted to the
+
-
4 4 2
) by addition of Mg(ClO ) to the
perchlorate salt (MeAcrH ClO
iodide salt, and purified by recrystallization from methanol.
15,16
I0/I ) 1 + KSV[D]
(1)
Likewise, 10-decylacridinium hexafluorophosphate was prepared by
the reaction of acridine with decyl iodide in acetone and the subsequent
metathesis with sliver hexafluorophosphate, followed by recrystalliza-
was obtained for the ratio of the emission intensities in the absence
and presence of quenchers (I /I) and the concentrations of quenchers
[D]. The fluorescence lifetimes τ of MeAcrH and DeAcrH were
determined as 31 ns in MeCN, 30 ns in CHCl and 29 ns in benzene
by single photon counting using a Horiba NAES-1100 time-resolved
8
tion from methanol. 9-Substituted 10-methylacridinium perchlorates
0
+
-
i
+
+
(
MeAcrR ClO
4
2
: R ) Pr , CH Ph, and Ph) were prepared by the
reaction of 10-methylacridone in dichloromethane with the correspond-
ing Grignard reagents (RMgX), then addition of sodium hydroxide for
the hydrolysis and perchloric acid for the neutralization, and purified
3
spectrofluorophotometer. The observed quenching rate constants k ()
q
1
7
-1
by recrystallization from ethanol-diethyl ether.
K τ ) were obtained from the Stern-Volmer constants KSV and the
SV
emission lifetimes τ.
(10) Although the radical/radical cation pair was observed in benzene
Electrochemical Measurements. Cyclic voltammetry measurements
were performed at 298 K on a BAS 100W electrochemical analyzer in
deaerated solvent containing 0.1 M tetrabutylammonium perchlorate
in competition with return electron transfer in photoinduced charge-shift
+
reaction from biphenyl to 10-decylacridinium cation (DeAcrH ), products
in such nonpolar solvent have yet to be reported in ref 8.
(TBAP) or tetrahexylammonium perchlorate (THAP) as supporting
(
11) Fukuzumi, S.; Suenobu, T.; Patz, M.; Hirasaka, T.; Itoh, S.;
Fujitsuka, M.; Ito, O. J. Am. Chem. Soc. 1998, 120, 8060.
12) Fukuzumi, S.; Wong, C. L.; Kochi, J. K. J. Am. Chem. Soc. 1980,
3, 2928.
electrolyte. A conventional three-electrode cell was used with a gold
2
(
working electrode (surface area of 0.3 mm ) and a platinum wire as
2
2
1
the counter electrode. The Pt working electrode (BAS) was routinely
(13) Fukuzumi, S.; Kuroda, S.; Tanaka, T. J. Chem. Soc., Perkin Trans.
1986, 25.
(17) Fukuzumi, S.; Ohkubo, K.; Tokuda, Y.; Suenobu, T. J. Am. Chem.
Soc. 2000, 122, 4286.
(18) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals; Butterworth-Heinemann: Oxford, 1988.
(19) Anne, A. Heterocycles 1992, 34, 2331.
(20) Fukuzumi, S.; Tokuda, Y.; Kitano, T.; Okamoto, T.; Otera, J. J.
Am. Chem. Soc. 1993, 115, 8960.
(
14) Patz, M.; Kuwahara, Y.; Suenobu, T.; Fukuzumi, S. Chem. Lett.
997, 567.
15) Roberts, R. M. G.; Ostovic, D.; Kreevoy, M. M. Faraday Discuss.
Chem. Soc. 1982, 74, 257.
16) Fukuzumi, S.; Koumitsu, S.; Hironaka, K.; Tanaka, T. J. Am. Chem.
Soc. 1987, 109, 305.
(
(

Photoalkylation of 10-Alkylacridinium Ion
J. Am. Chem. Soc., Vol. 123, No. 35, 2001 8461
Table 1. Fluorescence Quenching Rate Constants (k
Potential of Electron Donors, and the Free Energy Change of Photoinduced Electron Transfer (∆G et
q
) of MeAcrH+ and DeAcrH+ by Electron Donors in Various Solvents, Oxidation
0
)
k
q
,c M-1 s-1
CHCl
E0ox vs SCEa
in MeCN, V
MeCN
MeAcrH
3
6 6
C H
DeAcrH
+
+
+
no.
electron donor
benzene
toluene
ethylbenzene
cumene
m-xylene
o-xylene
1,3,5-trimethylbenzene
p-xylene
1,2,3-trimethylbenzene
1,2,4-trimethylbenzene
1,2,3,4-tetramethylbenzene
1,2,3,5-tetramethylbenzene
1,2,4,5-tetramethylbenzene
pentamethylbenzene
hexamethylbenzene
triphenylamine
∆G0et,b eV
MeAcrH
7
6
1
2
3
4
5
6
7
8
9
2.35
2.20
2.14
2.14
2.02
1.98
1.98
1.93
1.88 (1.98)
1.79 (1.89)
1.71 (1.81)
1.71 (1.77)
1.63 (1.75)
1.58 (1.68)
1.49 (1.60)
0.84
0.03
-0.12
-0.18
-0.18
-0.30
-0.34
-0.34
-0.39
-0.44
-0.53
-0.61
-0.61
-0.69
-0.74
-0.83
-1.48
-1.61
-1.95
-2.52
1.3 × 10
2.4 × 10
4.5 × 10
2.4 × 10
7.7 × 10
7.9 × 10
1.2 × 10
8.6 × 10
1.3 × 10
1.3 × 10
1.5 × 10
1.6 × 10
1.9 × 10
1.9 × 10
2.1 × 10
2.2 × 10
2.6 × 10
-
7.6 × 10
6.6 × 10
1.0 × 10
4.4 × 10
3.5 × 10
5.5 × 10
5.0 × 10
4.2 × 10
8.1 × 10
7.2 × 10
7.1 × 10
1.3 × 10
6.7 × 10
9.9 × 10
8.4 × 10
-
-
8d
7
7
7
7
9
9
9
3.5 × 10
4.5 × 10
5.7 × 10
1.8 × 10
2.0 × 10
2.7 × 10
-
8
8
9
9
1
9
1
1
1
1
1
1
1
1
1
8
8
d
d
0
d
0
0
0
0
0
0
0
0
0
9d
9d
9
9d
9
9
3.9 × 10
6.2 × 10
7.9 × 10
7.2 × 10
9.1 × 10
1.1 × 10
1.1 × 10
-
9
9
10
11
12
13
14
15
16
17
18
19
9
9
10
9
9
9
9
10
10
9
10
10
10
N,N-dimethylaniline
ferrocene
decamethylferrocene
0.71
0.37
-0.20
1.7 × 10
2.1 × 10
1.6 × 10
1.3 × 10
1
1
0
0
-
-
1.9 × 10
a
. b Determined from the E0red value of MeAcrH * () 2.32 V vs SCE in MeCN) and eq 2.
1
+
22
Values in parentheses are determined in CH
2
Cl
2
c
d
The experimental error is (5%. Taken from ref 23.
cleaned by soaking it in concentrated nitric acid, followed by repeating
rinsing with water and acetone, drying at 353 K prior to use to avoid
possible fouling of the electrode surface. The reference electrode was
light of λ ) 398 nm from a Shimadzu RF-5000 fluorescence
spectrophotometer. Under the conditions of actinometry experiments,
+
+
both the actinometer and MeAcrH or DeAcrH absorbed essentially
all the incident light. The light intensity of monochromatized light of
3
an Ag/0.01 M AgNO . The cyclic voltammograms were measured with
-
8
-1
various sweep rates in a deaerated solvent containing TBAP (0.10 M)
or THAP (0.50 M) used as a supporting electrolyte at 298 K.
The second harmonic ac voltammetry (SHACV) measurements were
performed on a BAS 100B electrochemical analyzer in deaerated MeCN
containing 0.10 M TBAP as a supporting electrolyte at 298 K to
determine the one-electron oxidation potentials of alkylbenzenes and
triphenylamine. The platinum working electrode (BAS) was polished
with BAS polishing alumina suspension and rinsed with acetone before
λ ) 398 nm was determined as 1.8 × 10 einstein s with the slit
width of 20 nm. The photochemical reaction was monitored by using
Hewlett-Packard 8452A and 8453 diode-array spectrophotometers. The
quantum yields were determined from the decrease in absorbance due
+
4
-1
-1 22
+
to MeAcrH (λ ) 358 nm, ꢀ ) 1.8 × 10 M cm
) and DeAcrH
4
-1
-1
(λ ) 362 nm, ꢀ ) 2.0 × 10 M cm ).
Laser Flash Photolysis. The measurements of transient absorption
+
spectra in the photochemical reactions of DeAcrH with R
4
Sn in
0
use. The counter electrode was a platinum wire (BAS). The E ox values
benzene were performed according to the following procedures. The
benzene solution was deoxygenated by argon purging for 10 min prior
to the measurement. The deaerated benzene solution containing
+
(
vs Ag/Ag ) are converted to those vs SCE by adding 0.29 V.
•
ESR Measurements. The MeAcrPh was generated by the electron-
transfer reduction of MeAcrPh ClO
+
-
-4
-3
(1.7 × 10 - 5.0 × 10 M)
+
-5
-3
4
DeAcrH (5.0 × 10 M) and R
Sn (1.0 × 10 - 2.5 M) was excited
4
-
5
with tetramethylsemiquinone radical anion (2.0 × 10 M) generated
by comproportionation of tetramethyl-p-benzoquinone and tetramethyl-
hydroquinone with tetrabutylammonium hydroxide. The solution
containing the radical was transferred to an ESR tube under an
atmospheric pressure of Ar. The ESR spectra were measured at various
temperatures with a JEOL X-band spectrometer (JES-RE1XE). The
ESR spectra were recorded under nonsaturating microwave power
conditions. The magnitude of modulation was chosen to optimize the
resolution and the signal-to-noise (S/N) ratio of the observed spectra.
The ESR measurements were also carried out to detect the radical
by a Nd:YAG laser (Quanta-Ray, GCR-130, 6 ns fwhm) at 355 nm
with the power of 30 mJ. A pulsed xenon flash lamp (Tokyo
Instruments, XF80-60, 15 J, 60 ms fwhm) was used for the probe beam,
which was detected with a Si-PIN module (Hamamatsu, C5331-SPL)
after passing through the photochemical quartz vessel (10 mm × 10
mm) and a monochromator. The output from Si-PIN module was
recorded with a digitizing oscilloscope (HP 54510B, 300 MHz). The
transient spectra were recorded using fresh solutions in each laser
excitation. All experiments were performed at 298 K.
+
-
4
intermediates produced in the photochemical reaction of MeAcrH ClO
Results and Discussion
-
2
t
-2
(1.0 × 10 M) with Bu
2
Me
2
Sn (4.5 × 10 M) in frozen MeCN at
7
7 K under irradiation of light with a high-pressure mercury lamp
Photoinduced Electron Transfer from Electron Donors
(
USH-1005D) focusing at the sample cell in the ESR cavity. The g
+
+
to MeAcrH and DeAcrH . Irradiation of the absorption band
2+
values were calibrated with an Mn marker and the hyperfine coupling
constants (hfc) were determined by computer simulation using a Calleo
ESR Version 1.2 program coded by Calleo Scientific on an Apple
Macintosh personal computer.
+
of MeAcrH results in fluorescence at λ ) 488 nm in MeCN
and CHCl3. The fluorescence lifetimes were determined by
single photon counting, as 31 and 30 ns in MeCN and CHCl3,
+
-
Quantum Yield Determinations. A standard actinometer (potassium
respectively. Due to its limited solubility of MeAcrH ClO4 ,
it cannot be used in nonpolar solvents such as benzene. To
overcome the solubility problem, 10-decylacridinium hexafluo-
2
1
ferrioxalate) was used for the quantum yield determination of the
+
+
photochemical reactions of MeAcrH and DeAcrH with R
4
Sn and
t
Bu BNAH. A square quartz cuvette (10 mm i.d.) which contained a
+
-
+
-
rophosphate (DeAcrH PF6 ) instead of MeAcrH ClO4 was
3
+
+
deaerated MeCN solution (3.0 cm ) of MeAcrH or DeAcrH (5.0 ×
+
dissolved in benzene. The fluorescence maximum of DeAcrH
498 nm) in benzene is nearly the same as that of MeAcrH in
MeCN (488 nm). The fluorescence lifetime of DeAcrH in
-
5
-4
-3
t
1
(
0
- 1.0 × 10 M) and R
4
Sn (1.0 × 10 - 2.5 M) or Bu BNAH
+
(
-
4
-3
5.0 × 10 - 1.0 × 10 M) was irradiated with monochromatized
+
(
21) (a) Hatchard, C. G.; Parker, C. A. Proc. R. Soc. London, Ser. A
1
956, 235, 518. (b) Calvert, J. C.; Pitts, J. N. In Photochemistry; Wiley:
New York, 1966; p 783.
(22) Fujita, M.; Ishida, A.; Takamuku, S.; Fukuzumi, S. J. Am. Chem.
Soc. 1996, 118, 8566.

8462 J. Am. Chem. Soc., Vol. 123, No. 35, 2001
Fukuzumi et al.
Scheme 1
reach a plateau value corresponding to the diffusion rate constant
1
0
-1 -1
(
2.0 × 10
becomes energetically more favorable (i.e., more exergonic).
Photoinduced electron transfer from an electron donor (D)
M
s ) as the photoinduced electron transfer
24
1
+
to MeAcrH * may occur as shown in Scheme 1, where k12
and k21 are diffusion and dissociation rate constants in the
encounter complex (D MeAcrH *), ket and k-et are the rate
+
+
constants of the forward electron transfer from D to MeAcrH *
and the back electron transfer to the ground state.2
2,26
The
observed quenching rate constant (kq) of photoinduced electron
transfer is given by eq 3.
k k
et 12
kq )
(3)
k + k
et
2
1
0
The dependence of ket on ∆G et for adiabatic outer-sphere
electron transfer has well been established by Marcus as given
by eq 4,
0
2
∆
G
et
1
+
(
)
kT
h
λ
λ
ket )
exp
[
-
]
(4)
(
)
( )
4
kT
where k is the Boltzmann constant, h is the Planck constant,
and λ is the reorganization energy of electron transfer. From
9
eqs 3 and 4 is derived eq 5
Figure 1. Plots of log k
vs ∆G0et for the fluorescence quenching of
q
+
-4
MeAcrH (2.0 × 10 M) by various electron donors in (a) MeCN
0
2
∆
G
+
-4
et
and (b) in CHCl
at 298 K.
3
and that of DeAcrH (2.0 × 10 M) (c) in benzene
1
+
(
)
λ
λ
k ) k Z exp
[
-
]
q
12
( )
4
/
kT
+
benzene (30 ns) is also nearly the same as that of MeAcrH in
MeCN (31 ns) (see Supporting Information, S1).
The fluorescence of the singlet excited state ( MeAcrH *)
is quenched efficiently by a variety of electron donors in MeCN
0
et
2
∆
G
1
+
(
)
1
+
λ
λ
(
k + Z exp
[
-
]
)
(5)
1
2
( )
4
kT
22,23
and CHCl3.
The quenching rate constants kq are determined
where Z ) [(kT/h)(k12/k21)] is the collision frequency which is
from the slopes of the Stern-Volmer plots and lifetime of the
singlet excited state MeAcrH *. The fluorescence of De-
AcrH * is also quenched efficiently by the same series of
electron donors in benzene and the kq values thus obtained are
summarized in Table 1.
The free energy change of photoinduced electron transfer
from electron donors to MeAcrH * (∆G et in eV) is given by
1
1
-1 -1 9
1
+
1
taken as 1 × 10
M
s . The k12 values in MeCN, CHCl3,
1
0
10
+
and benzene are taken as 2.0 × 10 , 1.2 × 10 , and 1.1 ×
1
0
-1 -1
26
1
0
M
s , respectively.
0
The dependence of kq on ∆G et for electron-transfer quenching
1
+
of MeAcrH * by electron donors in MeCN is calculated on
the basis of eq 5 using the best fit value of λ value (0.88 eV)
as shown by the solid line in Figure 1a. The kq values agree
well with the calculated values except for the kq values in the
highly exergonic region, ∆G et , -1 eV, which are significantly
larger than the calculated values (Figure 1a). The calculated
1
+
0
eq 2,
0
0
et
0
ox
0
∆
G
) e(E - E red)
(2)
0
dependence of kq on ∆G et (eq 5) predicts a decrease in the kq
0
0
where e is elementary charge, E ox and E red are the one-electron
oxidation potentials of electron donors and the one-electron
reduction potential of MeAcrH (2.32 V), respectively.
Since the E ox values of electron donors in MeCN have been
determined in this study (see Experimental Section), the ∆G et
values are determined by using eq 2 as listed in Table 1 for
comparison. Figure 1a shows a plot of log ket vs ∆G et in MeCN,
which exhibits a typical feature of an electron-transfer process;
value from a diffusion-limited value with increasing the driving
0
force of electron transfer (-∆G et) when the ket values become
1
+*
7,22
smaller than the diffusion-limited value in the Marcus inverted
0
0
region (∆G et < λ), provided that the λ value is constant in a
0
(
24) (a) Rehm, A.; Weller, A. Ber. Bunsen-Ges. Phys. Chem. 1969, 73,
34. (b) Rehm, A.; Weller, A. Isr. J. Chem. 1970, 8, 259.
25) In Scheme 1, the back electron transfer to the excited state is
neglected when the back electron transfer to the ground state is much faster
than the back electron transfer to the excited state.
8
0
(
0
the log ket value increases with a decrease in the ∆G et value to
(
26) Kavarnos, G. J. Fundamentals of Photoinduced Electron Transfer;
(23) Ohkubo, K.; Fukuzumi, S. Org. Lett. 2000, 2, 3647.
Wiley-VCH: New York, 1993.

Photoalkylation of 10-Alkylacridinium Ion
J. Am. Chem. Soc., Vol. 123, No. 35, 2001 8463
0
+
Table 2. One-Electron Reduction Potentials E (MeAcrR /
•
+
-
MeAcrR ) of MeAcrR ClO
4
Determined by the Cyclic
Voltammograms in Deaerated Various Solvents at 298 K
0
+
•
E (MeAcrR /MeAcrR ),
V vs SCE
solvent
MeCNa
ꢀ
R ) PhCH
2
R ) Ph
R ) Pri
-0.63c
c
c
37.5
7.8
4.8
-0.50
-0.55
a
c
c
c
CH
CHCl
2
Cl
3
b
2
-0.44
-0.49
-0.54
a
c
c
c
-0.44
-0.47
-0.52
d
d
-0.42d
C
H
6 6
2.3
-0.42
-0.45
a
b
c
Containing 0.1 M TBAP. Containing 0.5 M THAP. Sweep rate
-
1
d
-1
is 50 mV s
Figure S2.
.
Sweep rate is 10 mV s . See Supporting Information,
9
series of electron-transfer reactions. The absence of a Marcus
inverted region has well been recognized in forward photo-
induced electron-transfer reactions.24 In the case of back
electron-transfer reactions, however, the observation of the
2
7-30
Marcus-inverted region has been well established.
The
absence of an inverted region in forward photoinduced electron
0
transfer reactions in the highly exergonic region (∆G et < -λ)
can be explained by an increase in the λ value from the value
for a contact radical ion pair (CRIP) to the value for a solvent
separated radical ion pair (SSRIP) which has a larger distance
•
3
0
Figure 2. ESR spectrum of MeAcrPh in deaerated MeCN at 298 K
between the radical ions in the highly exergonic region. The
existence of low-energy excited states of radical cations may
and the computer simulation spectrum.
3
1
also contribute to the nonexistence of an inverted region.
benzene is also calculated on the basis of eq 5 as shown in
Figure 1b and Figure 1c, respectively. The ∆G et values CHCl3
and benzene are evaluated from the values in MeCN [∆G et
0
The solvent independent ∆G et value is confirmed by deter-
mination of the E ox values of electron donors and the E red
0
0
0
0
values of acridinium ions in solvents with different polarity.
(MeCN)] by using eq 6,
0
The E ox values of alkylbenzene derivatives have been deter-
mined in MeCN and CH2Cl2 as listed in Table 1 (see the values
0
0
0
∆G ) ∆G (MeCN) + C
(6)
in parentheses). The E red values of 10-methyl-9-substituted
et
et
+
acridinium ions instead of MeAcrH have also been determined
in MeCN, CH2Cl2, CHCl3 and benzene, since the reversible
redox couples can be obtained for the 9-substituted derivatives
in contrast to the case of MeAcrH (see Supporting Information,
where the constant value C, which is dependent on the solvent,
is determined by fitting the data to eq 5. The best fit values of
λ and C are determined as λ ) 0.62 eV, C ) 0.16 eV in CHCl3
and λ ) 0.53 eV, C ) 0.27 eV in benzene (see Supporting
+
0
0
S2). The E red values are also listed in Table 2. The E ox values
in a less polar solvent (CH2Cl2) are shifted to the positive
direction by about 0.1 V due to the less solvation of the radical
cations as compared to that in MeCN (Table 1). Similar positive
3
2,33
Information, S3).
The λ value decreases from the value in
MeCN (0.88 eV) with decreasing the solvent polarity due to
0
the smaller solvent reorganization energy, whereas the ∆G et
0
shifts are observed for the E red values of acridinium ions in
CH2Cl2 as compared to the E red values in MeCN, and the larger
value increases, since the difference in solvation between
aromatic donor radical cations and acridinium ion decreases with
decreasing the solvent polarity.
0
shifts are observed in CHCl3 and benzene (Table 2). Thus, the
0
0
∆
G et values obtained as the difference between the E ox and
Electron-Transfer Self-Exchange Reactions between Me-
0
•
+
E red values (eq 2) become rather solvent-independent because
AcrPh and MeAcrPh in Different Solvents. The decrease
in the solvent reorganization energy of electron transfer with
decreasing the solvent polarity is examined by determining the
rate constants of electron-transfer self-exchange reactions
of the cancellation of the solvation. However, the cancellation
0
is not complete, since the positive shifts in the E ox values are
0
slightly larger than the shifts in the E red values.
0
The dependence of kq on ∆G et for electron-transfer quenching
(32) Since the redox potentials have been determined in the presence of
1
+
of R′AcrH * by aromatic electron donors in CHCl3 and
the high concentration of electrolyte (Tables 1 and 3), the positive shifts of
the redox potentials in less polar solvents may be larger in the absence of
electrolyte as compared to those in its absence. This may be the reason the
estimated C value is larger than that expected from the difference in the
redox potentials determined in the presence of the high concentration of
electrolyte.
(27) (a) Closs, G. L.; Miller, J. R. Science 1988, 240, 440. (b) Miller, J.
R.; Calcaterra, L. T.; Closs, G. L. J. Am. Chem. Soc. 1984, 106, 3047. (c)
Asahi, T.; Mataga, N. J. Phys. Chem. 1989, 93, 6575. (d) Gould, I. R.;
Ege, D.; Moser, J. E.; Farid, S. J. Am. Chem. Soc. 1990, 112, 4290. (e)
Gould, I. R.; Farid, S. Acc. Chem. Res. 1996, 29, 522.
(33) The best fit is obtained by linear plots between (∆G ) vs ∆G0et
q 1/2
(28) (a) McLendon, G. Acc. Chem. Res. 1988, 21, 160. (b) Winkler, J.
(see Supporting Information, S3) which is derived from eq 5 as given by
q 1/2
1/2
0
1/2
q
-1
-1
R.; Gray, H. B. Chem. ReV. 1992, 92, 369. (c) McLendon, G.; Hake, R.
Chem. ReV. 1992, 92, 481.
(∆G ) ) λ0 /2 + ∆G et/(2λ ), where ∆G ) kT ln[Z(kq - k12 ) and
0
∆G et ) ∆G et(MeCN) + C. The λ and C values were determined from the
(
29) (a) Rau, H.; Frank, R.; Greiner, G. J. Phys. Chem. 1986, 90, 2476.
slope and the intercept of the least-squares linear plot. The correlation
coefficients F of the linear plots for the fluorescence quenching in MeCN,
CHCl3, and benzene are 0.97, 0.96, and 0.97, respectively. The experimental
errors in the λ and C values in MeCN, CHCl3, and benzene are estimated
using a standard graphical method as (2, (12, and (11%, respectively.
For a standard graphical method of determining the limit of error in an
intercept and slope of a linear plot, see: Shoemaker, D. P.; Garland, C.
W.; Steinfeld, J. I.; Nibler, J. W. Experiments in Physical Chemsitry;
McGraw-Hill Book Company: New York, 1981; p 51.
(
1
b) Stevens, B.; Biver, C. J., III; McKeithan, D. N. Chem. Phys. Lett. 1991,
87, 590. (c) Kikuchi, K.; Takahashi, Y.; Katagiri, T.; Niwa, T.; Hoshi,
M.; Miyashi, T. Chem. Phys. Lett. 1991, 180, 403.
30) Mataga, N.; Miyasaka, H., In Electron Transfer from Isolated
(
Molecules to Biomolecules, Part 2; Jortner, J., Bixon, M., Eds.; Wiley: New
York, 1999; p 431.
(31) Mataga, N.; Konda, Y.; Asahi, T.; Miyasaka, H.; Okada, T.;
Kakitani, T. Chem. Phys. 1988, 127, 239.

8464 J. Am. Chem. Soc., Vol. 123, No. 35, 2001
Fukuzumi et al.
+
of the linear correlations between ∆Hmsl and [MeAcrPh ] within
experimental error of (5% (Figure 3). The kex values in other
solvents were determined in a similar manner and are listed in
Table 3. The activation parameters were also determined by
the Arrhenius plots (see Supporting Information, S4) as listed
q
in Table 3. The small ∆S values in Table 3 are consistent with
the outer-sphere electron-transfer reactions. The reorganization
energies (λ) of the electron-transfer reactions are obtained from
the kex values using eq 9
-
1
-1
-1
λ
kT
[
(k ) - (kdiff) ] ) Z exp
(9)
ex
(
)
4
where kdiff is the diffusion rate constant which corresponds to
1
0
-1 -1
k12 in Scheme 1 (kdiff ) 2.0 × 10
M
s
in MeCN, 2.0 ×
1
0
-1 -1
10
-1 -1
1
0
M
s
in CH2Cl2, 1.2 × 10
M
s
in CHCl3 and 1.1
1
0
-1 -1
26
+
× 10
M
s
in benzene, respectively). The λ values thus
determined are also listed in Table 3. It is clearly shown that
the λ value decreases with decreasing the solvent polarity.
Photoalkylation of MeAcrH and DeAcrH with R4Sn.
Irradiation of the absorption band of MeAcrH ClO4 in a
deaerated MeCN solution containing R4Sn with a xenon lamp
for 1 h gave the alkyl adduct (MeAcrHR) as shown in eq 10
Figure 3. Plots of ∆Hmsl vs [MeAcrPh ] for the ESR spectra of
•
MeAcrPh in deaerated MeCN at various temperatures.
36
Table 3. Electron-Transfer Self-Exchange Rate Constants (kex),qthe
Reorganization Energies (λ), and the Activation Parameters (∆H
+
+
+
-
q
•
+
and ∆S ) for the MeAcrPh /MeAcrPh System
q
q
∆
H ,
∆S ,
ex,a M-1 s-1
-1
cal K-1 mol-1
solvent
MeCN
k
λ, eV
kcal mol
3
7
(
see Experimental Section). The quantum yields (Φ) of the
9
3.1 × 10
0.34
0.28
0.27
0.21
2.2
1.0
1.1
1.9
3.7
1.3
1.3
0.5
9
2
CH Cl
2
4.1 × 10
9
CHCl
3
4.2 × 10
9
C H
6 6
5.9 × 10
a
The experimental error is ( 5%.
+
between 9-phenyl-10-methylacridinium ion (MeAcrPh ) and the
•
corresponding one-electron reduced radical (MeAcrPh ) in
photoalkylation of MeAcrH+ and DeAcrH+ with R4Sn were
•
solvents with different polarity. The MeAcrPh radical was
determined from the spectral change under irradiation of
monochromatized light of λmax ) 398 nm (see Experimental
Section). The Φ value increases with an increase in the
concentration of R4Sn [R4Sn] to approach a limiting value (Φ∞)
in accordance with eq 11
+
produced by the electron-transfer reduction of MeAcrPh by
tetramethylsemiquinone radical anion (eq 7). ESR spectra for
Φ K [R Sn]
∞
obs
4
Φ )
(11)
1
+ Kobs[R Sn]
4
•
(see Supporting Information, S5). Equation 11 is rewritten by
MeAcrPh were persistent for several hours in deaerated MeCN.
-
1
eq 12 which predicts a linear correlation between Φ and
The hyperfine splitting constants and the maximum slope line
widths (∆Hmsl) were determined from a computer simulation
of the ESR spectra as shown in Figure 2. The ∆Hmsl value thus
determined increases linearly with an increase in the concentra-
-
1
[
R4Sn] .
Φ- ) Φ∞ [1 + (Kobs[R Sn]) ]
1
-1
-1
(12)
4
+
tion of MeAcrPh in MeCN as shown in Figure 3. Such line
From the slopes and intercepts of linear plots are obtained the
Φ∞ and Kobs values (see Supporting Information, S6). The Kobs
values can be converted to the corresponding rate constants (kobs)
using the relation Kobs ) kobsτ provided that the excited state of
width variations of the ESR spectra can be used to investigate
the rate processes involving the radical species.34 The rate
constants (kex) of the electron-transfer self-exchange reactions
+
•
between MeAcrPh and MeAcrPh were determined using eq
+
+
MeAcrH or DeAcrH involved in the photochemical reaction
is singlet. The kobs and Φ∞ values thus determined are listed in
Table 4.
8
,
7
0
1
.57 × 10 (∆H - ∆H msl)
msl
1
+
1
+
The fluorescence of MeAcrH * and DeAcrH * is quenched
efficiently by electron transfer from R4Sn in MeCN and benzene,
respectively. The quenching rate constants kq are also listed in
Table 4, where the kobs values derived from the dependence of
k )
(8)
ex
+
(
1 - P )[MeAcrPh ]
i
0
where ∆Hmsl and ∆H msl are the maximum slope line width of
the ESR spectra in the presence and absence of MeAcrPh ,
respectively, and Pi is a statistical factor which can be taken as
nearly zero.
+
(
36) The difference in the λ values for the electron-transfer self-exchange
of MeAcrPh+/MeAcrPh• between MeCN and benzene (0.13 eV) is smaller
3
4,35
The kex values are determined from the slope
than the corresponding difference for the electron-transfer reactions from
aromatic electron donors to the singlet excited state of acridinium ion (0.27
eV). This is consistent with the smaller solvation of acridinium ion as
compared to aromatic donor radical cations due to the more delocalized
charge.
(
34) (a) Chang, R. J. Chem. Educ. 1970, 47, 563. (b) Cheng, K. S.; Hirota,
N. In InVestigation of Rates and Mechanisms of Reactions; Hammes, G.
G., Ed.; Wiley-Interscience: New York, 1974; Vol. VI, p 565.
(
35) Fukuzumi, S.; Nakanishi, I.; Suenobu, T.; Kadish, K. M. J. Am.
(37) Fukuzumi, S.; Kuroda, S.; Tanaka, T. J. Chem. Soc., Chem.
Commun. 1986, 1553.
Chem. Soc. 1999, 121, 3468.

Photoalkylation of 10-Alkylacridinium Ion
J. Am. Chem. Soc., Vol. 123, No. 35, 2001 8465
Table 4. Rate Constants (k
q
) for the Fluorescence Quenching of
Sn, the Observed Rate Constants (kobs) Derived from
the Dependence of the Quantum Yields on [R Sn] in the
Sn in MeCN and Benzene at
+
R′AcrH by R
4
4
+
Photoalkylation of R′AcrH with R
4
0
2
98 K, and the Driving Force of Back Electron Transfer (-∆G bet
)
•
•+
from R′AcrH to R
4
Sn in MeCN
a
obs,b
-1 -1
-∆G0bet
,
k
q
,
k
M
-
1
s-1
b
eVc
R
4
Sn
solvent
benzene 1.1 × 10
MeCN
3.3 × 10
benzene 2.4 × 10
M
s
Φ
∞
8
8
9
1
9
9
8
Me
4
Sn
1.1 × 10
4.2 × 10
2.1 × 10
2.4 × 10
5.0 × 10
8.0 × 10
1.7 × 10
1.4 × 10
0.0088
0.00042
0.028
0.007
0.027
0.0028
0.0055
0.021
8
2.11
1.67
1.43
1.29
9
Et
4
Sn
0
9
MeCN
benzene 5.3 × 10
MeCN
7.7 × 10
Sn benzene 1.1 × 10
2.0 × 10
Pri
9
4
Sn
9
But
Me
2 2
10
10
10
1
0
MeCN
1.1 × 10
a
b
The experimental error is (5%. The experimental error is (10%.
The -∆G bet values are obtained using eq 2 where ∆G et is replaced
c
0
0
0
by ∆G bet
.
Scheme 2
•
Figure 4. (a) Transient absorption spectra of DeAcrH generated in
-
3
+
-5
electron transfer from Et
4
Sn (8.2 × 10 M) to DeAcrH (8.0 × 10
M) in MeCN at 150 µs after laser excitation (355 nm) at 298 K. (b)
•
Transient absorption spectra of DeAcrH generated in electron transfer
-
1
+
-4
from Me
4
Sn (1.0 × 10 M) to DeAcrH (5.0 × 10 M) in benzene
at 150 µs after laser excitation (355 nm) at 298 K.
R4Sn•+ is known to be cleaved to give the alkyl radical,39-42
which is coupled within the cage to yield the adduct selectively
Φ on [R4Sn] in both MeCN and benzene agree well with the
corresponding kq values of the fluorescence quenching. Such
an agreement indicates that the photochemical reaction proceeds
via photoinduced electron transfer from R4Sn to the singlet
•
without dimerization of free R′AcrH radicals escaped from the
43
cage, in competition with the back electron transfer to the
ground state (kbet). In such a case, the limiting quantum yield
Φ∞ is determined by the competition between the cleavage of
1
+
1
+
excited states ( MeAcrH * and DeAcrH *).
•+
Sn-C bond of R4Sn in the radical cation/radical pair (kc) and
•
The formation of DeAcrH in photoinduced electron transfer
•
•+
the back electron transfer from R′AcrH to R4Sn (kbet) as given
1
+
from Et4Sn to DeAcrH * in MeCN is confirmed by the laser
by eq 13.
+
flash photolysis of the DeAcrH -Et4Sn system as shown in
Figure 4a. The laser excitation (355 nm from a Nd:YAG laser)
kc
+
-5
of DeAcrH (8.0 × 10 M) in deaerated MeCN and benzene
Φ )
(13)
∞
-
3
kbet + kc
solutions containing Et4Sn (8.2 × 10 M) gives transient
•
absorption of DeAcrH which is essentially the same as that of
t
t
t
•+
In the case of Bu 2Me2Sn, the Sn-Bu bond of Bu 2Me2Sn
is cleaved selectively as compared to the Sn-Me bond,40 to
•
MeAcrH (a broad absorption band between 450 and 540
2
2,38
•
nm).
The formation of DeAcrH in photoinduced electron
•+
transfer from the weakest electron donor employed in this study,
(39) The transient absorption spectrum of Et4Sn is not observed because
•
+
+
of the facile Sn-C bond cleavage of Et4Sn ; see: ref 40.
that is Me4Sn to DeAcrH has also been confirmed in benzene
(
40) Fukuzumi, S.; Mochida, K.; Kochi, J. K. J. Am. Chem. Soc. 1979,
+
as shown in Figure 4b. Thus, the photoalkylation of R′AcrH
1
01, 5961.
(
R ) Me and De) with R4Sn proceeds via photoinduced electron
(41) (a) Walther, B. W.; Williams, F.; Lau, W.; Kochi, J. K. Organo-
metallics 1983, 2, 688. (b) Symons, M. C. R. Chem. Soc. ReV. 1984, 13,
1
+
transfer from R4Sn to R′AcrH * as shown in Scheme 2.
3
93.
(
The photoalkylation is initiated by photoinduced electron
42) Fukuzumi, S.; Kochi, J. K. J. Org. Chem. 1980, 45, 2654.
(43) No significant amount of the dimer was observed in the present
case, indicating that the radical coupling in the cage is highly efficient as
compared to the escape of radicals from the cage. In the case of the
photochemical reaction of diphenylmethane with MeAcrH , however, the
dimers, (MeAcrH)2 and (Ph2CH)2 have been obtained in addition to the
adduct; see: ref 22.
1
+
transfer (ket) from R4Sn to R′AcrH * to give the radical cation-
•
+
•
acridinyl radical pair (R4Sn R′AcrH ). The Sn-C bond of
+
(
38) (a) Peters, K. S.; Pang, E.; Rudzki, J. J. Am. Chem. Soc. 1982, 104,
535. (b) Poulos, A. T.; Hammond, G. S.; Burton, M. E. Photochem.
Photobiol. 1981, 34, 169.
5

8466 J. Am. Chem. Soc., Vol. 123, No. 35, 2001
Fukuzumi et al.
Table 5. Rate Constants (k
q
) for the Fluorescence Quenching of
+
t
R′AcrH by Bu BNAH, the Observed Rate Constants (kobs), and the
) Derived from the Dependence of the
Limiting Quantum Yields (Φ
∞
t
+
Quantum Yields on [Bu BNAH] in the Photoalkylation of R′AcrH
t
with Bu BNAH
1
obs, M-1 s-1
solvent
MeCN
k
q
, M- s-1
k
Φ
∞
1
1
1
0
0
0
10
2.1 × 10
2.0 × 10
2.3 × 10
2.3 × 10
0.04
0.12
0.81
10
CHCl
3
1.9 × 10
10
C H
6 6
2.1 × 10
competition between the cleavage of Sn-C bond of R4Sn•+ in
the radical cation/radical pair (kc) and the back electron transfer
•
Figure 5. ESR spectrum of acridinyl radical (MeAcrH ) and tert-butyl
t•
t
•
•+
radical (Bu ) generated in photoinduced electron transfer from Bu
2
-
from R′AcrH to R4Sn (kbet). The solvent effects on the
-
2
+
•+
Me
1
2
Sn (4.5 × 10 M) to the singlet excited state of MeAcrH (1.0 ×
cleavage rates of Sn-C bond of R4Sn have been studied
-
2
0 M) observed in frozen MeCN at 77 K under irradiation of UV
previously, and it has been shown that the kc values are rather
light by using high-pressure mercury lamp.
42,47
independent of solvent polarity.
Thus, the Φ∞ value is mainly
•
determined by the back electron-transfer rate from R′AcrH to
R4Sn (kbet); the smaller the kbet value, the larger is the Φ∞
value. The kbet value varies depending on ∆G bet of back
•
+
0
electron-transfer according to the Marcus equation (eq 4) as
0
shown in Figure 6 where the dependence of log kbet vs ∆G bet
9
in MeCN and benzene is drawn schematically.
0
The -∆G bet value at which the maximum kbet value is
obtained corresponds to the reorganization energy (λ) for the
back electron transfer. The λ value for the electron-transfer
-
1
oxidation of R4Sn is known to be large (λ ) 41 kcal mol
)
1
2,13
1
.78 eV).
This is confirmed by analyzing the kq values of
1
+
photoinduced electron transfer from R4Sn to MeAcrH * in
0
Table 4. The λ value can be obtained from the kq and ∆G et
values using eq 5. The λ values thus evaluated are λ ) 1.95
4
8
and 1.82 eV for Me4Sn and Et4Sn in MeCN, respectively.
Figure 6. Dependence of log kbet on ∆G0bet in MeCN (solid line) and
1
2,13
These values agree with the reported large value.
In such a
benzene (broken line). The lines are drawn schematically based on eq
0
0
case, the -∆G et value is changed from the value in the Marcus
inverted region in the case of Me4Sn to the normal region up
to 1.2 eV which is significantly smaller than the λ value in the
4
4
. The ∆G bet value at which kbet is maximum corresponds to -λ (eq
).
t
t
case of Bu 2Me2Sn. According to the Marcus equation (eq 4),
give MeAcrHBu rather than MeAcrHMe (see Experimental
Section). This is confirmed by the ESR spectrum observed in
the kbet value decreases with decreasing the λ value in the
inverted region, whereas the kbet value increases in the normal
region (Figure 6). The λ value is expected to decrease with
decreasing the solvent polarity (Table 3). In fact, the λ values
evaluated from the kq and ∆G et values in benzene using eq 5
λ ) 1.52 and 1.30 eV for Me4Sn and Et4Sn, respectively) are
significantly smaller than those in MeCN (vide supra). This
is the reason the Φ∞ value of Me4Sn in benzene is much larger
than the corresponding value in MeCN, whereas this is reversed
t
photoinduced electron transfer from Bu 2Me2Sn to the singlet
+
excited state of MeAcrH observed in frozen MeCN at 77 K
under photoirradiation with a high-pressure mercury lamp as
shown in Figure 5. The 10-line isotropic spectrum is ascribed
0
(
to the hyperfine coupling of an unpaired electron with the 9
49
t• 42,44
equivalent protons of Bu .
The broad signal at the center
•
45
can be assigned due to MeAcrH .
Solvent Dependence of the Limiting Quantum Yield. The
Φ∞ value varies drastically depending on the alkyl group (R)
t
for Bu 2Me2Sn, the Φ∞ value of which in MeCN is 38 times
-
4
larger than the corresponding value in benzene (Table 4). Thus,
whether the back electron transfer is in the Marcus inverted
region or in the normal region determines whether the Φ∞ value
increases or decreases with decreasing the solvent polarity. In
such a case, the solvent dependence of Φ∞ would be reversed
if the photoinduced electron-transfer system with the much
of R4Sn and solvents from the smallest value (4.2 × 10 ) for
-
2
i
Me4Sn in MeCN to the largest value (2.8 × 10 ) for Pr 4Sn in
MeCN as shown in Table 4. It is interesting to note that the Φ∞
value of Me4Sn in benzene is 21 times larger than the
corresponding value in MeCN. However, this is reversed for
t
Bu 2Me2Sn; the Φ∞ value in a polar solvent (MeCN) is 38 times
+
smaller λ value than the R4Sn-R′AcrH system is chosen (Vide
larger than the corresponding value in a nonpolar solvent
infra).
(
benzene). The ratio of the Φ∞ value in benzene to that in MeCN
t
changes systematically from Me4Sn to Bu 2Me2Sn with decreas-
ing the driving force (-∆G bet) of back electron transfer from
R′AcrH to R4Sn . The driving force is obtained from the
0
+
(
46) The E red value of DeAcrH is assumed to be the same as that of
0
+
MeAcrH . See ref 8.
•
•+
(47) The counterion may also affect the cleavage rates of Sn-C bond
•+
-
of R4Sn . In this study, however, the counterion remains the same (ClO4 ).
0
0
difference between the E ox value of R4Sn and E red value of
•+
The selectivities of the cleavage of Sn-C bond of RxR′4-xSn (x ) 1-3)
are known to be the same irrespective of the difference in the counterion
+
46
0
R′AcrH (eq 2), and the -∆G bet values are listed in Table
. According to eq 13, the Φ∞ value is determined by the
-
3- 42
4
between I and IrCl6
.
The selectivities are also unaffected by changes
in the solvent.42
i
(
44) Fessenden, R. W.; Schuler, R. H. J. Chem. Phys. 1963, 39, 2147.
b) Ascough, R. B.; Thomson, C. Trans. Faraday Soc. 1962, 58, 1477. (c)
Krusic, P. J.; Kochi, J. K. J. Am. Chem. Soc. 1968, 90, 7155.
45) (a) Fukuzumi, S.; Fujita, M.; Noura, S.; Ohkubo, K.; Suenobu, T.;
Araki, Y.; Ito, O. J. Phys. Chem. A 2001, 105, 1857.
(48) The λ value of Pr 4Sn is also determined as 1.82 eV. However, the
evaluation of the λ value from the kq value which is close to the diffusion-
limit involves a large experimental error.
(
0
(
(49) The ∆G et values in benzene are evaluated from those in MeCN by
using eq 6.

Photoalkylation of 10-Alkylacridinium Ion
J. Am. Chem. Soc., Vol. 123, No. 35, 2001 8467
•
t•
Scheme 3
or (b) But+ and BNA , the formation of Bu in the one-electron
t
oxidation of Bu BNAH has been confirmed by applying a rapid
mixing flow electron spin resonance (ESR) technique. Thus,
cleavage of Bu BNAH gives Bu that is coupled immediately
with MeAcrH to yield the adduct (MeAcrHBu ) in competition
with the back electron transfer from MeAcrH to Bu BNAH
Scheme 3).
By applying the steady-state approximation to the reactive
51
t
•+
t•
•
t
•
t
•+
(
t
species in Scheme 2, the dependence of Φ on [Bu BNAH] can
be derived as given by eq 15, which agrees with the observed
t
dependence of Φ on [Bu BNAH].
kc
t
k τ[Bu BNAH]
et
[
]
kc + k
bet
Φ )
(15)
t
1
+ k τ[Bu BNAH]
et
The limiting quantum yields Φ∞ corresponds to kc/(kc + kbet).
In such a case the Φ∞ value is determined by the competition
between the C(4)-C bond cleavage (kc) and the back electron
transfer (kbet). Since the reorganization energy (λ) of photo-
induced electron transfer reactions of BNAH and derivatives
1
6
in MeCN is relatively small (λ ) 0.5 eV), the back electron
•
t
•+
transfer from MeAcrH to Bu BNAH should be in the Marcus
0
inverted region (-∆G bet ) 1.1 eV . 0.5 eV) in contrast with
t
0
the case of Bu 2Me2Sn (-∆G bet ) 1.1 eV , 1.8 eV). Thus,
the solvent reorganization energy should decrease with decreas-
ing the solvent polarity, leading to the decrease in the back
electron-transfer rate in the Marcus inverted region as shown
∞
t
+
Photoalkylation of Bu BNAH with MeAcrH and De-
+
+
-
AcrH . Irradiation of the absorption band of MeAcrH ClO4
in a deaerated MeCN solution containing Bu BNAH with a
xenon lamp for 1 h gave an adduct (MeAcrHBu ) as shown in
t
t
in Figure 6 and thereby an increase in the Φ value (Table 5).
eq 14 (see Experimental Section). The same type of reaction
Summary and Conclusions
The present study has shown that the reactivity of charge-
shift type of photoinduced electron transfer reactions which lead
to stable products can be finely controlled by the solvent
polarity. The photoinduced electron transfer from electron
1
+
donors to R′AcrH * occurs efficiently even in a nonpolar
occurs when MeAcrH+ is replaced by DeAcrH+ in benzene.
solvent as well as in a polar solvent to yield the stable adducts
between R′AcrH and radicals produced by the bond-cleavage
•
+
Thus, the product derived from R′AcrH is the same as obtained
t
of the radical cations to yield the stable adducts. The limiting
quantum yield increases with decreasing the solvent polarity
in the photoalkylation with Bu 2Me2Sn. The quantum yields (Φ)
+
t
of the photoalkylation of R′AcrH with Bu BNAH were
determined from the spectral change under irradiation of
monochromatized light of λmax ) 398 nm (see Experimental
Section). The Φ∞ and kobs values are determined as the case of
the photoalkylation with R4Sn, and they are listed in Table 5
•
when the back electron transfer from R′AcrH to the radical
cations is in the Marcus inverted region, whereas the yield
decreases when the back electron transfer is in the Marcus
normal region.
(
see Supporting Information, S7). In contrast with the case of
Acknowledgment. This work was partially supported by a
Grant-in-Aid for Scientific Research Priority Area (No. 11228205)
from the Ministry of Education, Science, Culture and Sports,
Japan.
t
Bu 2Me2Sn (Table 4), the Φ∞ value increases with decreasing
the solvent polarity and the Φ∞ value in benzene is 20 times
larger than that in MeCN.
The kq values are also determined independently by the
t
fluorescence quenching via electron transfer from Bu BNAH
to MeAcrH * and DeAcrH *, and they are also listed in Table
5
agreement indicates that the photoalkylation of R′AcrH with
Bu BNAH proceeds via photoinduced electron transfer from Bu -
BNAH to R′AcrH * as shown in Scheme 3. The photoinduced
electron transfer from Bu BNAH to R′AcrH * produces the
radical cation/radical pair (Bu BNAH R′AcrH ). Sav e´ ant et
al. have reported that the electrochemical oxidation of Bu BNAH
results in the selective C(4)-C bond cleavage of Bu BNAH .
Although there are two possible modes of the carbon-carbon
bond cleavage in such reactions to generate (a) Bu and BNA
Supporting Information Available: Fluorescence decay
1
+
1
+
+
curves of R′AcrH (Figure S1), cyclic voltammograms of
in which the kq values agree with the kobs values. Such an
i+
-
MeAcrPr ClO4 in various solvents (Figure S2), plots of
(
+
q 1/2
0
+
∆G ) vs ∆G et for the fluorescence quenching of R′AcrH
t
t
by various electron donors in different solvents (Figure S3),
1
+
Arrhenius plots of kex for the electron-transfer self-exchange
t
1
+
•
+
reaction of MeAcrPh /MeAcrPh system (Figure S4), depen-
t
•+
•
+
dence of the quantum yields for photoalkylation of R′AcrH
with R4Sn (Figure S5), plots of Φ vs [R4Sn] (Figure S6),
plots of Φ vs [Bu BNAH] (Figure S7) (PDF). This material
t
-1
-1
t
•+ 50
-1
t
-1
is available free of charge via the Internet at http://pubs.acs.org.
t•
+
JA004311L
(50) Anne, A.; Moiroux, J.; Sav e´ ant, J.-M. J. Am. Chem. Soc. 1993, 115,
1
0224.
(51) Takada, N.; Itoh, S.; Fukuzumi, S. Chem. Lett. 1996, 1103.