Direct observation and kinetic characterization of o-quinodimethane and its radical cation variant generated in a photoinduced electron-transfer reaction of 1,2-bis(α-styryl)benzene

Article abstract of DOI:10.1016/j.tetlet.2005.01.124

Nanosecond time-resolved absorption spectroscopy on the laser flash photolysis of 1,2-bis(α-styryl)benzene (1) under N-methylquinolinium tetrafluoroborate-toluene-sensitized conditions in acetonitrile confirmed that an o-quinodimethane radical cation (2⁺, λ_max = 569 nm) decayed and the corresponding neutral prototype (2, λ_max = 444 nm) rose with rate constants of 5.6 and 5.9 × 10⁵ s ^-1, respectively, showing the first agreement in kinetics between a reactive radical cation intermediate intervening in chemical reaction and the corresponding neutral species formed by back electron transfer.

Full text of DOI:10.1016/j.tetlet.2005.01.124

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 1831–1835
Direct observation and kinetic characterization of
o-quinodimethane and its radical cation variant generated in a
photoinduced electron-transfer reaction of 1,2-bis(a-styryl)benzene
Hiroshi Ikeda,a,* Teruyo Ikeda,a Megumi Akagi,a Hayato Namai,a Tsutomu Miyashi,a
Yasutake Takahashib and Masaki Kamatac
aDepartment of Chemistry, Graduate School of Science, Tohoku University, Sendai 980-8578, Japan
bDepartment of Chemistry for Materials, Faculty of Engineering, Mie University, Tsu, Mie 514-8507, Japan
cDepartment of Chemistry, Faculty of Education and Human Science, Niigata University, Ikarashi, Niigata 950-2181, Japan
Received 17 December 2004; revised 20 January 2005; accepted 21 January 2005
Abstract—Nanosecond time-resolved absorption spectroscopy on the laser flash photolysis of 1,2-bis(a-styryl)benzene (1) under
N-methylquinolinium tetrafluoroborate–toluene-sensitized conditions in acetonitrile confirmed that an o-quinodimethane radical
cation (2Å+, kmax = 569 nm) decayed and the corresponding neutral prototype (2, kmax = 444 nm) rose with rate constants of 5.6
and 5.9 · 105 sꢀ1, respectively, showing the first agreement in kinetics between a reactive radical cation intermediate intervening
in chemical reaction and the corresponding neutral species formed by back electron transfer.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Previously, we reported on preliminary results of an
electrocyclization reaction of 1,2-bis(a-styryl)benzene
(1, Scheme 1, anodic peak potential = +1.51 V vs SCE
in acetonitrile) to o-quinodimethane (2), triggered by
photoinduced electron transfer (PET) using 9,10-di-
cyanoanthracene (DCA) or 1,4-dicyanonaphthalene
(DCN) as a sensitizer.1 Although the reaction is sup-
posed to proceed via two radical cations, 1Å+ and 2Å+, di-
rect observation of them by laser flash photolysis (LFP)
was not achieved. If 2 is stable under the LFP conditions
used, it becomes easy to analyze kinetic data and allows
us to study the kinetics for each process, especially the
back electron-transfer (BET) process, converting 2Å+ to
2. Such a study will be of value, because the kinetics
for the BET process involving reactive radical cation
intermediates intervening in chemical reactions and the
Ph
Ph
CN
CN
Sens.*
Sens.*
NC
1•+
2
•
Ph
kFN
+
BET
CN
1
kdecay(2•+)
Ph
Ph
2
3
Ph
1O2
Ph
and
krise(2)
4
1
4
Ph
2•+
3O2
Sens.*
O
3•+
O
Ph
Ph
3
5
Scheme 1. PET reactions of 1 under various conditions.
Keywords: Photochemistry; o-Quinodimethane; Radical cation; Back electron transfer; Kinetics.
*
Corresponding author. Fax: +81 22 217 6557; e-mail: ikeda@org.chem.tohoku.ac.jp
0040-4039/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.01.124

1832
H. Ikeda et al. / Tetrahedron Letters 46 (2005) 1831–1835
100
100
1
2
3
50
00
0
0
2
2
4
4
Time / h
Figure 1. Time-dependent changes of the relative yields of products in
the DCA-sensitized PET reaction of 1 in degassed dichloromethane-d2.
[1] = 50 mM.
corresponding neutral compounds is unprecedented in
spite of its great importance.2 For this purpose, we
studied nanosecond time-resolved absorption spectros-
copy on LFP and kinetics for the intermediates gener-
ated in the PET reactions of 1. Herein, we report on a
direct observation of 2Å+ and the first agreement of the
decay and rise rate constants of 2Å+ and 2, respectively.
Upon irradiating (80 W Rayonet lamp, 350 nm) DCA
for 15 min in degassed dichloromethane-d2 at 20 ꢁC, 1
produced 2 as the sole product in 27% yield (Fig. 1).
Similar irradiation for 2 h gave a dark-yellow mixture
of 1, 2, and 1,2-dihydro-1,4-diphenylnaphthalene (3),3
in 43%, 26%, and 31% yields, respectively. The reaction
mixture containing 2 was stable in the dark for a full
day, but readily converted to 3 in 87% yield after pro-
longed irradiation for 5 h. These findings indicate that
3 is formed secondarily and slowly by PET reaction of
2. A similar reaction also proceeded in degassed benzene
and acetonitrile. Interestingly, 2 was unstable in an aer-
ated solution in the dark, resulting in decomposition to
give a complex mixture.
Figure 2. (a) Time-resolved absorption spectra observed in the LFP of
1 under NMQþBFꢀ4 –toluene-sensitized conditions in aerated acetoni-
trile. [1] = 10 mM, ½NMQþBF4ꢀŠ ¼ 1 mM, [toluene] = 2 M. (b) UV–vis
absorption spectra observed during a DCN-sensitized PET reaction of
1 in degassed acetonitrile. [1] = 10 mM, [DCN] = 0.02 mM. Inset: a
photograph of the sample solution after irradiation for 30 s. (c)
Absorption spectra observed during the annealing (77 to ꢁ130 K) of
an n-butyl chloride glassy matrix of 1 irradiated with c-rays at 77 K.
[1] = 5 mM.
Nanosecond time-resolved absorption spectroscopy
upon LFP4 was performed with N-methylquinolinium
tetrafluoroborate ðNMQþBFꢀ4 Þ and toluene as a sensi-
tizer and a co-sensitizer, respectively.5 As shown in Fig-
ure 2a, laser excitation (355 nm) of NMQþBF4ꢀ with 1 in
aerated acetonitrile gave two transient absorption bands
with kmax at 444 and 569 nm. The differential optical
density (DOD) of the 444 nm band increased with
decreasing DOD of the 569 nm band. Similar absorption
spectra were observed under NMQþBF4ꢀ–toluene-sensi-
tized conditions in degassed dichloromethane and aceto-
nitrile, or under DCA–biphenyl (BP, co-sensitizer)-
sensitized conditions in degassed acetonitrile. The
444 nm band is assigned to 2 because this band corre-
sponds to the band with kmax at 442 nm of 2 observed
for a yellow solution (Fig. 2b and inset) of the DCN-sen-
sitized PET reactions (80 W Rayonet lamp, 300 nm) of 1
in acetonitrile.6 Conversely, the 569 nm band is attrib-
uted to 2Å+, but not to 1Å+, because the radical cation
of 1,1-diphenylethylene, a subunit of 1, did not have
any intense absorption bands at this region in an n-butyl
chloride glassy matrix irradiated with c-rays at 77 K.7
Interestingly, an absorption band with kmax at 578 nm
appeared in a similar matrix of 1 at 77 K, and disap-
peared with an appearance of 2 observed at ꢁ440–
450 nm upon annealing (77 to ꢁ133 K) (Fig. 2c). No
absorption band due to the precursor radical cation
1Å+ was observed upon LFP or c-ray irradiation. This
may be due to the fixed cisoid form of two styrene units
in 1, which allows 1 to cyclize rapidly upon oxidation.
The assignments are in accord with the decay and rise
behaviors of the observed species (viz. 2Å+ and 2) with
an isosbestic point at k = 494 nm (Fig. 2a).
A kinetic analysis for 2Å+ and 2 was achieved under
NMQþBFꢀ4 (2 mM) and toluene (0.4 M) conditions in
degassed acetonitrile containing 1 (0.3 mM) at 25 ꢁC
(Fig. 3).4 The decay rate constant for 2Å+ observed at
570 nm, kdecay(2Å+) = 5.6 · 105 sꢀ1, agrees with the rise
rate constant for
2
observed at 461 nm,
krise(2) = 5.9 · 105 sꢀ1, within the experimental error
(ꢁ5%), confirming that 2Å+ is quantitatively converted
to 2 by single-electron reduction, probably BET from
NMQÅ. To the best of our knowledge, this is the first
agreement of the decay and rise rate constants of a reac-
tive radical cation intervening in chemical reaction and
the corresponding species, respectively. Note that
kdecay(2Å+) and krise(2) are not the net values of the rate

H. Ikeda et al. / Tetrahedron Letters 46 (2005) 1831–1835
1833
constant, kbet, of BET from NMQÅ to 2Å+ in a radical ion
Ph
Ph
Ph
Ph
Ph
pair [2Å+//NMQÅ]rip, which was theoretically calculated
•
•
O
to be 5.2 · 1010 sꢀ1
,
by the latest electron-transfer
8
O
theory.9 Given the steady state approximation for
[2Å+//NMQÅ]rip generated in the BET sequence shown
in Scheme 2, kdecay(2Å+) and krise(2) are represented by
Eq. 1, where kass and kdis are the rate constants
Ph
8••
6
7
Chart 1.
for the association and dissociation for [2Å+//NMQÅ]rip
,
and
and are estimated to be 1.9 · 1010 Mꢀ1 sꢀ1
,
12
2.4 · 109 sꢀ1
,
respectively. A numerical constant k0 is
oxygen (1O2)1 but also the mechanism involving 2Å+ and
13
ꢀ1.6 · 105 sꢀ1
.
These assumptions conclude that the
3O2 contributed to the DCA-sensitized PET oxygenation
14
concentration of NMQÅ, [NMQÅ], is ꢁ4 · 10ꢀ2 mM,
which is reasonable, because the initial concentration
of NMQþBFꢀ4 is 2 mM. Consequently, the findings de-
scribed here indicate that the kinetic analysis used in this
work is quite appropriate in light of the latest electron-
transfer theory.9
giving 5. Probably, 2Å+ reacts slowly with O2 to give 5.
3
This may be due to the nondistonic nature of 2Å+; the spin
and charge are delocalized to the long p-conjugated sys-
tem of 2Å+. Needless to say, the spin densities at C-1 and
C-4 are not very high compared with those of distonic
radical cations that afford oxygenation products
efficiently.17
kasskbet
kdis þ kbet
Å
kdecayð2þÞ and kriseð2Þ ¼
½NMQ Š þ k0 ð1Þ
On the other hand, the cycloaddition reaction of 2 with
FN, giving 4, proceeded with the rate constant,
kFN = 1.9 · 10ꢀ2 Mꢀ1 sꢀ1, at 25 ꢁC in degassed benz-
ene.18,19 Interestingly, 1,4-diphenylcyclohexa-1,3-diene
(6, Chart 1) did not react with FN in benzene at
80 ꢁC. Moreover, the decay rate constant for 8ÅÅ, which
is formed from a structurally related furan derivative
7, with FN, is 1.2 · 107 Mꢀ1 sꢀ1 at 25 ꢁC in dichloro-
methane under nitrogen.20 The high reactivity of 2 com-
pared with 6 is reasonably ascribed to the recovery of
aromaticity in 4. Conversely, the slow reactivity of 2
compared with 8ÅÅ is ascribed to the nonradical nature
of 2.
We also investigated the reactivity of 2Å+ and 2 toward
molecular oxygen (3O2) and fumaronitrile (FN) by
absorption spectroscopy in connection with the DCA-
sensitized PET oxygenation of 1, giving 5,1 and the PET
cycloaddition reaction of 1 with FN (Scheme 1).1 We
found that under DCA–BP-sensitized conditions in ace-
tonitrile, DOD of 2Å+ was not changed very drastically,
depending upon the concentration of 3O2 on the timescale
of LFP,15 whereas the change in DOD of 2 was not exam-
ined because the absorption bands of BPÅ+ and 2 overlap
each other. However, 5 was formed in 55% yield at 81%
conversion in the tris(4-bromophenyl)aminium hexa-
ꢀ
Åþ
In conclusion, we directly observed an o-quinodime-
thane radical cation 2Å+, which intervenes in the PET
electrocyclization of 1 to 2. Kinetic analyses for 2Å+
and 2 confirmed for the first time that the decay rate
constant of a reactive radical cation intervening in
chemical reaction agrees with the rise rate constant of
the corresponding reduced species formed by BET. A ki-
chloroantimonate ½ð4ꢀBrC6H4Þ N SbCl6 Š (0.4 mM)-
3
catalyzed oxygenation of 1 (2 mM) under oxygen in
dichloromethane (50 mL) in the dark, clearly indicating
that 3O2 added to radical cation,16 2Å+. These findings sug-
gest that not only the mechanism involving 2 and singlet
netic study of 2Å+ toward O2 and of 2 toward FN also
3
revealed the kinetic characteristics of 2 and 2Å+. Further
studies are now in progress, and will be published
elsewhere.
kass
kdis
Scheme 2. A sequence of BET from NMQÅ to 2Å+
kbet
[2•+//NMQ•]rip
2 + NMQ+
2•+ + NMQ•
.
Acknowledgments
H.I. and M.K. gratefully acknowledge financial support
by a Grant-in-Aid for Scientific Research on Priority
Areas (No. 417) from the Ministry of Education, Cul-
ture, Sports, Science, and Technology of Japan and by
The Uchida Energy Science Promotion Foundation,
respectively. We also thank Dr. T. N. Inada (Kitasato
University) for helpful discussions and Professor M.
Ueda (Tohoku University) for his generous
considerations.
461 nm
570 nm
0.2
0
OD
0
10
Time / µs
15
5
Figure 3. Time-profiles of DOD of 2Å+ and 2 monitored at 570 and
461 nm, respectively, in the LFP of 1 under NMQþBF4ꢀ–toluene
conditions in degassed acetonitrile. [1] = 0.3 mM, ½NMQþBF4ꢀŠ ¼
2 mM, [toluene] = 0.4 M. Data from 0.76 to 7.8 ls were used for
kinetic analyses. See notes 4 and 5.
Supplementary data
An energy diagram for NMQþBFꢀ4 -sensitized PET reac-
tion of 1. Supplementary data associated with this

1834
H. Ikeda et al. / Tetrahedron Letters 46 (2005) 1831–1835
ꢀ
ꢁ
1=2
4p3
article can be found, in the online version, at
doi:10.1016/j.tetlet.2005.01.124.
kbet
¼
h2kskbT
"
#
ꢀ
ꢁ
1
2
ꢀssw
ðks þ DGbet þ whmÞ
4kskbT
X
e
2
j V j
exp ꢀ
ð2Þ
ð3Þ
w!
w¼0
References and notes
1. Takahashi, Y.; Ohya, Y.; Ikeda, H.; Miyashi, T. J. Chem.
Soc., Chem. Commun. 1995, 1749–1750.
S ¼ kv=hm
2. For the importance of BET in the PET reactions, see: (a)
Wong, P. C.; Arnold, D. R. Tetrahedron Lett. 1979, 2101–
2104; (b) Kumar, C. V.; Chattopadhyay, S. K.; Das, P. K.
J. Chem. Soc., Chem. Commun. 1984, 1107–1109; (c)
Ikeda, H.; Minegishi, T.; Miyashi, T. J. Chem. Soc., Chem.
Commun. 1994, 297–298; (d) Karki, S. B.; Dinnocenzo, J.
P.; Farid, S.; Goodman, J. L.; Gould, I. R.; Zona, T. A. J.
Am. Chem. Soc. 1997, 119, 431–432; (e) Ikeda, H.;
Minegishi, T.; Abe, H.; Konno, A.; Goodman, J. L.;
Miyashi, T. J. Am. Chem. Soc. 1998, 120, 87–95; (f) Ikeda,
H.; Nakamura, T.; Miyashi, T.; Goodman, J. L.; Akiy-
ama, K.; Tero-Kubota, S.; Houmam, A.; Wayner, D. D.
M. J. Am. Chem. Soc. 1998, 120, 5832–5833; (g) Ikeda, H.;
Takasaki, T.; Takahashi, Y.; Konno, A.; Matsumoto, M.;
Hoshi, Y.; Aoki, T.; Suzuki, T.; Goodman, J. L.; Miyashi,
T. J. Org. Chem. 1999, 64, 1640–1649; (h) Miyashi, T.;
Ikeda, H.; Takahashi, Y. Acc. Chem. Res. 1999, 32, 815–
824; (i) Roth, H. D. J. Photochem. Photobiol. C 2001, 2,
93–116; (j) Ikeda, H.; Akiyama, K.; Takahashi, Y.;
Nakamura, T.; Ishizaki, S.; Shiratori, Y.; Ohaku, H.;
Goodman, J. L.; Houmam, A.; Wayner, D. D. M.; Tero-
Kubota, S.; Miyashi, T. J. Am. Chem. Soc. 2003, 125,
9147–9157.
where the parameters V, ks, kv, m, and DGbet are, respec-
tively, the electronic coupling matrix element (57 cmꢀ1),
solvent reorganization energy (1.52 eV), vibration reorga-
nization energy (0.25 eV), single average frequency
(1500 cmꢀ1), and free energy change for BET process
(ꢀ1.03 eV).11 In addition, h, kb, and T are PlanckÕs con-
stant, BoltzmannÕs constant, and the temperature
(298 K), respectively.
9. (a) Miller, J. R.; Beitz, J. V.; Huddleston, R. K. J. Am.
Chem. Soc. 1984, 106, 5057–5068; (b) Siders, P.; Marcus,
R. A. J. Am. Chem. Soc. 1981, 103, 741–747; (c) Siders, P.;
Marcus, R. A. J. Am. Chem. Soc. 1981, 103, 748–752; (d)
Van Duyne, R. P.; Fischer, S. F. Chem. Phys. 1974, 5,
183–197; (e) Ulstrup, J.; Jortner, J. J. Chem. Phys. 1975,
63, 4358–4368.
10. Inada, T. N.; Miyazawa, C. S.; Kikuchi, K.; Yamauchi,
M.; Nagata, T.; Takahashi, Y.; Ikeda, H.; Miyashi, T. J.
Am. Chem. Soc. 1999, 121, 7211–7219.
11. The value of DGbet = ꢀ1.03 eV was estimated by analyses
for an energy diagram for the NMQþBF4ꢀ-sensitized PET
reaction of 1. See the Supplementary data.
3. Abegg, V. P.; Hopkinson, A. C.; Lee-Ruff, E. Can. J.
Chem. 1978, 56, 99–103.
4. Nanosecond time-resolved absorption spectroscopy
upon LFP was carried out with a pulsed YAG laser
12. The value of kass is supposed to correspond to the diffusion
at
rate
constant
in
acetonitrile
25 ꢁC,
kdif = 1.9 · 1010
M
ꢀ1 sꢀ1
.
13. The value of kdis for [2Å+//NMQÅ]rip with DGbet = ꢀ1.03 eV
(Continuum Surelite-10, Nd, THG,
55 mJ) and a Xe arc lamp (150 W) as the monitoring
kex = 355 nm,
is supposed to correspond to that for [anisidine radical
DGbet =
cation//N-methylacridinium
radical]
10
with
ꢀ1.06 eV and kdis = 2.4 · 109 sꢀ1
.
light. A cutoff filter (Y-46, k > 435 nm) was used for
2 so as to avoid
the kinetic analysis of 2Å+ and
14. The value of k0 was determined as the intercept of a linear
relationship between kdecay(2Å+) or krise(2) and the absorp-
tion efficiency of NMQ+ (100-transmittance) on the
assumption that [NMQÅ] is proportional to the absorption
efficiency of NMQ+.
possible side reactions of 1 and 2 initiated by absorp-
tion of the monitoring light. Reactions of 1 affording 2
were also observed under these co-sensitized PET
conditions.
15. The values of half lifetime (s1/2) of 2Å+ under oxygen, air,
and degassed conditions in acetonitrile were 1.7, 1.6, and
1.6 ls, respectively.
5. A use of cationic sensitizers and chemically unreac-
tive aromatic hydrocarbons so-called co-sensitizers is
effective to observe radical cations in the LFP experi-
ments. Under these conditions,
cation is formed via electron transfer from
sensitizer to photoexcited sensitizer followed by
a
substrate radical
16. Barton, D. H. R.; Haynes, R. K.; Leclerc, G.; Magnus, P.
D.; Menzies, I. D. J. Chem. Soc., Perkin Trans. 1 1975,
2055–2065.
a
co-
a
hole transfer from a co-sensitizer radical cation to a
substrate.
6. Because DCN is transparent in the vis region, DCN was
used as a sensitizer instead of DCA in the monitoring
experiment of the time-dependent changes of absorption
spectra of the PET reaction of 1.
7. A n-butyl chloride solution containing 1 (5 mM) was
degassed by repeating five freeze (77 K)–pump
(10ꢀ3 mmHg)–thaw (ambient temperature) cycles and
sealed at 10ꢀ3 mmHg at 77 K. This matrix was irradiated
at 77 K for 40 h with c-rays from a 4.0 TBq 60Co source at
17. For recent examples of spectroscopic studies of chemical
capturing distonic radical cation intermediates with 3O2,
see Ref. 2e–g,j, and: Mizuno, K.; Tamai, T.; Hashida, I.;
Otsuji, Y.; Kuriyama, Y.; Tokumaru, K. J. Org. Chem.
1994, 59, 7329–7334.
18. The DOD of 2Å+ did not significantly change before or after
adding FN (0.185 mM) in the LFP experiment of 1
(0.3 mM) under NMQþBF4ꢀ–toluene-sensitized conditions
in dichloromethane.
19. A benzene solution containing 1 (1 mM), DCA (0.5 mM),
and FN (0–100 mM) was degassed by repeating five
freeze (77 K)–pump (10ꢀ3 mmHg)–thaw (ambient temper-
ature) cycles and sealed at 10ꢀ3 mmHg at 77 K. After
irradiation for 5 min with a Rayonet lamp (15 W, 350 nm)
at 20 ꢁC, the time-dependent change of the absorbance at
480 nm of the sample solution was monitored at constant
the Cobalt 60
c Ray Irradiation Facility, Tohoku
University.
8. The value of kbet for the BET from NMQÅ to 2Å+ at 25 ꢁC
in acetonitrile was calculated using the following Eqs.9 2
and 3 and parameters reported by Kikuchi and co-
workers10
temperature with
a
JASCO V-570 UV–vis/NIR

H. Ikeda et al. / Tetrahedron Letters 46 (2005) 1831–1835
1835
spectrophotometer equipped with ETC-505 Peltier-type
thermostatic cell holder. The apparent rate constant, kap
was given as the slope of a linear relationship between
absorbance and time. Plotting kap and [FN] gave kFN at a
temperature as the slope. Similar experiments were
performed at 15.1, 20.0, 22.3, 24.7, 27.1, and 29.4 ꢁC.
Arrhenius plot analysis for kFN and the theory of absolute
reaction rate finally afforded the following kinetic
,
parameters; Ea = 9.3 kcal molꢀ1
,
A = 1.3 · 105 Mꢀ1 sꢀ1
,
DGà(25 ꢁC) = 19.8 kcal molꢀ1, DHà(25 ꢁC) = 8.7 kcal molꢀ1
,
DSà(25 ꢁC) = ꢀ37.1 cal molꢀ1 Kꢀ1
.
20. Yamada, M. Master Thesis, Tohoku University, 1995.