Bond-Coupled Electron Transfer Reactions: Photoisomerization of Norbornadiene to Quadricyclane

Article abstract of DOI:10.1021/jp992884i

In this paper, we report on the bond coupled electron transfer (BCET) isomerization reaction of norbornadiene (NB) to quadricyclane (Q). We have used several triplet sensitizers to examine electron and energy quenching by NB. In the case of acetophenone, energy transfer generates the detectable ³NB, which rearranges to give ³Q which relaxes to Q. Using chloranil, benzoquinone, and 2,5-dichlorobenzoquinone as triplet sensitizers, the triplet ion pairs are generated, but cannot access either ³NB or ³Q, and so the ion pair undergoes return electron transfer to the ground state. However, with 3,3',4,4'-benzophenonetetracarboxylic dianhydride as the triplet sensitizer, the triplet ion pair is generated, but can undergo BCET to ³Q efficiently, which then relaxes to Q. To our knowledge, this is the first example of an efficient electron-transfer mechanism to photoisomerize NB to Q without the intermediacy of ³NB. Consequently, in the valence isomerization of NB to Q, it is possible to directly access the NB-Q triplet surface by controlling the energy of the triplet ion pair photogenerated, and in effect, the quantum yield can be modulated.

Full text of DOI:10.1021/jp992884i

©
Copyright 1999 by the American Chemical Society
VOLUME 103, NUMBER 51, DECEMBER 23, 1999
LETTERS
Bond-Coupled Electron Transfer Reactions: Photoisomerization of Norbornadiene to
Quadricyclane
†
,†
,†
,‡
A. Cuppoletti, J. P. Dinnocenzo,* J. L. Goodman,* and I. R. Gould*
Center for Photoinduced Charge-Transfer, UniVersity of Rochester, Rochester, New York 14627-0219,
Department of Chemistry, UniVersity of Rochester, Rochester, New York 14627-0216, and
Department of Chemistry, Arizona State UniVersity, Tempe, Arizona 85287
ReceiVed: August 16, 1999; In Final Form: October 27, 1999
In this paper, we report on the bond coupled electron transfer (BCET) isomerization reaction of norbornadiene
(
NB) to quadricyclane (Q). We have used several triplet sensitizers to examine electron and energy quenching
3
by NB. In the case of acetophenone, energy transfer generates the detectable NB, which rearranges to give
3
Q which relaxes to Q. Using chloranil, benzoquinone, and 2,5-dichlorobenzoquinone as triplet sensitizers,
3
3
the triplet ion pairs are generated, but cannot access either NB or Q, and so the ion pair undergoes return
electron transfer to the ground state. However, with 3,3′,4,4′-benzophenonetetracarboxylic dianhydride as
3
the triplet sensitizer, the triplet ion pair is generated, but can undergo BCET to Q efficiently, which then
relaxes to Q. To our knowledge, this is the first example of an efficient electron-transfer mechanism to
3
photoisomerize NB to Q without the intermediacy of NB. Consequently, in the valence isomerization of NB
to Q, it is possible to directly access the NB-Q triplet surface by controlling the energy of the triplet ion pair
photogenerated, and in effect, the quantum yield can be modulated.
2
Introduction
ring to form a 1,3-biradical. In fact, BCET can be significantly
faster than RET and result in high quantum efficiencies for
Return electron transfer (RET) processes in photoinduced
electron transfer reactions are typically viewed as undesirable
energy-wasting steps. Indeed, a considerable amount of work
has been directed toward strategies to increase the efficiencies
of photoinduced electron transfers by decreasing the rate of RET,
or increasing the rates of useful competing processes such as
isomerization. In this paper, we now report on the bond coupled
electron-transfer isomerization reaction of norbornadiene (NB)
to quadricyclane (Q).
The mechanism of valence isomerization of norbornadiene
3
to quadricyclane has been extensively studied. Triplet sensitiza-
3
1
tion of norbornadiene generates its triplet state, NB, which
ion pair separation or follow-up ion radical reactions. Instead
undergoes adiabatic isomerization to the triplet state of quad-
of RET competing with the useful processes, however, an
alternative approach would be to have RET itself do useful work.
We recently reported on such a process with 1,2-diphenyl-
cyclopropane in which bond-coupled electron transfer (BCET)
resulted in cis-trans isomerization via initial cleavage of the
3
ricyclane, Q, which very rapidly decays to its ground state,
4
Scheme 1.
On the basis of our previous work, we postulated that valence
isomerization may also be possible from the triplet ion pair state,
3
•-
•+
3
(
A /NB ), directly to Q via BCET in competition with RET.
†
Herein, we detail our mechanistic investigation of BCET
isomerization of NB using a variety of triplet sensitizers.
University of Rochester.
Arizona State University.
‡
1
0.1021/jp992884i CCC: $18.00 © 1999 American Chemical Society
Published on Web 12/04/1999

11254 J. Phys. Chem. A, Vol. 103, No. 51, 1999
Letters
SCHEME 1
of the sensitizers and the ion pair energies are given in Table
1
. The ion radical pair energies are calculated according to ∆G
+
-
2
)
Eox(NB/NB ) - Ered(A /A) - e /ꢀr using the oxidation
7
potential of NB, 1.54 V vs SCE, the reduction potentials of
1
6
8
A, and the Coulomb term, ∼ 0.2 V in CDCl3.
Laser Flash Photolysis. The rate of quenching of the triplet
state of the sensitizers was measured by nanosecond absorption
spectroscopy. Using the absorption of the triplet state of the
sensitizers, the rate of decay was measured as a function of NB
concentration. The triplet states of all the triplet sensitizers were
quenched by NB at approximately diffusion controlled rates,
9
-1 -1
∼
5 × 10 M s . Quenching of the triplet states of acetophe-
none and BTDA with NB in CDCl3 yielded no observable
transient on either the nanosecond or picosecond time scale.
However, quenching of the CHL triplet (λmax ) 510 nm)
produced a transient (λmax ) 450 nm) with a lifetime < 5 ns.
-
•
We assign this transient to be CHL on the basis of its known
3
9
spectrum, and is formed by electron transfer from NB to CHL.
Discussion
The rate of reaction of NB with triplet acetophenone occurs
3
at diffusion control and generates NB as demonstrated by PAC.
The reaction presumably occurs via energy transfer and not by
sequential electron transfers as the initial transfer to form the
TABLE 1
3
ion pair is endothermic by ∼10 kcal/mol, Table 1. NB has a
a
b
c
∆Get(3sens)d
sensitizer
Φ
E
red
E
T
E
irp
lifetime of ∼7 ns and an energy of 61 kcal/mol and decays to
acetophenone
BTDA
TCB
BQ
DCBQ
CHL
>0.9
>0.9
0.34
0.0
0.0
0.0
-1.85
-0.81
-0.65
-0.51
-0.18
+0.02
73.6
65.9
64.8
53
82.8
58.8
55.1
51.9
44.3
39.7
+9.2
-7.1
-9.7
-1.1
-5.7
-10.3
3
Q with high efficiency presumably via Q which is not observed.
6
These results are in good agreement with those of Caldwell.
3
3
In contrast, the reactions of NB with BQ, DCBQ and
CHL, which occur at diffusion control, do not yield NB as no
3
3
50
50
transient is observed by PAC. In addition, quantum yield
measurements and PAC show that no photoisomerization of NB
a
in V (vs SCE), see ref 16. b kcal/mol, see ref 17. c kcal/mol, see
text. d kcal/mol, ∆Get ) Eirp - E
.
3
-•
to Q occurs. Quenching of CHL by NB does produce CHL ,
which implies initial electron transfer to give the triplet ion pair,
T
-
•
+
Results
CHL /NB . Decay of this ion pair occurs rapidly, <5ns, via
return electron transfer to regenerate NB.
Photoacoustic Calorimetry. Triplet sensitization of NB
The reaction of NB with 3BTDA is significantly more
(0.5M) using acetophenone in CH2Cl2 results in two heat
interesting than with the other sensitizers. Whereas quenching
occurs at diffusion control and yields Q efficiently, it does not
depositions. The first deposition yields a transient with energy
of 61 kcal/mol relative to NB and ground-state acetophenone,
and a lifetime of 7.2 ns. The overall energy retained in the
3
generate NB. So although both triplet energy transfer and
electron transfer are energetically favorable (see Table 1), triplet
energy transfer apparently does not occur. In fact, other
sensitizers with similar triplet energies to BTDA but which
cannot react with NB by electron transfer do not efficiently
5
photoacoustic experiment is 23.3 kcal/mol. These results are
in excellent agreement with those previously obtained by
6
3
Caldwell and suggest the transient is NB, which decays to Q
with quantum efficiency >0.9. The 3Q was not detected,
presumably because of its short lifetime, <1 ns.
10
isomerize NB. We postulate that electron transfer occurs
-•
+
instead to initially form the triplet BTDA /NB ion pair.
Unfortunately, picosecond absorption spectroscopy experiments
Triplet sensitization of NB with 3,3′,4,4′-benzophenone-
tetracarboxylic dianhydride (BTDA), results in only one heat
deposition, with an overall energy storage of 22.5 kcal/mol. This
3
showed the quenching of BTDA (<100 ps, 1M NB), but no
evidence for the formation of the ion pair. However, this can
be explained if the lifetime of the ion pair is short, <100ps. If
the ion pair is indeed formed, its decay must yield Q efficiently,
3
requires that NB is not formed but Q is produced rapidly, < 1
ns, and efficiently. Triplet sensitization of NB using 1,4
benzoquinone (BQ), 2,5-dichloro-1,4-benzoquinone (DCBQ),
and chloranil (CHL) also result in only one heat deposition but
1
2
but how does this occur?
Assuming initial formation of the triplet ion pair, two
reasonable pathways could explain the formation of Q, Scheme
3
no energy is stored which implies that neither NB nor Q are
1
3
formed using these sensitizers. In addition, both NB and these
sensitizers are regenerated which indicates no chemical reaction
occurs.
2.
Pathway (a) involves initial isomerization of NB+• to Q+•
3
followed by return electron transfer to yield either Q or Q
directly. Pathway (b) involves concerted isomerization and return
Quantum Yields Measurements. Quantum yields for isomer-
ization of NB to Q were determined using several triplet
sensitizers and are given in Table 1. For acetophenone and
BTDA, the quantum yields for isomerization are quite high
3
electron transfer to generate Q, which then decays to Q via
intersystem crossing.
Pathway (a) requires that the unimolecular rearrangement of
•
+•
>
0.9. In contrast, BQ, DCBQ, and CHL do not effect isomer-
NB+ to Q must be extremely rapid as given by the lifetime
-
•
+
ization. A quantum yield of 0.34 was obtained using the triplet
state of 1,2,4,5-tetracyanobenzene, TCB. The triplet energies
of the triplet BTDA /NB ion pair, <100 ps. However, this
rearrangement is reported to occur on the microsecond time

Letters
J. Phys. Chem. A, Vol. 103, No. 51, 1999 11255
SCHEME 2
DCBQ were recrystallized and/or sublimed prior to use.
Norbornadiene, quadricyclane, and acetophenone were distilled
prior to use.
Absorption Spectroscopy. Nanosecond laser plash photolysis
experiments were carried out using an excimer (Lambda Physik
Lextra 50, XeCL,308 nm) pumped dye laser(Lambda Physik
LPD 3002, 7ns pulse width) as the excitation source. Tuning
of the excitation wavelength was possible through a dye laser
chamber (Lambda Physik). Transient absorptions were moni-
tored at right angles to excitation with a conventional xenon
lamp, monochromator, photomultiplier tube arrangement. The
signal from the multiplier tube was recorded and digitized with
a Tektronix TDS 620 digitizing oscilloscope and then passed
to a microcomputer for storage and analysis. The excitation
beam energy was typically attenuated to less than 2-3 mJ per
pulse, and appropriate long pass filters were placed on either
side of the sample to prevent analyzing light photolysis.
scale,11 effectively ruling out this pathway. Moreover, the
rearrangement rate should not be dependent on sensitizer. Using
CHL as a triplet sensitizer, in which the ion pair is observed,
no rearrangement occurs. This observation cannot be explained
by pathway (a).
The bond coupled electron transfer pathway (b) can reason-
ably explain all the experimental observations. The triplet state
of Q has not been directly observed and its energy has not been
determined. However, ab initio calculations have predicted its
triplet energy to be 57.5 kcal/mol above NB and 23.6 kcal/mol
Picosecond laser flash photolysis experiments were carried
out using a Continuum model PY-61C-10 Nd:YAG laser, which
operates at 10 Hz and uses active mode locking and cavity
dumping. The laser produces ca. 60 mJ at 1064 nm, which can
be converted into ca. 30 mJ at 532 nm after appropriate
frequency conversion in KDP. The residual 1064 nm is separated
from the harmonics using dichroic mirrors and is directed to
an optical delay. The delay line consists of a solid BK-7
retroreflector mounted on a 4-foot long Velmex model
MB4051K2J-S4 translation stage. The excitation and analyzing
beams pass through Vincent Associates model 225L2A0Z5
shutters before hitting the sample. The two monochromatic
beams that exit the monochromator corresponding to the sample
and the reference signals are directed: (i) for spectra analysis,
toward a CCD model which is connected to the computer for
storage and analysis, (ii) for kinetic analysis, toward Hamamatsu
model HC120-5 PMT. The outputs of the PMT are electroni-
cally filtered and digitized using a LeCroy model 9304 175 MHz
oscilloscope, which is connected to the computer for storage
and analysis.
1
4
3
15
above Q, and it is also known that Q efficiently yields Q.
Consequently, only triplet ion pairs with energies near or above
5
3
7.5 kcal/mol can access Q. Of the triplet sensitizers employed,
only BTDA yields a triplet ion pair with sufficient energy to
3
efficiently produce Q. Sensitizers BQ, DCBQ and CHL produce
3
ion pairs whose energies are significantly below Q and so they
decay by RET with no net observable isomerization of NB. The
-
•
+
3
TCB / NB ion pair, 55.1 kcal/mol, apparently can access Q
in competition with RET. In addition, the short lifetime of the
-
•
+•
BTDA /NB ion pair can be rationalized as occurring by a
spin allowed, slightly exothermic bond coupled electron-transfer
reaction.
Conclusions
In photoinduced electron transfer reactions, RET has always
been viewed as a problem to overcome. However, RET can be
rendered useful by coupling electron transfer to a chemically
productive event involving bond-breaking or bond-making. In
the valence isomerization of NB to Q, it is possible to directly
access the NB-Q triplet surface by controlling the energy of
the photogenerated triplet ion pair and, in effect, the quantum
yield can be modulated.
Photoacoustic Measurements. PAC measurements were
made using a front face irradiation cell with 347 nm excitation
provided by a P. R. A. model LN1000 laser. Data were collected
with a 10 MHz transducer (Panametrics A611S-SB) with a 25
µm shim. The optical density of the transducer wave was
matched to that of the experimental wave. Each measurement
consisted of 50 averaged shots. Ferrocene and 2-hydroxyben-
zophenone were used for the T-waves. CH2Cl2 was used as the
solvent.
We have used several triplet sensitizers to examine electron
and energy quenching by NB. In the case of acetophenone,
3
energy transfer generates the PAC detectable NB, which
3
rearranges to give Q followed by relaxation to Q. Using CHL,
Quantum Yields Measurements. Quantum yields measure-
ments were made using an optical bench consisting of an Oriel
BQ, and DCBQ as triplet sensitizers, the triplet ion pairs are
3
3
generated, but cannot access either NB or Q and so undergo
RET to the ground state. With BTDA as the triplet sensitizer,
electron transfer also generates a triplet ion pair, but this ion
6
2
6002 Universal Lamp housing equipped with an Oriel 6283
00W Hg lamp, connected to an Oriel 68805 power supply,
and mounted on Oriel 11150 high stability optical bench.
Attached in series to the lamp housing was an Oriel 71430
shutter, and an Oriel 6194 Liquid Filter filled with water, and
Schott Glass Technologies UG 11 band-pass filter.
3
pair can undergo efficient BCET to Q, which then relaxes to
Q. To our knowledge, this is the first example of an efficient
electron-transfer mechanism to photoisomerize NB to Q without
3
the intermediacy of NB.
A quartz cuvette equipped with a Teflon stopcock and a
Teflon coated stir bar was charged with 2.0 mL of a solution
of the sensitizer (OD > 1.0), norbornadiene (0.01-0.1 M) and
norbornane (0.1-1 mM) as the internal standard, in CDCl3. The
solution was purged with a stream of argon and photolyzed using
an appropriate band-pass filter. The solution was then analyzed
by GC using an Altech Econo-Cap SE-30 column. The
quadricyclane was quantified using norbornane as an internal
standard after appropriate response factor correction. For TCB
Experimental Section
Materials. Spectroscopic grade solvents were used as re-
ceived. CDCl3 was stored over poly(4-vinylpyridine). Ferrocene,
2-hydroxybenzophenone, 3,3′,4,4′-benzophenone-tetracarboxylic
dianhydride (BTDA), benzophenone, and TCB were recrystal-
lized several times from appropriate solvents. CHL, BQ, and

11256 J. Phys. Chem. A, Vol. 103, No. 51, 1999
Letters
experiments, 1.2 mM NB was used, and the quantum yield was
corrected for interception of the TCB singlet by NB.
2937. (b) Gorman, A. A.; Hamblett, I.; McNeeney, S. P. Photochem.
Photobio. 1990, 51, 145.
2
(11) Ishiguro, K.; Khudyakov, I. V.; McGarry, P. F.; Turro, N. J. J.
Actinometry was performed using Aberchrome 540 (J. Chem.
Soc., Perkin Trans. II 1981, 341). A 2.0 mL solution of
Aberchrome 540 (0.005 M) in toluene was irradiated and the
moles of photoproduct were determined from the change in
optical density at 494 nm and the extinction coefficient (8200
Am. Chem. Soc. 1994, 116, 6933.
(12) Photoisomerization of NB to Q occurs using 9,10-anthraquinone,
AQ, as the sensitizer. The quantum yield for isomerization is high, 0.9.
3
Picosecond absorption experiments show AQ is rapidly quenched by NB,
but the ion pair if formed is not observed. PAC experiments indicate the
-1
-1
formation of an intermediate with a lifetime of ∼7 ns. This is consistent
M cm ). The known quantum yield of photoproduct (0.20)
was used to calculate the photon flux.
3
with the formation of NB via electron transfer from the ion pair or energy
3
transfer directly from AQ. However, the partitioning of the heat depositions
between the fast and slow components in the PAC experiments suggest
3
that NB is not formed with high efficiency (<0.9). The simplest explanation
Acknowledgment. We are grateful to the National Science
Foundation for research support (Grants CHE-9120001 and
CHE-9617264).
that fits the PAC data is that the ion pair formed via initial electron-transfer
3
3
yields both NB and Q directly. This is consistent with the fact that the
3
ion pair energy (∼61 kcal/mol) is close to that of NB and so, unlike BTDA,
-
•
+•
3
the AQ /NB ion pair can form in competition with Q. This does not
occur with BTDA because the ion pair is slightly lower in energy than that
with AQ.
References and Notes
(
1) Gould, I. R.; Farid, S. Acc. Chem. Res. 1996, 29, 522.
(13) A third pathway is plausible which involves the concerted isomer-
ization and return electron transfer directly to Q without the formation of
Q. Although we cannot exclude this pathway, it does requires rearrangement
(2) Karki, S. B.; Dinnocenzo, J. P.; Farid, S.; Goodman, J. L.; Gould,
I. R.; Zona, T. A. J. Am. Chem. Soc. 1997, 119, 431.
3) (a) Bren, V. A.; Dubonosov, A. D.; Minkin, V. I.; Chernoivanov,
3
(
and spin forbidden return electron transfer to occur simultaneously.
V. A. Russ. Chem. ReV. 1991, 60, 451. (b) Jones, G.; Chiang S. H.; Xuan,
P. T. J. Photochem. 1979, 10, 1. (c) Grutsch, P. A.; Kutal, C. J. Am. Chem.
Soc. 1986, 108, 3108.
(
14) Bach, R. D.; Schilke, I. L.; Schlegel. B. H. J. Org. Chem. 1996,
6
1, 4845.
(
4) (a) Hammond, G. S.; Turro, N. J.; Fischer, A. J. Am. Chem. Soc.
(15) (a) Sluggett, G. W.; Turro, N. J.; Roth. H. D. J. Am. Chem. Soc.
1
961, 83, 4674. (b) Murov, S. L.; Cole, R. S.; Hammond, G. S. J. Am.
Chem. Soc. 1968, 90, 2957. (c) Gorman, A. A.; Hamblett, I.; McNeeney
Photchem. Photobiol. 1990, 51, 145.
1995, 117, 9982. (b) Turro, N. J.; Cherry, W. R.; Mirbach, M. F.; Mirbach,
M. J. J. Am. Chem. Soc. 1977, 99, 7388. (c) Roth, H. D.; Manion Schilling,
M. L. J. Am. Chem. Soc. 1981, 103, 7210. (d) Roth, H. D.; Manion Schilling,
M. L.; Jones, G., II J. Am. Chem. Soc. 1981, 103, 1246.
(5) Rogers, D. W.; Choi, L. S.; Girellini, R. S.; Holmes, T. J.; Allinger,
N. L. J. Phys. Chem. 1980, 84, 1810.
(16) (a) Poever, M. E. J. Chem. Soc. 1962, 4540. (b) Mattes, S.; Farid,
(
(
6) Helms, A. M.; Caldwell, R. A. J. Am. Chem. Soc. 1995, 117, 358.
7) Gassman, P. G.; Yamaguchi, R.; Koser, G. F. J. Org. Chem. 1978,
S. Org. Photochem. 1983, 6, 233. (c) Pykhtina, E. V.; Glukhoedov, N. P.;
Karkash, I. E.; Pravednikov, A. N. Zh. Fiz. Khim. 1978, 52, 1217. (d) Meiter,
L.; Zuman, P. CRC Handbook Series in Organic Electrochemistry; CRC
Press: Cleveland, OH, 1976; Vol. 1.
4
3, 4392 and references therein.
8) Arnold, B. R.; Farid, S.; Goodman, J. L.; Gould, I. R. J. Am. Chem.
Soc. 1996, 118, 5482.
9) Gschwind, R.; Haselbach, E. HelV. Chim. Acta 1979, 62, 941.
10) (a) Arai, T.; Oguchi, T.; Wakabayashi, T.; Tsuchiya, M.; Nishimura,
Y.; Oishi, S.; Sakuragi, H.; Tokumaru, K. Bull. Chem. Soc. Jpn. 1987, 60,
(
(
(17) (a) Loutfy, R. O.; Winnik, M. A. J. Phys. Chem. 1978, 82, 1304.
(b)Scaiano, J. C. Handbook of Organic Photochemistry; CRC Press: Boca
Raton, FL, 1989; Vol. 1.
(