Generation of Polyphenylene Radical Cations and Their Cosensitization Ability in the 9,10-Dicyanoanthracene-Sensitized Photochemical Chain Reactions of 1,2-Bis(4-methoxyphenyl)cyclopropane

Article abstract of DOI:10.1021/jo971649y

Cosensitization effects of polyphenylene compounds (PP) such as biphenyl (BP), terphenyls (o-, m-, p-TP), and phenanthrene (Phen) in photoinduced electron-transfer reactions were examined. The 9,10-dicyanoanthracene (DCA)-sensitized cis-trans photoisomerization of 1,2-bis(4-methoxyphenyl)-cyclopropane (CP), which proceeds in a chain reaction via free radical cation of CP (CP^.+) as a chain carrier, was accelerated by adding PP, particularly by TP. A similar accelerating effect was observed in the DCA-sensitized photooxygenation of CP as another example. BP and TP were more stable under the oxygenation condition than phenanthrene and naphthalene, which also accelerate the photooxygenation CP. CP^.+ is generated by the direct electron transfer from CP to the excited singlet state of DCA (¹DCA*) and also by the secondary electron transfer from CP to PP^.+, which is generated by the primary electron transfer from PP to ¹DCA*. The laser flash photolysis study revealed that the quantum yield for the formation of free CP^.+ in the direct electron transfer from CP to ¹DCA* (Φ_CP.+ ≈ 0.1) was smaller than that in the presence of PP. This is due to the high yield of free PP^.+ generation by the primary electron transfer and the efficient secondary electron transfer from CP to PP^.+. The secondary electron transfers were found to take place in nearly diffusion-controlled rates (0.9-1.5 × 10¹⁰ M^-1 s^-1). The high yield of PP^.+ as free radical ions does not seem to be the sole factor of the cosensitization of PP for the DCA-sensitized photoreactions of CP. The ratio of the quantum yields of the photoreactions to that of the initial CP^.+ formation (turnover) also increased by the addition of PP from 3 (isomerization) and 15 (oxygenation) to 32-90 for both reactions. The second-order rate constant for the decay of CP^.+ in aerated acetonitrile was decreased by a factor of 0.5-0.8 by the addition of PP. We concluded that the cosensitization effect in the photoreaction involves a π-complex formation between CP^.+ and PP assisting the chain reaction as well as initial CP^.+ formation.

Full text of DOI:10.1021/jo971649y

3
204
J . Org. Chem. 1998, 63, 3204-3212
Gen er a tion of P olyp h en ylen e Ra d ica l Ca tion s a n d Th eir
Cosen sitiza tion Ability in th e 9,10-Dicya n oa n th r a cen e-Sen sitized
P h otoch em ica l Ch a in Rea ction s of
1
,2-Bis(4-m eth oxyp h en yl)cyclop r op a n e
,
†
,‡,§
‡
‡
Toshiyuki Tamai,* Nobuyuki Ichinose,* Tomoko Tanaka, Tsuneo Sasuga,
†
,⊥
Isao Hashida, and Kazuhiko Mizuno*
Osaka Municipal Technical Research Institute, 1-6-50 Morinomiya, J oto-ku, Osaka 536-8533, J apan,
Kansai Research Establishment, J apan Atomic Energy Research Institute, 25-1 Mii-minamimachi,
Neyagawa, Osaka 572-0019, J apan, and Department of Applied Chemistry, Osaka Prefecture University,
1
-1 Gakuen-cho, Sakai, Osaka 599-8531, J apan
Received September 4, 1997
Cosensitization effects of polyphenylene compounds (PP) such as biphenyl (BP), terphenyls (o-, m-,
p-TP), and phenanthrene (Phen) in photoinduced electron-transfer reactions were examined. The
9
,10-dicyanoanthracene (DCA)-sensitized cis-trans photoisomerization of 1,2-bis(4-methoxyphenyl)-
•
+
cyclopropane (CP), which proceeds in a chain reaction via free radical cation of CP (CP ) as a
chain carrier, was accelerated by adding PP, particularly by TP. A similar accelerating effect was
observed in the DCA-sensitized photooxygenation of CP as another example. BP and TP were
more stable under the oxygenation condition than phenanthrene and naphthalene, which also
accelerate the photooxygenation CP. CP is generated by the direct electron transfer from CP to
the excited singlet state of DCA ( DCA*) and also by the secondary electron transfer from CP to
•
+
1
•
+
1
PP , which is generated by the primary electron transfer from PP to DCA*. The laser flash
photolysis study revealed that the quantum yield for the formation of free CP in the direct electron
•+
1
transfer from CP to DCA* (ΦCP•+ ≈ 0.1) was smaller than that in the presence of PP. This is due
•+
to the high yield of free PP generation by the primary electron transfer and the efficient secondary
•
+
electron transfer from CP to PP . The secondary electron transfers were found to take place in
1
0
-1 -1
•+
nearly diffusion-controlled rates (0.9-1.5 × 10
M
s ). The high yield of PP as free radical
ions does not seem to be the sole factor of the cosensitization of PP for the DCA-sensitized
photoreactions of CP. The ratio of the quantum yields of the photoreactions to that of the initial
•
+
CP formation (turnover) also increased by the addition of PP from 3 (isomerization) and 15
oxygenation) to 32-90 for both reactions. The second-order rate constant for the decay of CP•+ in
aerated acetonitrile was decreased by a factor of 0.5-0.8 by the addition of PP. We concluded that
(
•
+
the cosensitization effect in the photoreaction involves a π-complex formation between CP and
PP assisting the chain reaction as well as initial CP formation.
•
+
In tr od u ction
Sch em e 1
The photoinduced electron-transfer reaction has re-
ceived considerable attention from synthetic and mecha-
1
nistic viewpoints for these two decades. In the primary
process of the photoinduced electron-transfer reactions
(Scheme 1), diffusion-controlled quenching of an excited
state of an electron acceptor (A) by a donor (D) or vice
versa gives a solvent-separated radical ion pair (SSRIP)
•
-
•+
(
s
(A /D ) ), which undergoes further chemical processes
mainly after dissociation into free ions or deactivation
by back-electron transfer (BET). The BET process, which
has attracted much interest from the theoretical and
experimental chemists because its high exothermicity is
favorable for the study of the “inverted region” in the
conversion. A direct electron transfer from an electron-
donating substrate molecule (S) to an excited state of an
electron acceptor molecule (A*) often gives less amount
of free S because of a fast BET process suppressing the
separation of SSRIP. It has been reported that several
organic and inorganic additives induce the charge sepa-
Marcus theory,2,3 often dominates resulting in the inef-
•
+
ficient free ion formation. This deactivation process is
undesirable for the purpose of photochemical energy
†
ration against the back-electron transfer within a radical
Osaka Municipal Technical Research Institute.
Kansai Research Establishment.
Phone: +81-720-0759. Fax: +81-720-0596. e-mail: ichinose@apr.
‡
4-10
ion pair.
Inorganic salts such as metal perchlorates
§
enhance the separation of some SSRIP’s through an
jaeri.go.jp.
⊥
Osaka Prefecture University.
(
1) Photoinduced Electron Transfer; Fox, M. A., Chanon, M., Eds.;
(2) Marcus, R. A. J . Chem. Phys. 1956, 24, 966.
(3) Gould, I. R.; Farid, S. Acc. Chem. Res. 1996, 29, 522.
Elsevier: Amsterdam, 1988.
S0022-3263(97)01649-6 CCC: $15.00 © 1998 American Chemical Society
Published on Web 04/18/1998

Generation of Polyphenylene Radical Cations
J . Org. Chem., Vol. 63, No. 10, 1998 3205
Ch a r t 1
or imcompletely deoxygenated conditions. We reported
•+
an oxygenation of BP upon photoirradiation in oxygen-
saturated acetonitrile giving benzoic acid in a high yield
6
in the presence of DCA as an electron acceptor. It was
a surprising experience for the authors to observe that
the benzene ring had been completely decomposed under
the mild condition. We have studied the DCA-sensitized
photooxygenation of polyphenylenes (PP) (biphenyl, BP;
o-, m-, and p-terphenyl, o-, m-, and p-TP; quaterphenyl,
QUP; quinquephenyl, QIP) and some polyacenes such as
phenanthrene (Phen) and their cosensitization ability in
the DCA-sensitized cis-trans photoisomerization and
photooxygenation of cis- and trans-1,2-bis(4-methoxyphe-
4
,5
nyl)cyclopropanes (c-CP and t-CP).
We discuss the
•+
chemical properties of PP as a cosensitizing hole carrier
and a role of the neutral PP in the cosensitization
through the product analysis and laser flash photolysis
studies.
electrostatic interaction with radical ions.4-6,9,10 On the
other hand, aromatic hydrocarbons (ArH) such as pyrene
or biphenyl (BP, Chart 1) form their free radical cations
Exp er im en ta l Section
•
+
•+
(ArH ) through a photoexcitation with A, and ArH
Gen er a l. All the conventional analytical instruments are
undergoes a secondary electron transfer from S to give
the same as reported previously.4
,5
•
+ 7,8
ArH and S .
(
When the free energy change of BET
Ma ter ia ls. Acetonitrile was distilled three times over P
2 5
O
•-
•+
∆G-et) for an SSRIP of A and ArH ((A /ArH ) ) is highly
s
and then once over anhydrous K CO before use. Polyphe-
2
3
negative, separation of SSRIP takes place efficiently as
a consequence of slow BET rate as demonstrated by Farid
in 1990.11 For example, the quantum yield of 0.72 has
been reported for the formation of free radical cation of
nylenes (PP) (biphenyl, BP; o-, m-, and p-terphenyl, o-, m-, and
p-TP; quaterphenyl, QUP) and other organic compounds were
purified by recrystallization before use. cis- and trans-1,2-Bis-
(4-methoxyphenyl)cyclopropanes (c-CP and t-CP) were pre-
pared and purified according to the previous papers.
quephenyl (QIP) and inorganic chemicals were purchased and
used without further purification.
DCA-Sen sitized P h otooxygen a tion of P P . A 16 mL
portion of an acetonitrile solution containing p-TP (115 mg,
.5 mmol), DCA (5 mg, 0.022 mmol), and Mg(ClO
0.25 mmol) was irradiated with a 500 W high-pressure
mercury lamp (Eikosha) through an aqueous NH -CuSO
filter solution (>400 nm) under an oxygen stream for 5 h. The
solvent was removed under reduced pressure. The residue was
dissolved in benzene-ether (1:1) and extracted three times
with a 20% NaOH aqueous solution. The aqueous layer was
4
,5
•+
Quin-
BP (BP ) upon irradiation with 9,10-dicyanoanthracene
DCA) as an electron accepting sensitizer. A diffusive
(
•
+
secondary electron transfer from S to BP , which is
•
+
formed by competitive quenching of A*, gives free S in
a high quantum yield when BP is added to the S-A
system and the oxidation potential of S is lower than that
of BP by > 0.4 V. This cosensitizing effect often can meet
the synthetic purposes. The redox cycle will reproduce
ArH without consumption of ArH if chemical stability of
ArH is high enough. Since cosensitizers are generally
used to accelerate the DCA-sensitized photoreactions in
a high concentration of 0.01-0.2 M, their stability have
0
4 2
) (56 mg,
3
4
•
+
neutralized with a 20% H
benzene-ether. The organic layer was dried over Na
2
SO
4
solution and extracted with
SO , and
2
4
•
+
not been studied deeply. Reactivity of ArH toward
oxygen would be one of the most important facors because
preparative reactions are often conducted under aerated
the solvent was removed to give a mixture benzoic acid (1, 36%;
on the assumption that 1 M p-TP yields 2 molar equiv of 1)
and 4-phenylbenzoic acid (2e, 16%). The mixture was treated
with diazomethane. The yields and product ratios were
1
determined by GLC and H NMR spectra of the mixture
(4) (a) Mizuno, K. Kamiyama, N.; Ichinose, N.; Otsuji, Y. Tetrahe-
compared with those of authentic compounds. Photooxygen-
ation of other PP was performed in a similar manner.
Mea su r em en t of Qu a n tu m Yield s for th e DCA-Sen si-
tized P h otor ea ction s of CP . Quantum yields were deter-
dron 1985, 41, 2207. (b) Tamai, T.; Mizuno, K.; Hashida, I.; Otsuji, Y.
J . Org. Chem. 1991, 56, 5338.
(
(
5) Mizuno, K.; Ichinose, N.; Otsuji, Y. J . Org. Chem. 1992, 57, 1855.
6) (a) Mizuno, K.; Ichinose, N.; Tamai, T.; Otsuji, Y. Tetrahedron
Lett. 1985, 26, 5823. (b) Tamai, T.; Mizuno, K.; Hashida, I.; Otsuji, Y.
Photochem. Photobiol. 1991, 54, 23.
mined by the procedure of Murov, using a potassium ferriox-
12
alate actinometer.
The light source was a 500 W high-
(
7) Pac, C.; Nakasone, A.; Sakurai, H. J . Am. Chem. Soc. 1977, 99,
pressure mercury lamp (Eikosha), and the 405 nm Hg line was
isolated with an aqueous NH -CuSO filter solution, a Toshiba
5
806. Majima, T.; Pac, C.; Sakurai, H. Ibid. 1980, 102, 5265. Majima,
T.; Pac, C.; Nakasone, A.; Sakurai, H. Ibid. 1981, 103, 4499. Pac. C.
Pure Appl. Chem. 1986, 58. 1249.
3
4
UV-39 glass filter, and a KL-40 interference filter. The light
intensity was measured twice before and after actual photo-
reactions. Photoreactions were carried out in a quartz cuvette
(
8) (a) Steichen, D. S.; Foote, C. S. J . Am. Chem. Soc. 1981, 103,
1
1
855. (b) Schaap, A. P.; Lopez, L.; Gagnon, S. D. J . Am. Chem. Soc.
983, 105, 663. Schaap, A. P.; Siddiqui, S.; Gagnon, S. D.; Lopez, L.
(
1 × 1 × 4 cm) up to 30-70% conversion of the starting
Ibid. 1983, 105, 5149. (c) Lewis, F. D.; Bedell, A. M.; Dykstra, R. E.;
Elbert, J . E.; Gould, I. R.; Farid, S. J . Am. Chem. Soc. 1990, 112, 8055.
materials. The reaction mixtures were analyzed by GLC to
determine the conversion or yield of the reaction. The
quantum yields were estimated from the slope of linear parts
of plots in time course conversion.
(
d) Yamashita, T.; Yasuda, M.; Isamai, T.; Tanabe, K.; Shima, K.
Tetrahedron 1994, 50, 9275.
9) (a) Pac, C.; Ishitani, O. Photochem. Photobiol. 1988, 48, 767. (b)
(
Goodson, B.; Schuster, G. B. Tetrahedron Lett. 1986, 27, 3123. (c)
Gollnick, K.; Wellnhofer, G. J . Photochem. Photobiol. 1993, A74, 137.
A 4 mL portion of an aerated acetonitrile solution containing
-
4
-2
DCA (5 × 10 M) and c-CP (1 × 10 M) in a quartz cuvette
was irradiated at 405 nm to determine the quantum yield for
the t-CP formation. The progress of the isomerization was
followed by GLC. The photochemical isomerization of c-CP
(
d) Loupy, A.; Tchoubar, B.; Astruc, D. Chem. Rev. 1992, 92, 1141 and
references cited therin.
10) (a) Recently a simple polarity increase of solvents by the
(
addition of salts affecting the ∆G-et for BET within a radical ion pair
has been repored. (b) Thompson, P. A.; Simon, J . D. J . Am. Chem. Soc.
1
1
993, 115, 5657.
11) Gould, I. R.; Ege, D.; Moser, J . E.; Farid, S. J . Am. Chem. Soc.
990, 112, 4290.
(
(12) Murov, S. L.; Hug, S. Handbook of Photochemistry, 2nd ed.;
Marcel Dekker: New York, 1992.

3
206 J . Org. Chem., Vol. 63, No. 10, 1998
Tamai et al.
in the presence of PP was carried out by the irradiation of
solutions (4 mL) in a Pyrex tube (i.d. 10 mm) at >400 nm with
Ch a r t 2
a high-pressure Hg lamp with an aqueous NH
3
-3
4
-CuSO filter
5
solution. The concentration of PP was 5 × 10 M except for
-4
that of QUP (saturated, [QUP] < 4 × 10 M). Quantum yields
of t-CP formation in the presence of PP were determined from
the relative yield of t-CP to that in the absence of PP. In a
similar manner, quantum yields of the photooxygenation of
-
2
t-CP (2.5 × 10 M) in the absence and presence of PP (2.5 ×
-2
1
0
M) were determined. The progress of the photooxygen-
ation was followed by the decrease of t-CP or the increase of
4
the products using GLC. The products of the photooxygen-
ation of cis- and trans-3,5-bis(4-methoxyphenyl)dioxolane de-
composed at the injection part of GLC (250 °C) upon the GLC
analysis and gave the signals corresponding to the decomposi-
tion products, 4-methoxybenzaldehyde, 4-methoxyacetophe-
none, and 2-(4-methoxyphenyl)oxyrane in a ratio of 1:>0.9:
<
0.1. The yield of the dioxolane was quantified by GLC with
the signal of 4-methoxybenzaldehyde. Since t-CP was con-
verted to the dioxolane qantitatively and the dioxolane did not
give its decomposition products during the photooxygenation
until cyclopropane was consumed, the quantum yields obtained
from the decrease of t-CP and from the increase of the
dioxolane were identical each other within experimental error
yields 2 molar equiv of 1, see below) and 16% yields,
respectively (Chart 2). Without Mg(ClO , the efficiency
of the comsumption of p-TP was about half of that in the
presence of Mg(ClO , and a consumption of DCA giving
anthraquinone was observed.
4 2
)
4 2
)
(5%).
6
b,13
In the case of o-TP,
La ser F la sh P h otolysis. Acetonitrile solutions of DCA (5
-
4
-3
small amounts of phthalic anhydride (3) and diphenic
acid (4) were obtained as well as 1 but not 2-phenylben-
zoic acid (2c). Since biphenyl derivatives bearing an
electron-withdrawing group are inert under the same
conditions, a secondary oxygenation of 2c to 3 and 4
seems unlikely. The formation of 3 and 4 from phenan-
threne (see below) suggested that a cyclization of o-TP
was involved during the oxygenation. However, no
rearrangement products of o-TP such as triphenylene
were observed upon the irradiation under nitrogen or in
the presence of CP. These compounds would be formed
as a primary product via a cyclization of an oxygenated
intermediate to form a C-C bond between 2- and
×
10 mol dm ) containing PP and CP were irradiated with
-
1
a third harmonic pulse (355 nm, 6 ns, 15 mJ pulse ) from an
Nd:YAG laser (Continuum Surelite I-10) or a pulse from an
excimer laser (Lumonics EX-460) operated with an XeCl laser
gas mixture (308 nm, 12 ns, 70 mJ pulse ). A focused Xe arc
300-800 nm, pulse width 2 ms) which was synchronized with
-
1
(
the laser pulse by the use of a digital deley generator (Stanford
Research Systems DG535) was passed through the sample as
a monitor light for transient absorption measurement. The
arc was monitored as a spectral or a temporal response signal
of transient species with a gated-multichannel spectrometer
(
Hamamatsu PMA-50, gate time 5-100 ns) or with a mono-
chromatic ns-transient measurement system (Otsuka NSP-
000G) equipped with a photomultiplier (Hamamatsu R-1894)
and a digital oscilloscope (Tektronics TDS380P). The quantum
1
2
′-carbons of o-TP. Similar photooxygenation of m-TP
•
+
yields of CP in the presence of various cosensitizers and
•
+
afforded 1 and 3-phenylbenzoic acid (2d ) in 76% and 38%
molar equivalent yields, respectively. Since the sum of
the yields exceed unity, the central ring must decompose
into two carboxylic groups. Therefore, the yield of 1 was
corrected to be 38%. This also will be the case for o- and
p-TP. The photooxygenation of QUP and QIP was
conducted in a mixed solvent of acetonitrile-benzene (8:
extinction coefficient of CP were determined according to
1
1
Farid et al. from the transient absorption spectra at 100 ns
after the laser excitation of aerated acetonitrile solutions
containing PP, t-CP, and DCA. An aerated acetonitrile solu-
-
4
tion of BP (0.2 M) and DCA (5 × 10 M) was used as a
1
1
standard actinometer.
The rates for the secondary electron transfer from CP to
PP• was determined by the laser flash photolysis of aerated
acetonitrile solutions of 1,4-dicyanonaphthalene (DCN) ([DCN]
+
2
) because of their low solubility in acetonitrile. The
-
4
-4
irradiation of QUP afforded 1 and 2e in low yields. In
the case of QIP, consumption of QIP giving a complex
mixture containing 1 was not completed owing to a rapid
consumption of DCA. Alkyl-substituted derivatives such
as 4-methylbiphenyl (MeBP) showed a high reactivity
toward oxygenation affording R-carbonyl compunds 4-phe-
nylbenzaldehyde (5) and 2e via an oxidation of R-carbon
)
1 × 10 M) containing PP (0.02-0.2 M) and CP (1 × 10
-
3
•
+
to 5 × 10 M) to monitor the decay curves of PP around
•
+
6
50 nm and the rise time of the decay curves of CP at 560
nm. The use of DCN instead of DCA was to avoid the
disturbance by the strong fluorescence, ground-state absorp-
tion, and transient absorption of DCA in the wavelength of
4
00-650 nm. The rates were calculated from the slope of the
•
+
plots of the first-order decay constant for PP or rise time of
CP versus CP concentration and were almost identical within
the error of 10%. The second-order rate constants for the decay
of t-CP by back-electron transfer from O
by the decay curves of CP at 560 nm in the absence and
6
b
•+
as well as 1 and 2b via the ring oxidation.
The
disappearance of Phen and Naph in their photooxygen-
ation was much faster than that of PP and their photo-
oxygenation gave 3 and 4 from Phen and 3 from Naph,
respectively. PP derivatives were stable and recovered
completely after the irradiation in nitrogen-saturated
methanol in the presence of DCA.14 The relatively low
•
+
•-
2
were determined
•
+
-
2
presence of PP (2.5 × 10 M). The rate constants were
measured three times and averaged.
Resu lts a n d Discu ssion
(
13) Mizuno, K.; Tamai, T.; Nakanishi, I.; Ichinose, N.; Otsuji, Y.
DCA-Sen sitized P h otooxygen a tion of P P . Irradia-
tion of an acetonitrile solution (16 mL) containing p-TP
Chem. Lett. 1988, 2065.
14) Several nucleophilic addition reactions to Phen•+ and Naph•+
(
(0.5 mmol), DCA (0.02 mmol), and Mg(ClO
4
)
2
(0.25 mmol)
have been reported. (a) Yasuda, M.; Pac. C.; Sakurai, H. J . Org. Chem.
981, 46, 788. (b) Mizuno, K.; Pac, C.; Sakurai, J , Chem. Soc., Chem.
Commun. 1975, 533. Yasuda, M.; Pac. C.; Sakurai, H. J , Chem. Soc.,
Perkin Trans. 1 1981, 746. (c) Yasuda, M.; Yamashita, T.; Matsumoto,
T.; Shima, K. J . Org. Chem. 1985, 50, 3667.
1
with an oxygen bubbling for 5 h comsumed p-TP com-
pletely giving benzoic acid (1) and 4-phenylbenzoic acid
(2e) in 72 (or 36% on the assumption that the 1 M p-TP

Generation of Polyphenylene Radical Cations
J . Org. Chem., Vol. 63, No. 10, 1998 3207
Ta ble 1. DCA-Sen sitized P h otooxygen a tion of
Ch a r t 3
a
P olyp h en ylen e Com p ou n d s
product(s)/%
compd
irradn time/h
1
2
3
4
BP
o-TP
m-TP
p-TP
QUP
MeBP
Phen
Naph
5
5
5
5
8
1
1
1
76
35
38
36
20
32
b
b
b
12
3
38 (2d )
16 (2e)
31 (2e)
Ta ble 3. Effect of P olyp h en ylen e Com p ou n d s on th e
c
DCA-Sen sitized P h otoisom er iza tion of c-CP to t-CP a n d
d
•+
th e Tu r n over of CP
e
20 (2b), 5 (2e)
(Φsep)a
concn/M
Φcftb
0.28g
5.6
15.3
10.8
20.5
0.48
c
Φcft/ΦCP•+d
additive
ΦCP•+
8
86
8
none
BP
(0.12)
(0.72)e
(0.64)
(0.32)
0.09
0.14
0.17
3
40
90
83
186
-3
5 × 10
5 × 10
5 × 10
5 × 10
<4 × 10
5 × 10
1 × 10
1 × 10
a
-2
-4
[
Polyphenylene compounds] ) 3.1 × 10 M, [DCA] ) 5 × 10
-3
-3
-3
-4 f
-3
-2
-2
o-TP
m-TP
p-TP
QUP
Me-BP
Phen
Naph
-
2
M, [Mg(ClO4)2] ) 1.6 × 10 M in dry acetonitrile. Conversions of
0.13
b
the reaction were >95%. Yields on the assumption that 1 M TP
0.11h
c
yields 2 molar equiv of 1 (see text). p-TP was suspended in
d
acetonitrile because of its low solubility. QUP was suspended in
7.2
e
CH3CN-C6H6 (8:2). A small amount (2%) of 4-phenylbenzalde-
(0.62)e
(0.58)e
13.0g
0.31
0.28
42
45
hyde (5) was also formed.
12.5g
a
Separation efficiency of SSRIP (counter radical anion: DCA•-)
into free radical cation obtained by laser flash photolysis. Quan-
Ta ble 2. Ra te Con sta n ts for th e F lu or escen ce
Qu en ch in g of DCA, Oxid a tion P oten tia ls a n d Ion iza tion
P oten tia ls of P olyp h en ylen e Com p ou n d s, a n d Ca lcu la ted
b
tum yields of the photoisomerization of c-CP to t-CP; [c-CP] ) 0.01
M. [DCA] ) 5 × 10 M in dry acetonitrile. c Quantum yields of
-4
∆
GEt Va lu es for th e On e-Electr on Tr a n sfer P r ocess fr om
P olyp h en ylen e Com p ou n d s to 1DCA* in Aceton itr ile
•+
free CP generated by direct and indirect sensitization. The
1
DCA* quenching rates and concentrations of CP and PP were
taken into account. Turnover of CP in the photoisomerization,
compd
kq /M s-1
a
-1
Eox1/2 /V
b
IP /eV
c
∆Get /kJ mol
d
-1
d
•+
9
e
f
BP
3.6 × 10
1.45
1.42
1.40
1.32
5.87
-16.4
-30.9
-18.3
-26.1
see text. The values were taken from ref 11. Saturated solution.
9
g
The values were taken from ref 5. h Directly measured by laser
o-TP
m-TP
p-TP
QUP
QIP
MeBP
Phen
Naph
t-CP
7.0 × 10
9
5.5 × 10
flash photolysis.
1
0
1.2 × 10
5.97
5.90
5.87
for the photooxygenation of PP. The efficiency of the
photooxygenation of QUP and QIP was lower than that
of p-TP. The oxidation potentials of QUP and QIP could
not be measured in acetonitrile because of their low
solubility but would be almost equal to or slightly smaller
than that of p-TP on the comparison of their ionization
potentials (Table 2).16 The low reactivity of QUP and QIP
would be attributed to a low reactivity of their radical
cations QUP and QIP toward oxygen owing to the
delocalization of their radical cites over the whole mol-
ecules.
Cosen sitiza tion of P P in th e DCA-Sen sitized
P h otoisom er iza tion of Dia r ylcyclop r op a n es. As
previously reported, irradiation of an aerated acetonitrile
1
1
1
1
1
0
0
0
0
0
1.1 × 10
1.4 × 10
1.4 × 10
1.8 × 10
1.7 × 10
1.1 × 10
1.34
1.17
1.22
0.55
0.65
1.27
-27.0
-43.4
-38.6
-103.2
-93.6
-33.8
c-CP
DX
e
10
a
Rate constants for the fluorescence quenching of DCA in
•+
•+
-
4
aerated acetonitrile: [DCA] ) 4 × 10 M; τ(DCA, air) ) 16.1 ns.
b
Oxidation potentials (V vs Ag/AgClO4) were determined as half-
peak potentials by cyclic voltammetry: Pt electrode, tetraethy-
c
lammonium perchlorate (0.1 M) in acetonitrile. Solid-state ion-
d
ization potentials taken from ref 16. Calculated free energy
changes for the one-electron-transfer process from polyphenylene
compounds to DCA* in acetonitrile. See ref 15. Reduction
potential of DCA; -1.33 V vs Ag/AgClO4 in acetonitrile. Cis
isomer, the values are taken from ref 4b.
1
-
4
e
solution containing DCA (5 × 10 M) and cis- or trans-
1
1
,2-bis(4-methoxyphenyl)cyclopropanes (c-CP, t-CP, 1 ×
M) at >400 nm (Chart 3) caused the cis-trans
-
2
0
isomerization of the cyclopropanes, affording the photo-
stationary state (c-CP:t-CP ) 5:95).5 The quantum yields
for the isomerization from c-CP to t-CP were measured
in the absence and presence of PP compounds such as
BP, o-, m-, and p-TP, and also Phen. The results are
summarized in Table 3. The addition of PP showed a
remarkable enhancement in the quantum yields of the
photoisomerization. The order of enhancement in the
reaction efficiency was p-TP > o-TP > m-TP > MeBP >
BP (> QUP). Undoubtedly the quantum yields exceeded
unity indicating the reaction proceeds in a chain mech-
anism. The limiting quantum yield in the absence of PP
also exceeded unity (1.8).5 Addition of polyacene com-
pounds, phenanthrene (Phen) and naphthalenes (Naph),
also accelerated the photoisomerization. It was con-
firmed that all the PP compounds were not consumed
even upon prolonged irradiation.
reactivity of PP in the photooxygenation as well as
nucleophilic addition has proved that PP is suitable for
cosensitizer. The results are summarized in Table 1.
•
+
Formation of PP via photoinduced electron transfer
1
(
PET) from PP to DCA* must be involved in this
6
photooxygenation as discussed in our previous papers.
Although the detailed pathway for the decomposition of
the benzene ring of PP has not been completely clarified,
•
+
3
2
PP would be attacked by O to give the oxygenated
products. The free energy changes for one-electron
transfer from PP to 1DCA* (∆Get) estimated by the
Rehm-Weller equation15 are largely negative and the
fluorescence of DCA was quenched by PP in nearly
diffusion-controlled rates (Table 2). The generation of
•
+
PP was readily observed by laser flash photolysis. No
oxygenated products were obtained by the irradiation of
PP, Phen, and Naph in less polar solvents in the presence
Cosen sitiza tion of P P in th e DCA-Sen sitized
P h ot ooxygen a t ion of Dia r ylcyclop r op a n es. The
of DCA or by the irradiation in CH
2 2
Cl in the presence
of methylene blue under oxygen to generate singlet
1
2
oxygen ( O ). These results support the PET mechanism
(16) Matsuoka, S.; Fujii, H.; Yamada, T.; Pac, C.; Ishida, A.;
Takamuku, S.; Kusaba, M.; Nakashima, N.; Yanagida, S.; Hashimoto,
K.; Sakata, T. J . Phys. Chem. 1991, 95, 5802.
(15) Rehm, D.; Weller, A. Isr. J . Chem. 1970, 8, 259.

3
208 J . Org. Chem., Vol. 63, No. 10, 1998
Tamai et al.
transient absorption of DCA•- was quenched by the
presence of oxygen, no fast quenching for the BP or no
new transient species were observed within 1 µs by
Ta ble 4. Effect of P olyp h en ylen e Com p ou n d s on th e
DCA-Sen sitized P h otooxygen a tion of t-CP a n d th e
•+
•
+
Tu r n over of CP
ΦDX/ΦCP•+ d
additive
(Φsep)a
concn/M
ΦDXb
ΦCP•+ c
saturation of the solution with oxygen. The rate constant
•
+
none
BP
o-TP
m-TP
Phen
(0.12)
(0.72)
1.6f
1.9
11.6
9.3
0.11
0.16
0.21
0.13
0.34
15
12
55
71
32
for the oxygen attack on BP was, therefore, smaller
e
-2
8
-1 -1
2.5 × 10
2.5 × 10
2.5 × 10
2.5 × 10
than 10 M
s . This stability allowed us to assume
-
-
2
2
(0.64)
(0.32)
(0.62)
•+
that a secondary electron transfer from CP to BP will
afford CP and BP in unit yield in the DCA-BP-t-CP
•+
e
-2
10.9
-3
system. The addition of t-CP (5 × 10 M) to the aerated
a
•-
Separation efficiency of SSRIP (counter radical anion: DCA
)
-4
acetonitrile solution ([BP] ) 0.2 M and [DCA] ) 5 × 10
into free radical cation obtained by laser flash photolysis. b
Quantum yields for the formation of DX by the photooxygenation
M) gave a broad transient absorption at 450-580 nm and
>650 nm after 400 ns from the laser pulse (Figure 1a).
-
2
-4
of t-CP; [t-CP] ) 2.5 × 10 M. [DCA] ) 5 × 10 M in dry
•+
c
•+
The transient was assigned to t-CP by comparison with
acetonitrile. Quantum yields of free CP generated by direct and
1
indirect sensitization. The DCA* quenching rates and concentra-
tions of CP and PP were taken into account. Turnover of CP
those observed upon the low-temperatue γ-radiolysis and
d
•+
17a
pulse radiolysis of t-CP.
Assuming complete reduction
e
in the photooxygenation of CP, see text. The values were taken
•+
of initially formed BP by t-CP giving the same amount
from ref 11. f The values taken from ref 4a.
•
+
•+
of t-CP and a negligibly small contribution of c-CP in
1
8a
Ch a r t 4
the transient absorption, the absorption coefficient of
•
+
-1
-1
t-CP was determined to be 7500 M cm at 500 nm
Figure 1a). From this extinction coefficient, a qunantum
yield of the formation of t-CP upon the photolysis of an
(
•
+
-
2
aerated solution of t-CP (5 × 10 M) containing DCA (5
10 M) was determined to be 0.11, which was correc-
ted to 0.12 for the incomplete quenching at the concen-
tration of t-CP (Φ
-
4
×
DCA-sensitized photooxygenation of c-CP or t-CP gave
cis- and trans-3,5-bis(4-methoxyphenyl)-1,2-dioxolanes
f
/Φf0 ) 0.089). This value corresponds
(
cis/trans ) 7/3) almost quantitatively.5 The quantum
to the quantum yield for the charge separation (separa-
yield of the dioxolane formation was 1.6, also indicating
that the photoreaction involves a chain mechanism. As
shown in Table 4, the addition of PP and also Phen to
the reaction system increased the quantum yield ef-
ficiently (Chart 4). Neither any changes in the isomeric
ratio of dioxolanes nor consumption of PP during the
photooxygenation of CP was observed.
tion efficiency, Φsep) if the SSRIP is formed in a unit
•
+
quantum yield. Obviously the free t-CP yield was in-
creased by the addition of BP as indicated by the tran-
•
+
sient absorption spectra of t-CP in Figure 2. On the
•+
other hand, although transient absorption of c-CP (Fig-
ure 1b) was obtained only by the photolysis of an aerated
-2
acetonitrile solution of c-CP (5 × 10 M) containing DCA
in a similar manner, its extinction coefficient and yield
could not be determined because of the too efficient
isomerization to t-CP during the accumulation of the
transient spectra in the presence of BP. The generation
Deter m in a tion of Ra d ica l Ca tion Yield s by La ser
F la sh P h otolysis. The mechanism of DCA-sensitized
photoisomerization and photooxygenation of CP which
involves photoinduced electron transfer and a chain
•
+
•+
•+
process mediated by CP has been discussed in the
previous reports.4 The proposed mechanisms for the
photoisomerization and photooxygenation are shown in
Scheme 2. The PP radical cations also could be the key
intermediate for the cosensitization in these reactions as
well as the photooxygenation of PP. The fluorescence of
DCA in acetonitrile was also quenched by CP at nearly
diffusion-controlled rates. The values of ∆Get for c- and
t-CP were largely negative as given in Table 2, suggesting
that radical cations of CP could be generated by one-
of t-CP and c-CP by the direct photolysis with DCA
,5
was not affected by the nitrogen purge suggesting that
•-
•-
oxygen quenched the free DCA but not DCA in SSRIP.
The transient absorption measurement of o-TP- and
m-TP-DCA solutions ([TP] ) 0.02 M) indicated that o-
•
+
and m-TP were stable within 1 µs in the presence of
oxygen.1 When the concentration of TP was increased,
8b
•
+
the transient absorption of TP decreased considerably
and was not observed at 0.2 M. This can be due to a
•
-
•+
quenching of a radical ion pair (DCA /TP ) by TP. The
1
electron transfer from these compounds to DCA*. How-
rates for the secondary electron transfer from t-CP to
•
+
•+
ever, the turnover values in the chain isomerization and
photooxygenation of CP are still unknown. Our questions
are (1) how large are the quantum yields of free CP
TP were determined by the decay time of TP at 650
•
+
nm for o-TP or the rise time of t-CP at 560 nm for o-,
m-, and p-TP upon the photolysis of TP-1,4-dicyanon-
aphthalene (DCN) solutions containing various concen-
tration of t-CP. Typical decay and rise curves of the
transient species are given in Figure 3a. All the rates
•
+
formation in the absence and presence of PP and (2) how
the quantum yields for the reactions of CP reflect those
•
+
of free CP formation.
1
0
Direct evidence for the generation of radical cation of
PP upon the DCA-sensitized photoreaction was obtained
from the nanosecond laser flash photolysis (LFP). The
LFP of an acetonitrile solution of BP (0.2 M) containing
showed nearly diffusion-controlled rates (o-TP, 1.4 × 10 ;
9
10
-1 -1
•+
m-TP, 9.7 × 10 ; p-TP, 1.3 × 10
M
s ). Since TP
was stable within the timeframe for secondary electron
-
4
DCA (5 × 10 M) under nitrogen was performed by a
conventional nanosecond transient absorption measure-
ment system using an Nd:YAG or XeCl excimer laser.
(17) (a) Toki, S.; Komitsu, S.; Tojo, S.; Takamuku, S.; Ichinose, N.;
Mizuno, K.; Otsuji, Y. Chem. Lett. 1988, 433. (b) Mizuno, K.; Ichinose,
N.; Otsuji, Y. Unpublised results. See also ref 5.
•
+
(18) (a) The quantitative charge transfer from CP to PP was
•
-
The transient absorption was readily assigned to DCA
confirmed by the observation of isosbestic points in the time-resolved
absorption spectra of acetonitrile solutions containing DCN, PP, and
t-CP in the period of 0-1 µs. (b) The transient species were readily
•
+
11
•+
and BP as reported by Farid et al. The free BP yield
of 0.72 and the molar extinction coefficient at 670 nm of
•
+
assigned to TP by the comparison of the spectra with the reported
-
1
-1
1
5 000 M cm were employed to determine the values
of these parameters for CP and PP . Though the
•+
ones of TP . Shida, T. Electronic Absorption Spectra of Radical Ions;
•
+
•+
Elsevier: Amsterdam, 1988.

Generation of Polyphenylene Radical Cations
J . Org. Chem., Vol. 63, No. 10, 1998 3209
Sch em e 2
F igu r e 2. Transient absorption spectra of t-CP•+ generated
by the indirect (cosensitization) and direct sensitization upon
laser pulse excitation of aerated acetonitrile solutions of t-CP
-
3
-4
(
1 × 10 M) and DCA (5 × 10 M) in the presence and
absence of BP (0.02 M). The spectra were taken at 400 ns after
the laser pulse. Excitation wavelength: 355 nm. Laser inten-
sity: 15 mJ pulse
-
1
.
•
+
bance at a time of 400 ns due to t-CP generated by the
•
+
use of TP and BP, free TP yields were determined
•
+
assuming direct formation of t-CP was small because
of a low concentration of t-CP employed as compared to
•
+
that of TP. The quantum yields of the formation of TP
were 0.47 (o-TP) and 0.22 (m-TP), but that for p-TP could
not be determined owing to its low solubility in acetoni-
trile. The values were also corrected for the imcomplete
quenching to 0.64 (o-TP) and 0.32 (m-TP). Obviously, the
•
+
free CP yield was increased also by the addition of TP.
Tu r n over of th e Ch ain Reaction s. The photochemi-
cal isomerization and photooxygenation of CP involve
chain processes without doubt, because their quantum
yields largely exceed unity. The triplet excited state of
F igu r e 1. Transient absorption spectra of (a) BP and t-CP•+
•
+
•
+
and (b) c-CP observed upon laser pulse excitation of an
aerated acetonitrile solution containing (a) BP (0.2 M) and
3
DCA ( DCA*) was hardly observed throughout the pho-
tolysis study, ruling out the incorporation of triplet
species for these reactions. From the values of the
separation efficiencies (Φsep) and the fluorescence quench-
ing rates for CP and PP, the initial quantum yields for
-
4
-3
DCA (5 × 10 M) and BP (0.2 M), t-CP (5 × 10 M), and
-
4
-2
DCA (5 × 10 M) and (b) c-CP (5 × 10 M) and DCA (5 ×
-4
1
1
0
M). The spectra were taken at (a) 100 and 400 ns and (b)
00 ns after the laser pulse, respectively. Excitation wave-
length: 308 nm. Laser intensity: 70 mJ pulse . The extinction
coefficient was determined according to ref 11.
-1
•+
CP generation (ΦCP•+) both through the direct and
indirect electron-transfer processes under the photoi-
somerization and photooxygenation conditions were cal-
culated (Tables 3 and 4). The ratio of the quantum yield
transfer (400 ns) upon the photolysis of aerated soluions
-
3
-4
of TP containing t-CP (5 × 10 M) and DCA (5 × 10
of the reactions (Φ ) obtained by the stationary-state
r
M), the concentration of t-CP•+ formed was expected to
experiments to ΦCP•+ (Φ /Φ •+) will be the turnover of the
r
CP
•+ 18a
•+
be equal to that of the initially formed TP .
By
chain reactions of CP . The results are also shown in
comparison of the values of the 500 nm transient absor-
Tables 3 and 4. Though the values of ΦCP•+ vary from

3
210 J . Org. Chem., Vol. 63, No. 10, 1998
Tamai et al.
Sch em e 3
F igu r e 3. (a) Decay and rise curves of the transient absorp-
•
+
•+
tion of o-TP (650 nm, closed circle) and t-CP (560 nm, open
than 50 when c-CP•+ was produced in the dry acetonitrile
4 2
in dark by the use of cationic oxidants such as Cu(BF ) ,
circle) observed upon laser pulse excitation of an acetonitrile
-
4
solution of o-TP (0.025 M) in the presence of DCN (1 × 10
-
4
M) and t-CP (5 × 10 M) under nitrogen. An undershoot
•+
-
+
- 17b
(4-BrC
6
H
4
)
3
N SbCl
6
, and NO BF
4
.
However, neu-
around 0 ns was due to the fluorescence of DCN. Excitation
tral oxidants such as 2,3-dichloro-5,6-dicyano-1,4-benzo-
-
1
wavelength: 355 nm. Laser intensity: 15 mJ pulse . (b) Plot
1
7b
-
1
quinone (DDQ) gave smaller turnover values of ≈5.
of the reciprocal of transient absorption at 560 nm ((∆OD)
)
These results strongly suggest that the turnover value
depends on the charge of the electron donor for CP in
the back-electron-transfer processes, i.e., Cu , (4-
versus time obtained by the laser flash photolysis of an aerated
•
+
acetonitrile solution of t-CP (0.025 M) containing DCN (1 ×
-4
+
1
0
M) in the absence (open circle) and presence (closed circle)
•
-
•-
of o-TP (0.025 M). Excitation wavelength: 308 nm. Laser
6 4 3
BrC H ) N, NO, (benzonitrile) , or DDQ .
-
1
intensity: 70 mJ pulse
.
The turnover in the photooxygenation was also in-
creased by the addition of PP although the value in the
absence of PP was higher than that of the isomerization.
Interestingly the values of the turnover were strongly
dependent on the additive used and the orders in the
effects obtained for both reactions agreed with each
other: BP < Phen, Naph < o-, m-TP. This strongly
0
.09 to 0.31 depending on the aromatic additive used, the
quantum yields of photoisomerization range from 0.28
up to 20. The turnover in the photoisomerization was 3
in the absence of PP and greatly increased up to 40-190
in the presence of PP. Since the reciprocal plot of the
quantum yield of the isomerization Φcft in the absence
-
1
-1
•+
of PP and the concentration of c-CP ((Φcft
)
vs [c-CP] )
suggests that the high yield of CP due to the high
showed a linear relationship for a range of [c-CP] from
separation efficiency of solvent-separated radical ion pair
-
3
•-
•+
1
0
to 0.2 M, the turnover could be considered to be
(SSRIP) of DCA and PP (DCA /PP ) into free ions is
not the sole factor of the cosensitization of PP for the
photoreactions of CP (Scheme 3).
constant. We reported that the turnover of the isomer-
ization of c-CP to t-CP induced by the 60Co γ-ray irradia-
tion in benzonitrile was ≈5 at a concentration of CP of 1
A simple equilibrium between CP•+ and PP does not
seem plausible because the oxidation potentials of PP are
-
2
17a
×
10 M.
We also observed a high turnover of more

Generation of Polyphenylene Radical Cations
J . Org. Chem., Vol. 63, No. 10, 1998 3211
higher than that of CP by 0.6-0.9 V (redox equilibrium
reducing the reactivity but it would prevent CP•+ from
its disappearance by a nucleophilic attack by impurities
such as water. Increase of the lifetime of CP by the
-
9
constants , 10 ). The uphill electron transfer from
•+
•+
allyltrialkylsilane and -stannane derivatives to PP has
been found to have a limit for the endothermicity of 0.4
π-complex formation could be fairly plausible.
1
9
V.
In fact, we observed that the secondary electron
For the photooxygenation of CP, the chain carrier must
•
+ 24
transfer took place in diffusion-controlled rates and a
be a radical cation of the dioxolane (DX ). The oxida-
tion potential of DX is higher than that of t-CP by 0.7 V
and comparable to that of TP, Phen, and Naph. In this
•
+
complete decay of PP within 1 µs after the laser pulse.
•+
We did not observe PP after >1 µs after the laser pulse.
•
+
•+
If there is a redox equilibrium with a constant of larger
case, π-complex formation between DX and PP [DX /
•
+
than 0.1, PP should have been detected with our
transient absorption measurement system. Though the
irreversible chemical change in the substrate may assist
the uphill electron transfer by shifting the equilibrium,8
this would not be the case.
PP] and also a redox equilibrium would be more likely
•
+
than that between CP and PP. However, the lifetime
•
+
of DX would be too short to form the π-complex or to
a-c
reach the equilibrium in the presence of CP (see below).
•
+
Therefore, the π-complex formation between PP and CP
•
+
The high turnover can be attributed to a specific
rather than DX would play an important role both in
the photoisomerization and photooxygenation.
•
.+
interaction between CP
and PP like a π-complex
•+
formation of [CP /PP]. In the previous paper, we have
reported that the inhibition of a photochemical [4π + 2σ]
cycloaddition of DCA and CP, which proceeds via their
radical ion pair, by the addition of aromatic hydrocarbons
such as Phen.5 The inhibition of the reaction could be
Since the material balance in the photoisomerization
4,5
and photooxygenation was close to unity, the termina-
tion of the chain would be the diffusive back-electron
•-
•+
•+
transfer from free O
2
to CP or DX . The larger value
of the turnover for the oxygenation than that of the
isomerization in the absence of the additives could be
explained by the competition between chain propagation
•
-
•+
explained by the separation of a pair of DCA and CP
through the complex formation as explained by Pac et
al. for the endothermic sensitization reactions instead of
the endothermic hole transfer.5,7 Since radical cations
of aromatic hydrocarbons tends to form dimer radical
cations as a result of charge resonance stabilization
25
and the termination by the back-electron transfer. The
•+
hole transfer in the isomerization takes place from t-CP
to c-CP with a free energy change of -0.1 eV and its rate
constant would be close to those of self-exchange reac-
whose transition band has been observed in the IR region
8
9
-1 -1
tions of 10 -10 M
s . On the other hand, the hole
of 1000-2000 nm,2
0-22
the “hetero-type” π-complex for-
•+
transfer from DX to t-CP with an exothermicity of -0.7
eV in the oxygenation would take place in a diffusion-
mation would be acceptable.2 The charge resonance
2j
stabilization in the π-complex formation would be large
10
-1 -1
controlled rate (≈10
M
s ), although following attack
•+
when the SOMO level of CP is close to the HOMO level
of PP. Cross-type cycloaddition of alkenes may proceed
via hetero-type dimer radical cations.23 The reactivity
of two alkenes is high owing to a high perturbation
interaction between their SOMO and HOMO levels when
their oxidation potentials are close each other. A weak
•+
•+
of oxygen on CP reproducing DX is a slow process
8
-1 -1
•+
(
<10 M
s
). Therefore, the lifetime of DX would be
-2
much shorter (<10 ns) in the presence of 10 M CP as
compared to that of CP and the deactivation of the both
reactions must be due to electron transfer from free O
to the long-lived CP .
•+
•-
2
•+
•
+
perturbation between CP and PP which does not lead
We have measured the second-order rate constants for
to any reaction may raise the SOMO level of CP•+
•+
the decay of t-CP in the absence and presence of PP
upon the photolysis of an aerated acetonitrile solution of
t-CP containing DCN. As shown in Figure 3b, the plots
of the reciprocal of the transient absorbance at 560 nm
(
19) Nakanishi, K.; Mizuno, K.; Otsuji, Y. Bull. Chem. Soc. J pn.
993, 66, 2371.
20) (a) Badger, B.; Brocklehurst, B.; Russel, R. D. Chem. Phys. Lett.
967, 1, 122. (b) Badger, B.; Brocklehurst, B. Nature 1968, 219, 263.
1
1
(
(
(
(1/∆OD) showed a linear relationship with time indicat-
c) Badger, B.; Brocklehurst, B. Trans Faraday Soc. 1969, 65, 2582.
d) Badger, B.; Brocklehurst, B. Ibid. 1969, 65, 2588. (e) Badger, B.;
Brocklehurst, B. Ibid. 1970, 66, 2939. (f) Ekstrom, A. J . Phys. Chem.
970, 74, 1705. (g) Rodgers, M. A. J . Chem. Phys. Lett. 1971, 9, 107.
h) Rodgers, M. A. J . J . Chem. Soc., Faraday Trans. 1 1972, 68, 1278.
i) B u¨ hler, R. E.; Funk, W. J . Phys. Chem. 1975, 79, 2098. (j) Irie, S.;
ing the second-order kinetics. This can be attributed to
•
-
•+
the diffusive back-electron transfer from O
2
to CP as
1
(
(
shown in Scheme 3. The second-order rate constants for
•
+
the decay of t-CP were determined from the slope of
Horii, H.; Irie, M. Macromolecules 1980, 13, 1355.
21) (a) Kira, A.; Arai, S.; Imamura, M. J . Chem. Phys. 1971, 54,
890. (b) Arai, S.; Kira, A.; Imamura, M. Ibid. 1972, 56, 1777. (c) Kira,
A.; Arai, S.; Imamura, M. J . Phys. Chem. 1972, 76, 1119. (d) Kira, A.;
Imamura, M.; Shida, T. Ibid. 1976, 80, 1445. (e) Egusa, S.; Arai, S.;
Kira, A.; Imamura, M.; Tabata, Y. Radiat. Phys. Chem. 1977, 9, 419.
the plots, the molar extinction coefficient at 560 nm (4800
(
4
(23) (a) Mizuno, K.; Kaji, R.; Okada, H.; Otsuji, Y. J , Chem. Soc.,
Chem. Commun. 1978, 594. Mizuno, K.; Ueda, H.; Otsuji, Y. Chem.
Lett. 1981, 1237. Mizuno, K.; Ishii, M.; Otsuji, Y. J . Am. Chem. Soc.
1981, 103, 5570. (b) Farid, S.; Hartman, S. E.; Evans, T. R. The
Exciplex; Gordon, M., Ware, W. R., Ed.; Academic Press: New York,
1975; p 327. (c) Maroulis, A. J .; Arnold, D. R. J . Chem. Soc., Chem.
Commun. 1979, 351. (d) Calhoun, G. C.; Shuster, G. B. Tetrahedron
Lett. 1984, 27, 911. (e) Pabon, R. A.; Bellville, D. J .; Bauld, N. L. J .
Am. Chem. Soc. 1983, 105, 5158; 1984, 106, 2730.
(
f) Kira, A.; Imamura, M. J . Phys. Chem. 1979, 83, 2267.
(22) (a) Tsujii, Y.; Tsuchida, A.; Yamamoto, M.; Nishijima, Y.
Macromolecules 1988, 21, 665. (b) Tsujii, Y.; Tsuchida, A.; Yamamoto,
M.; Nishijima, Y.; Wada, Y. Polym. J . 1988, 20, 837. (c) Yamamoto,
M.; Tsujii, Y.; Tsuchida, A. Chem. Phys. Lett. 1989, 154, 559. (d)
Tsuchida, A.; Tsujii, Y.; Ito, S.; Yamamoto, M. J . Phys. Chem. 1989,
•
+
•+
9
3, 1244. (e) Tsuchida, A.; Tsujii, Y.; Ohoka, M.; Yamamoto, M. Nippon
(24) The oxygenation of CP giving DX must involve two DX forms
•
+
•+
Kagaku Kaishi 1989, 1285. (f) Tsujii, Y.; Takami, K.; Tsuchida, A.;
Ito, S.; Onogi, Y.; Yamamoto, M. Polym. J . 1990, 22, 319. (g) Tsuchida,
A.; Tsujii, Y.; Ohoka, M.; Yamamoto, M. J . Phys. Chem. 1991, 95, 5797.
of an open-ring DX and a closed-ring DX , although the structure
•
+
•+
of DX as a chain carrier is not clear. The open-ring DX seems to
have a lifetime enough for an internal rotation giving a stable isomer
of DX by ring closure. Ichinose, N.; Mizuno, K. Tamai, T.; Otsuji, Y. J .
Org. Chem. 1990, 55, 4079.
(
1
h) Tsuchida, A.; Takamura, H.; Ito, S.; Yamamoto, M. Macromolecules
991, 24, 4061. (i) Tsuchida, A.; Takamura, H.; Yamamoto, M.; Lee,
B.; Ikeda, T.; Tazuke, S. Bull. Chem. Soc. J pn. 1992, 65, 909. (j)
Tsuchida, A.; Takamura, H.; Yamamoto, M. Chem. Phys. Lett. 1992,
(25) Since the quantum yield of the photoisomerization under
nitrogen in the absence of additives is small (<0.01) and cis-trans
ratio in the stationary state is the same as that in the presence of
additives, the contribution of a 1,3-biradical formed by the BET
within the SSRIP of DCA and CP would be small.
(26) Karki, S. B.; Dinnocenzo, J . P.; Farid, S.; Goodman, J . L.; Gould,
I. R.; Zona, T. A. J . Am. Chem. Soc. 1997, 119, 431.
1
98, 193. (k) Tsuchida and Yamamoto also observed the decrease of
2
6
the rate constant for the reduction of dimer radical cations by
triethylamine. They attributed this to the reduction in free energy
changes for the electron transfer by the stabilization energy for the
dimer formation.2
2a,b,e,h

3
212 J . Org. Chem., Vol. 63, No. 10, 1998
Tamai et al.
CP•+ and O
•-, and (ii) reduction of the diffusion constant
Ta ble 5. Effects of P olyp h en ylen e Com p ou n d s on th e
Secon d -Or d er Ra te Con sta n ts for th e Deca y of t-CP
2
•
+
by the increase of the molecular volume. The mechanism
for the retardation of the diffusive back-electron transfer
by the π-complex formation or hetero-dimer radical cation
is currently being studied.
Gen er a ted by th e P h otoin d u ced Electr on Tr a n sfer fr om
1
t-CP to DCN* in Aer a ted Aceton itr ile
10- kCP•+/M-1 s-1 a
9
additive
concn/M
none
BP
o-TP
m-TP
p-TP
Naph
14 ( 2
11 ( 1
7.3 ( 1
9.0 ( 1
9.5 ( 1.5
8.0 ( 1
-
2
2
2
3
2
2.5 × 10
2.5 × 10
2.5 × 10
5.0 × 10
2.5 × 10
-
-
-
-
Con clu sion
We have determined the turnover values for the
photochemical chain reactions of 1,2-bis(4-methoxyphe-
nyl)cyclopropane (CP) radical cation by a comparison
between the quantum yields for the chemical reactions
and the radical cation yield in the absence and presence
of polyphenylene (PP) cosensitizers such as o- and m-
terphenyls. The role of these compounds in the DCA-
CP system is summarized in Scheme 3. These com-
pounds effectively quench the fluorescence of DCA more
than biphenyl giving their radical cation in fairly high
yields and are more stable than phenanthrene or naph-
thalene which showed a similar cosensitization effect.
The turnover depends on the cosensitizer used and not
a
Obtained by the laser flash photolysis of aerated acetonitrile
-3
solutions containing DCN as a sensitizer: [DCN] ) 1 × 10 ; [t-CP]
-2
)
1
2.5 × 10 M. Excitation wavelength: 355 nm. Laser intensity:
-1
5 mJ pulse . The rate constants were calculated from the slope
of the plots of 1/∆OD vs time (see text).
M- cm-1) of t-CP , and a path length (0.6 cm). The
results are listed in Table 5. Though the rate constant
in the absence of PP showed a nearly diffusion-controlled
1
•+
1
0
-1 -1
rate of 1.4 × 10
M
s , the addition of PP decreased
the rate constants by 0.5-0.7. The effects of TP on the
decrease in the rate constants were larger than that of
BP, being similar to those on the increase of turnover
values for both reactions. This could be due to a stronger
•
+
on the initial yield of CP . Though the mechanism
involved has not been completely clarified, an interaction,
•
+
possibly π-complex formation between CP and cosen-
sitizer, must play an important role in stabilizing CP
as a hole carrier.
•
+
interaction of TP with CP than that of BP possibly
owing to the larger π-system of TP as compared with that
of BP. The results could support the retardation of the
•
+
chain termination and strongly suggest the formation of
the π-complex formation between t-CP•+ and PP. The
Ack n ow led gm en t. One of the authors (K.M.) is
grateful for a Grant-in-Aid from the Ministry of Educa-
tion, Science, Sports, and Culture of J apan.
chain termination would be retarded for the π-complex
formation for the reasons (i) the dispersion of the positive
charge which reduces the Coulobmic interaction between
J O971649Y