Catalytic photocarboxylation of 1,1-diphenylethylene with N,N,N′,N′-tetramethylbenzidine and carbon dioxide

Article abstract of DOI:10.1016/j.tet.2007.02.013

Photocarboxylation of 1,1-diphenylethylene with N,N,N′,N′-tetramethylbenzidine (TMB) in MeCN under bubbling of CO₂ proceeded with high catalytic efficiency, giving 3,3-diphenylacrylic acid (DPA) and 3-hydroxy-3,3-diphenylpropionic acid (20). The turnover number (TON=(DPA+20)/TMB) reached 17. Similarly, 1-phenyl-1-cyclohexene yielded cis-2-acetamido-2-phenylcyclohexanecarboxylic acid with TON 5.9. As compared with related N,N-dimethylaniline derivatives, TMB is more resistant to photodecomposition, has the much larger absorbance in the S₀→S₁ transition, and has the lower quenching efficiency by CO₂. Probably these factors are partly responsible for the high TON observed for TMB.

Full text of DOI:10.1016/j.tet.2007.02.013

Tetrahedron 63 (2007) 3108–3114
Catalytic photocarboxylation of 1,1-diphenylethylene with
0
0
N,N,N ,N -tetramethylbenzidine and carbon dioxide
*
Yoshikatsu Ito
Department of Synthetic Chemistry and Biological Chemistry, Graduate School of Engineering, Kyoto University,
Katsura, Kyoto 615-8510, Japan
Received 15 January 2007; revised 5 February 2007; accepted 6 February 2007
Available online 9 February 2007
0
0
Abstract—Photocarboxylation of 1,1-diphenylethylene with N,N,N ,N -tetramethylbenzidine (TMB) in MeCN under bubbling of CO pro-
2
ceeded with high catalytic efficiency, giving 3,3-diphenylacrylic acid (DPA) and 3-hydroxy-3,3-diphenylpropionic acid (20). The turnover
number (TON¼(DPA+20)/TMB) reached 17. Similarly, 1-phenyl-1-cyclohexene yielded cis-2-acetamido-2-phenylcyclohexanecarboxylic
acid with TON 5.9. As compared with related N,N-dimethylaniline derivatives, TMB is more resistant to photodecomposition, has the
much larger absorbance in the S /S transition, and has the lower quenching efficiency by CO . Probably these factors are partly responsible
0
1
2
for the high TON observed for TMB.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
Nearly two decades ago we reported that 1,1-diphenylethyl-
ene (DPE, 18) underwent photolysis in MeCN in the
A variety of methods were performed toward photofixation
of carbon dioxide into organic compounds (photocarboxyl-
ation): for example, (a) metalloporphyrins and other
organometallic or metal complexes, which cause insertion,
cycloaddition, or carboxylation, (b) particular semiconduc-
presence of N,N-dimethyl-p-toluidine (DMT, 7) and CO to
2
yield 3,3-diphenylacrylic acid (DPA, 19) with the DPA/
1
2
8e
8
DMT ratio of 0.16. Inspection of the reported papers has
suggested that the radical anion of DPE appears to satisfy
most the aforementioned requisite. To develop a new useful
3
tors and polymers, which usually fix CO through reduction
2
CO photofixation system, therefore, the author resumed
2
ꢀ
ꢁ 4
to CO , (c) photocatalysts or photosensitizers that are con-
jugated with enzyme-catalyzed reactions, and (d) trapping
by reactive species such as 1,3-dipolar species, carbenes,
this investigation by using a series of para-substituted N,N-
dimethylanilines (DMAs) as sensitizer (Eq. 1). The para-
substitution precludes the known para-coupling, which
2
5
6
7
8
8e
0
0
and radical anions. Among these, catalytic reactions seem
to be most attractive to realize the practical photofixation. In-
deed, high turnover numbers (TONs) were reported for the
consumes DMA. Here N,N,N ,N -tetramethylbenzidine
(TMB, 10) is demonstrated to be a respectable photocatalyst.
5
c
malic acid formation (TON 1074) and the malonic acid
derivative formation (TON 93), although their reaction
systems are complex. Traditional organic photochemistry
NMe2
2
a
hν
Ph
Ph
Ph
Ph
Ph
CO
2
+
+
HO
MeCN
CO H
2
CO H
2
6
–8
Ph
20
might not look very promising in this field.
Irradiation
X
X
of unsaturated hydrocarbon–amine systems, where the amine
is used as electron donor or sensitizer, can efficiently gener-
ate radical anions from the former, but unfortunately the mo-
lar ratio between the carboxylated product versus the amine
DPE (18)
DPA (19)
NMe2
NMe2
NO2
(1)
(2)
(3)
(4)
Ph
CN
8
HO
CH3
is very low (<0.1) in most cases. Radical anions, which
have enough nucleophilicity to attack CO , form a stable
ð1Þ
COMe
COOMe
Cl
Ph
2
Ph Ph
CH CO H
2 2
bond to CO , and end up with final products with high reac-
2
21
(5)
Ph
8
g,9
Ph
tion selectivity, are necessary.
H
DMA (6)
DMT (7)
(8)
23
22
Me
OMe
NMe2
C H NMe
0
0
Keywords: N,N,N ,N -Tetramethylbenzidine; Carbon dioxide; Catalytic
photocarboxylation; High turnover number.
TMPD (9)
TMB (10)
6
4
2
*
Tel./fax: +81 75 383 2718; e-mail: yito@sbchem.kyoto-u.ac.jp
0
040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.02.013

Y. Ito / Tetrahedron 63 (2007) 3108–3114
3109
NMe2
CH2
NMe2
Ph
Ph
N
HC
NMe2
NMe2
NMe2
Me
N
N
Me
N
NMe2
NMe2
11
12
13
14
15
16
17
2
. Results and discussion
calculated. This ratio may be called the turnover number
TON) for the photocarboxylation. As summarized in Table
(
2
.1. Photocarboxylation of DPE
2, higher TONs were obtained by decreasing the concentra-
tion of TMB relative to DPE. Notably, at DPE/TMB¼48
([TMB]¼4.6 mM, [DPE]¼0.22 M), the TON reached 17
(run 5). The major products DPA and 20 could be easily iso-
lated, as described in Section 4. The quantum yield for the
formation of DPA under the conditions of run 5 was esti-
Neither the DMA derivatives having an electron-withdraw-
ing para-substituent 1–4 and 13 nor the DMA analogs 14
and 15 photocarboxylated DPE upon photolysis in DMSO-
d6. As shown in Table 1, however, the DMA derivatives
without a strong electron-withdrawing group 5–12 gave
DPA (DPA/amine>0.06), although the main product in the
case of 5 was 22 formed via a triplet phenyl cation, and
the major reaction by DMA (6) in MeCN was para-coupling
to give 23, thereby consuming DMA. The reaction by
N,N,N ,N -tetramethyl-p-phenylenediamine (TMPD, 9) was
considerably solvent dependent. In fact, as seen from
the results for 5–9 (runs 1–11), the DPA formation is always
more efficient in DMSO or DMF than in MeCN. For TMB
1
1
mated to be 0.018 at 313-nm irradiation. This is as good
as those for the previous high TON reactions (F 0.017
and 0.019 ).
2
a
1
0
5c
8
e
2.2. Possible pathways for photocarboxylation of DPE
by TMB
0
0
Possible reaction pathways are described in Scheme 1. The
8
e
previously proposed mechanism was slightly modified.
First the photoinduced electron transfer occurs from the
TMB excited state. Since the reduction potentials of DPE
(
10), however, MeCN was a better solvent than DMF. The
yield of DPA in MeCN (71%, run 13) overwhelms others.
The finding that DPA/amine¼1.5 is surprising, because the
corresponding ratios were <0.1 in the earlier cognate photo-
1
2
13
(about ꢁ2.3 V ) and CO (about ꢁ2.2 V ) are comparable,
2
ꢀ
–
ꢀ
–
both DPE and CO can be generated, leading to formation
2
8
carboxylation.
of the distonic radical anion A according to step a or step b,
respectively. Similar electron transfer paths have already
been demonstrated to occur, e.g., see Ref. 8b for step a
and Refs. 3a and 4b for step b. After regeneration of TMB
by back-electron transfer, the resulting zwitterion B will
be transformed to DPA or, in the presence of impurity water,
to the hydration product 20. Because DPA was slowly con-
verted to 20 under similar photolysis conditions, a part of
The very encouraging result found by TMB led the author
to vary the molar ratio between DPE and TMB with the
intention of raising the DPA/amine ratio. Since considerable
production of 3-hydroxy-3,3-diphenylpropionic acid (20)
and 1,1-diphenylethanol (21) was observed, their yields
were also determined and the ratio (DPA+20)/TMB was
a
Table 1. Photocarboxylation of 1,1-diphenylethylene (DPE) with N,N-dimethylaniline derivatives 5–12
b
Run
Amine
Solvent
Products, %
Recovery, %
Amine
DPA/amine
DPA
Others
DPE
1
2
3
4
5
6
7
8
9
5
DMSO
MeCN
DMSO-d
MeCN
DMSO-d
MeCN
DMSO-d
MeCN
DMSO-d
DMF-d
MeCN
DMF-d
MeCN
DMSO-d
DMSO-d
10
6
22, 18
22, 16
23, Trace
23, 40
—
—
—
—
—
—
—
—
70
75
24
45
39
75
63
69
69
88
100
24
9
72
70
53
53
81
93
100
92
91
100
95
90
80
98
98
0.098
0.063
0.21
0.12 (0.005 )
0.27
0.17 (0.16 )
DMA (6)
DMT (7)
8
6
6
6
6
21
12
26
17
25
9
28
9
w0
43
71
7
c
c
0.25
0.093
0.28
0.081
w0
0.94
1.5
TMPD (9)
10
11
12
13
14
15
7
TMB (10)
7
d
20, 8; 21, 1
—
—
11
12
6
30
15
0.073
0.084
6
9
a
A solution containing 0.13 M of amine (except TMB) and 0.13 M of DPE was irradiated under bubbling CO
filter, >290 nm) for 10 h. In the case of TMB, a solution containing (a) 0.065 M of TMB and 0.13 M of DPE in DMF-d
2
with a 400-W high pressure mercury lamp (Pyrex
or (b) 0.018 M of TMB and 0.037 M
7
of DPE in MeCN was irradiated. Product yields and recoveries were estimated by NMR and HPLC analyses of the photolysates.
Yields are based on the initially used DPE.
The isolation yield of DPA was used to calculate the ratio.
b
c
d
8e
Benzophenone (0.6% yield).

3110
Y. Ito / Tetrahedron 63 (2007) 3108–3114
0
0
Table 2. Photocarboxylation of 1,1-diphenylethylene (DPE) with N,N,N ,N -tetramethylbenzidine (TMB) in MeCN
a
a
b
Run
[TMB], mM
[DPE], M
DPE/TMB
Carboxylated products, %
Noncarboxylated products, %
Recovery, %
(DPA+20)/TMB
DPA
20
21
Others
DPE
TMB
c
c
c
c
c
c
c
1
2
3
4
5
6
7
18
18
9.0
4.6
4.6
16, 4.6
17, 4.6
0.037
0.11
0.11
0.11
0.22
0.11
0.11
2
6
71
47
44
34
22
10
12
8
4
6
13
14
8
1
0
2
9
10
3
9
38
35
33
48
51
66
80
77
54
34
20
16, 3
17, 0
1.6
3.1
6.0
11
17
4.5
4.1
12
24
48
24
24
5
2
a
Yields are based on the initially used DPE.
The molar ratio between the carboxylated products (DPA+20) versus the initially used TMB (or 16 or 17).
A trace amount of benzophenone was formed (0.3–1% yield).
b
c
Ph C CH2
2
1
2
Me N
NMe2
Ph C CH2
2
+
X
NR R
2
hν
TMB
DPE
TMB
CO2 hν
CO2
step a
CO2
+
TMB
?
Ph C CH CO
2
2
2
Ph2C CH2CO2H
OH
A
2
0
Ph C CH2
2
TMB
H O
2
TMB
hν
H2O
step b
Ph C CH CO
Ph C CHCO H
2
2
2
2
2
B
DPA
Scheme 1. Possible mechanism.
2
0 produced may be a secondary photoproduct of DPA. DPE
potential E* is sufficiently more negative than the reduction
ox
was irradiated in the presence of added water (2 M) under
the conditions similar to run 5 of Table 2. The conversion
of DPE was complete (recoveries of DPEw0% and TMB
26%), but most of the product was 21 (90% yield). Only
a small amount of 20 (5% yield) and a trace of DPAwere ob-
tained (TON 2.5).
potential of DPE or CO . All the sensitizers in Table 3 appear
2
to meet this requirement and therefore the difference in Eo*x
is probably not the reason for the observed advantage of
TMB. The oxidation of A into B with the concomitant
back-electron transfer to regenerate TMB (Scheme 1) is
also a crucial step. TMB is not energetically advantageous
in this step when compared with DMA and DMT, because
Eox of TMB is slightly lower (w0.1 V) than that of DMA
or DMT (Table 3). However, the high TON found for
TMB suggests that this step is pretty efficient in the case
of TMB, although the reason is unclear.
2
.3. Reflection on the merit of TMB
Why TMB is a better photocarboxylation catalyst than other
DMA derivatives? The answer will allow us to design even
better photocatalysts. Since the TMB molecule consists of
two DMA moieties, bis[4-(dimethylamino)phenyl]methane
The fluorescence measurement for DMT, TMPD, and TMB
(Fig. 1B) shows that under these conditions, the TMB singlet
is quenched more by DPE rather than by CO , whereas the
(
16) and leucocrystal violet (17) were tested. Their TONs,
however, were much lower than that of TMB: compare
runs 6 and 7 with run 4 in Table 2. Furthermore, 16 and 17
are more photolabile than TMB: their recoveries were 3, 0,
and 34%, respectively. Next, some photophysical properties
of TMB were examined.
2
reverse is the case for the TMPD singlet and that the degree
of quenching of the DMT singlet by DPE and CO is approx-
2
y
imately equal. As already seen in Table 1, the efficiency for
DPA formation in MeCN is TMB>DMT>TMPD. This cor-
relation between the fluorescence quenching and the DPA
The absorption spectra of DMT, TMPD, TMB, DPE, and
DPA are shown in Figure 1A. Evidently, TMB (l
max
ꢁ
1
ꢁ1
3
(
10 nm, 3 42,000 M cm ) absorbs the Pyrex-filtered light
>290 nm) much more effectively than DMT or TMPD.
y
The fluorescence quenching constant k
q s 2
t by CO or DPE was estimated
from Figure 1B for each of DMT, TMPD, and TMB. The values were 3.7,
While the absorbance of DPE is very weak at >290 nm,
that of DPA is substantial at this region. Hence, in view of
a considerable inner filter effect by DPA, TMB will be
a much better sensitizer than DMT or TMPD.
ꢁ1
ꢁ1
2
) or 61, 51, 180 M (DPE), respectively. Since the sin-
15 16
6
.7, 6.7 M (CO
glet lifetime t for DMA, TMPD, and TMB in MeCN is 3.8, 1.3, and
s
1
10 ns, respectively, the quenching rate k by CO or DPE was calculated
7
q 2
8
9
8
ꢁ1 ꢁ1
10 10
as 9.7ꢂ10 , 5.2ꢂ10 , 6.7ꢂ10 M
s
(CO
ꢁ1 ꢁ1
s (DPE), respectively. Thus, k
2
) or 1.6ꢂ10 , 3.9ꢂ10
,
by DPE appears to be
1
0
1
.8ꢂ10
M
q
1
0
ꢁ1 ꢁ1
nearly diffusion controlled (kdiff¼1.9ꢂ10
M s
by CO
in MeCN at
decreases in the
1
5 C ) regardless of the amine, whereas k
4a
2
order TMPD>DMT>TMB.
The occurrence of photoinduced electron transfer from an
amine sensitizer requires that its excited-state oxidation
ꢃ
q
2

Y. Ito / Tetrahedron 63 (2007) 3108–3114
3111
Other carboxylated products were also formed (see Section
). The major product 25 could be isolated simply by recrys-
4
tallization of the photolysate. Incorporation of MeCN in the
product structure presumably proceeded through the Ritter-
1
8
like reaction, giving a heterocyclic intermediate, which
was then hydrolyzed during the work-up (Eq. 2).
Ph
Ph
Ph
N
O
Me
MeCN
N
Me
O
CO2
CO2
work-up
(hydrolysis)
hν/CO2
ð2Þ
Ph
hν
CO
MeCN
Ph
NHCOMe
2
+
TMB
CO H
2
24
25
24 : 1
major product
5 % yield (TON 5.9)
2
3. Conclusion
Photocarboxylation via traditional organic photochemistry
6
–8
might not look very promising. The present author now
found the highly catalytic photocarboxylation of DPE by us-
ing TMB as sensitizer, yielding DPA and 20. The turnover
number (TON¼(DPA+20)/TMB) reached 17. The formation
of 25 is an interesting type of photocarboxylation reaction.
As compared with related DMAs, TMB is more resistant
to photodecomposition, has the much larger absorbance in
the S /S (E ) transition, and has the lower quenching
Figure 1. (A) Absorption spectra for DMT, TMPD, TMB, DPE, and DPA in
ꢁ
5
MeCN: [amine]¼5.0ꢂ10 M. (B) Fluorescence spectra of DMT (dotted
broken line, lex¼305 nm), TMPD (broken line, lex¼330 nm), and TMB
ꢁ
5
(
solid line, lex¼330 nm) in MeCN: [amine]¼6–11ꢂ10 M. The fluores-
cence measurements were conducted under three different conditions: (a)
in argon-saturated MeCN, (b) in CO -saturated MeCN ([CO ]w 0.32 M ),
2 2
1
a
0
1
00
efficiency by CO . Probably these factors are partly respon-
and (c) in argon-saturated MeCN in the presence of DPE (0.017 M).
2
sible for the observed high TON. It is assumed that the
regeneration of TMB through possibly the back-electron
transfer from the radical anion A occurs pretty efficiently.
Analogous highly catalytic photocarboxylation models are
likely to be devised by appropriate choices of substrate
and catalyst.
a
Table 3. Electron-donor properties of the sensitizers in MeCN
Sensitizer Eox
versus
SCE
E
00
E
o
x
Sensitizer Eox
versus
SCE
E
00
ox
E*
versus
SCE
versus
SCE
b
b
b
DMA
DMT
0.53 V 3.89 V ꢁ3.36 V TMPD
0.50 3.77
ꢁ3.27 TMB
0.32 V 3.41 V ꢁ3.09 V
0.43 3.60
ꢁ3.17
4
. Experimental
a
The properties for DMA, TMPD, and TMB were derived from Ref. 14.
The Eox value for DMT was assumed from Ref. 12.
The excited-state oxidation potential E
4
.1. Instruments and materials
b
ox
¼EoxꢁE00, where Eox is the
oxidation potential and E00 is the excited singlet state energy.
1
13
H and C NMR spectra were measured on a JEOL EX-
2
70J, AL-300, or JUM-A400 spectrometer. Measurements
of 2D NMR were carried out with JEOL JUM-A400.
Mass, IR, and absorption spectra were recorded on JEOL
JMS-HX 110A, SHIMADZU FTIR-8400, and SHIMADZU
UV-2400PC spectrometers, respectively. Fluorescence spec-
tra were gathered with a SHIMADZU RF-5300PC spec-
trometer and excited near the absorption maximum. HPLC
analyses were performed with a JASCO PU-980 pump and
a SHIMADZU SPD-6AV detector (fixed at 254 nm) by using
a Cosmosil 5C -AR column (4.6 mm i.d.ꢂ150 mm) and
formation may be taken to indicate that DPA is formed via
the route involving step a, rather than step b (see Scheme 1).
2
.4. Photocarboxylation of 1-phenyl-1-cyclohexene
Photocarboxylation of a-methylstyrene by TMB was not ef-
ficient, similar to that by DMT. trans-Stilbene was not pho-
8
e
8
e
tocarboxylated by TMB, neither by DMA. Biphenyl was
photocarboxylated to give 4-phenylbenzoic acid cleanly
but only in a low efficiency (by TMB, 2.7% yield, TON
1
8
were eluted with a mixture of MeOH–acetate buffer
8
e
0
.16; by DMA, 3.5% yield, TON 0.03). On the other
hand, 1-phenyl-1-cyclohexene (24) yielded cis-2-acetamido-
-phenylcyclohexanecarboxylic acid (25) with reason-
able efficiency (Eq. 2): 25% yield, TON (¼25/TMB) 5.9.
(20 mM, pH 3.6) or a mixture of MeOH–H O.
2
2
Most of the materials employed here were purchased
from the company. 4-Chloro-N,N-dimethylaniline (5) and

3112
Y. Ito / Tetrahedron 63 (2007) 3108–3114
N,N-dimethyl-p-anisidine (8) were prepared by literature
methods. trans-2,3-Diphenyl-1-p-tolylaziridine (15) was
available from our previous work.
(run 5). Next, both parts were combined, the solvent was re-
moved by rotary evaporation at 40 C, and subsequently the
1
9
ꢃ
residue was redissolved in CHCl (8 mL). A white solid ap-
2
0
3
peared in a little while. After the solution was left overnight
at room temperature, the precipitated solid was collected by
filtration to afford 42 mg of pure 3-hydroxy-3,3-diphenyl-
propionic acid (20). Then, the filtrate was subjected to pre-
4
.2. Photocarboxylation (general)
The photocarboxylation experiments (Tables 1 and 2)
were conducted as follows. An equimolar mixture contain-
ing 0.13 M of N,N-dimethylanilines (DMAs) or analogs
1
vent was placed in a Pyrex (or NMR) tube and was irra-
diated externally, under bubbling CO , with a 400-W high
2
pressure mercury lamp (Pyrex filter, >290 nm) for 10 h.
During the irradiation, the solution was maintained at
either 20 C (for DMSO-d or DMSO solvent) or 10 C
6
(for MeCN or DMF-d solvent) with thermostated circu-
7
lating water. In the case of TMB (10), a 1:2 mol/mol
mixture containing (a) 0.065 M of TMB and 0.13 M of
DPE in DMF-d or (b) 0.018 M of TMB and 0.037 M
of DPE in MeCN was irradiated. TMB is insoluble in
DMSO.
parative TLC on silica gel (CHCl –MeOH 30:1 v/v). From
3
the lowest band just above the original sample band,
16 mg of additional 20 was obtained as a pale brown solid:
total yield of 20, 58 mg (16%). From the second lowest
band, 78 mg (23% yield) of pure 3,3-diphenylacrylic acid
(DPA, 19) was obtained as a pale brown solid. From the third
lowest band, 23 mg (8%) of nearly pure 1,1-diphenylethanol
(21) was obtained as a pale greenish brown solid. All the up-
per bands were discarded since they should contain mainly
DPE, biphenyl, and small amounts of TMB and benzophe-
none.
–15 (except TMB) and 0.13 M of DPE in suitable sol-
ꢃ
ꢃ
7
ꢃ
ene/hexane) (lit. mp 156–157.5 C); H NMR (270 MHz,
DPA (19): colorless prisms, mp 162–164 C (from 2:1 benz-
2
2
ꢃ
1
1
3
CDCl ) d 7.4–7.15 (10H, m), 6.32 (1H, s); C NMR
3
Product yields and recoveries were estimated by NMR and
HPLC analyses of the photolysates. In several experiments
(67.7 MHz, CDCl ) d 170.37, 158.87, 140.70, 138.29,
3
129.66, 129.17, 128.44, 128.39, 128.32, 127.86, 116.26;
IR (KBr) 3200–2500 (br), 1700 (s), 1613 (m), 1284 (m),
1216 (m), 774 (m), 696 (m) cm ; UV (MeCN, Fig. 1)
(
runs 3–6, 12, and 13 (¼run 1 in Table 2) in Table 1), a small
1
ꢁ1
H NMR peak at around d 8.5, which is probably ascribed to
the aldehyde proton of Ar–N(CHO)CH , was visible in the
+
+
l 274 (3 26,000) nm; MS (EI ) m/z 224 (M , 100), 223
(77), 207 (17), 179 (61), 178 (78); HRMS (EI ) calcd for
3
+
spectra. Also, in all the experiments in Table 2 (run 13 in
Table 1¼run 1 in Table 2), formation of a trace amount of
benzophenone was observed (0.3–1% yield from HPLC),
probably due to photooxidation of DPE with the residual
oxygen in CO gas. However, the production of 20, 21,
2
and benzophenone was examined only for the experiments
using TMB (Table 2).
C H O 224.0837, found 224.0837.
15 12 2
ꢃ
23
217 C); H NMR (270 MHz, DMSO-d ) d 12.38 (1H, br s),
20: White crystals, mp 232–233 C (from MeCN) (lit. mp
ꢃ
7.47–7.43 (4H, m), 7.29–7.23 (4H, m), 7.18–7.12 (2H, m),
1
6
1
3
5.78 (1H, br s), 3.28 (2H, s); C NMR (67.7 MHz,
DMSO-d ) d 172.87, 147.13, 127.72, 126.25, 125.35,
6
The photocarboxylation by TMB progressed rapidly at first
(
diation. This is probably due to an inner filter effect by DPA
as well as to the decomposition of TMB. The product ratio
between DPA and 20 did not change significantly with the
irradiation time.
75.46, 45.10; IR (KBr) 3477 (s), 3300–2500 (br), 1691 (s),
1442 (s), 1371 (m), 1232 (s), 1062 (m), 1022 (m), 925
(m), 753 (m), 694 (s) cm ; MS (EI ) m/z 242 (M , 3),
irradiation time<2 h), but almost stopped after 10 h of irra-
ꢁ
1
+
+
224 (11), 183 (86), 182 (70), 180 (59), 165 (29), 105
(100), 77 (15); HRMS (EI ) calcd for C H O
+
1
5
12
2
–
+
(¼M ꢁH O) 224.0837, found 224.0838; HRMS (FAB )
2
–
calcd for C H O (¼[MꢁH] ) 241.0865, found 241.0871.
1
5 13 3
As an illustrative example of photocarboxylation, the exper-
imental procedure for run 5 in Table 2 is described below. In
addition, the photocarboxylation experiment of 1-phenyl-1-
cyclohexene (24) is described.
For the quantum yield measurement, the light that was iso-
lated from a 400-W high pressure mercury lamp with
a K CO (0.67%)–K CrO (0.067%) filter solution (mainly
2
3
2
4
3
13 nm) was employed and the reaction was stopped at
4
1
.3. Photocarboxylation of 1,1-diphenylethylene (DPE,
0
a small conversion (the DPAyield<1.3 %). Photocyclization
of 2,4,6-triisopropylbenzophenone into its benzocyclobute-
nol in MeCN under argon (F ¼0.67) was used as actino-
0
8) with N,N,N ,N -tetramethylbenzidine (TMB, 10)
3
13
1
1
A solution containing 7.8 mg (0.032 mmol) of TMB and
75 mg (1.52 mmol) of DPE in MeCN (7 mL) was placed
in a Pyrex tube. The solution was irradiated for 10 h under
metry. Although each sample was irradiated separately
without using a merry-go-round apparatus, two runs for
the quantum yield measurement agreed within ꢄ2%.
2
ꢃ
bubbling of CO at 2 C. Only in this run, the photolysis
2
ꢃ
the dissolved CO in MeCN increases a little. The photo-
ꢃ
was carried out at 2 C rather than at 10 C, expecting that
4.4. Photocarboxylation of 1-phenyl-1-cyclohexene (24)
0
2
1
0
with N,N,N ,N -tetramethylbenzidine (TMB, 10)
2
lysate was divided into two equal parts. One part was ana-
lyzed by HPLC after addition of biphenyl (7.8 mg) as an
internal standard. Another part was rotary-evaporated at
Under bubbling of CO , a solution containing 9.4 mg
2
(0.039 mmol) of TMB and 151 mg (0.94 mmol) of 24 in
MeCN (8 mL) was irradiated as described above at 10 C
ꢃ
1
ꢃ
for 10 h. From the analysis of the divided photolysates by
4
0 C and was analyzed by H NMR. The yields and recov-
eries of DPA, DPE, TMB, and Ph CO were determined from
2
1
the HPLC data and the yields of 20 and 21 were estimated
from the H NMR data. The results are listed in Table 2
HPLC and H NMR as described above, formation of
many products was disclosed. The recoveries of 24 and
1

Y. Ito / Tetrahedron 63 (2007) 3108–3114
3113
TMB were 48 and 0%, respectively. After rotary evaporation
of the recombined photolysate, the residue was redissolved
Acknowledgements
in 3 mL of CHCl and the solution was allowed to stand at
3
room temperature for 5 days. Pale brownish yellow crystals
appeared and these were collected by filtration to obtain
The author thanks Ms. Hiromi Yoshida and Mr. Haruo Fujita
for measuring mass and 2D NMR spectra, respectively.
4
9 mg (20%) of cis-2-acetamido-2-phenylcyclohexanecarb-
oxylic acid (25). Since an HPLC peak corresponding to
5 was not observable directly after the photolysis, 25
Supplementary data
2
1
13
must have been formed during the work-up of the photoly-
sate. The assumed heterocyclic intermediate shown in
The H and C NMR spectra (Fig. 2) and the NOESY spec-
trum (Fig. 3) for cis-2-acetamido-2-phenylcyclohexanecar-
boxylic acid (25). Supplementary data associated with this
article can be found in the online version, at doi:10.1016/
j.tet.2007.02.013.
1
8
Eq. 2 may be formed via the Ritter-like reaction. The fil-
trate was separated with preparative TLC on silica gel
(
(
CHCl –MeOH 30:1 v/v), followed by preparative HPLC
3
Cosmosil 5C , 20 mm i.d.ꢂ250 mm, MeCN–H O 70:30
1
8
2
v/v). This gave 12 mg of additional 25 along with 13 mg
(
(
7%) of crude 2-phenylcyclohex-2-enecarboxylic acid
26). The total yield of 25 is 61 mg (25%) and thus the
References and notes
2
4
TON for 25 (¼25/TMB) is 5.9. Many other carboxylated
products (total 30 mg), including 2-phenylcyclohex-1-
enecarboxylic acid (27) and trans- and cis-2-hydroxy-2-
phenylcyclohexanecarboxylic acid (28), were formed in
minor amounts, judging from the H NMR, IR, MS, and
HPLC data, but these products were not further separated.
1. Our previous papers on this topic: (a) Masuda, K.; Ito, Y.;
Horiguchi, M.; Fujita, H. Tetrahedron 2005, 61, 213–229; (b)
Ito, Y.; Ushitora, H. Tetrahedron 2006, 62, 226–235; (c) Ito,
Y. Photochemistry 2002, 33, 205–212; (d) Ref. 8e.
2. (a) Komatsu, M.; Aida, T.; Inoue, S. J. Am. Chem. Soc. 1991,
113, 8492–8498; (b) Hirai, Y.; Aida, T.; Inoue, S. J. Am.
Chem. Soc. 1989, 111, 3062–3063; (c) Inoue, S.; Nukui, M.;
Kojima, F. Chem. Lett. 1984, 619–622; (d) Inoue, S.; Takeda,
N. Bull. Chem. Soc. Jpn. 1977, 50, 984–986.
3. (a) Morgenstern, D. A.; Wittrig, R. E.; Fanwick, P. E.; Kubiak,
C. P. J. Am. Chem. Soc. 1993, 115, 6470–6471; (b) Lemke,
F. R.; Delaet, D. L.; Gao, J.; Kubiak, C. P. J. Am. Chem. Soc.
1988, 110, 6904–6906; (c) Johnston, R. F.; Cooper, J. C.
Organometallics 1987, 6, 2448–2449; (d) Cocolios, P.;
Guilard, R.; Bayeul, D.; Lecomte, C. Inorg. Chem. 1985, 24,
2058–2062; (e) Kodaka, M.; Lee, A. L.; Tomohiro, T.;
Okuno, H. Chem. Express. 1990, 5, 233–236.
4. (a) Wada, Y.; Ogata, T.; Hiranaga, K.; Yasuda, H.; Kitamura, T.;
Murakoshi, K.; Yanagida, S. J. Chem. Soc., Perkin Trans. 2
1998, 1999–2004; (b) Fujiwara, H.; Kanemoto, M.; Ankyu,
H.; Murakoshi, K.; Wada, Y.; Yanagida, S. J. Chem. Soc.,
Perkin Trans. 2 1997, 317–321; (c) Sclafani, A.; Palmisano,
L.; Farneti, G. J. Chem. Soc., Chem. Commun. 1997, 529–
530; (d) Kawai, T.; Kuwabara, T.; Yoshino, K. J. Chem. Soc.
Faraday Trans. 1992, 88, 2041–2046.
1
ꢃ
H NMR (400 MHz, CD OD) d 7.44–7.40 (2H, quasi-d,
2
5: White plates, mp 245–247 C (from MeOH–acetone);
1
x
3
x
Jw7 Hz), 7.26–7.20 (2H, quasi-t, Jw7.5 Hz), 7.16–7.11
x
(
1H, quasi-t, Jw7 Hz), 3.34–3.31 (1H, m), 2.73–2.68 (2H,
m), 2.13–2.02 (1H, m), 1.9–1.83 (1H, m), 1.89 (3H, s),
1
3
1
d 176.01 (CONH), 172.18 (COOH), 146.72, 128.70,
.74–1.52 (4H, m) (Fig. 2); C NMR (100 MHz, CD OD)
3
1
2
27.52, 127.22, 59.84, 50.15, 28.07, 26.41, 23.57, 22.51,
1.92 (Fig. 2); IR (KBr) 3369 (s), 1711 (s), 1640 (s), 1545
(
(
2
s), 1374 (m), 1179 (s), 1155 (s), 1125 (s), 769 (m), 697
ꢁ
1
+
+
m) cm ; MS (EI ) m/z 261 (M , 13), 243 (11), 218 (18),
02 (93), 184 (19), 176 (31), 156 (56), 132 (100), 119
33), 104 (51), 91 (64), 77 (30), 60 (27); HRMS (EI ) calcd
+
(
for C H NO 261.1365, found 261.1367.
19
15
3
The cis configuration of CO H and NHCOMe was assigned
2
on the basis of a pronounced NOESY correlation between
the methine hydrogen (H ) of CH CO H (d 3.34–3.31)
a
a
2
and the ortho hydrogen of Ph (d 7.44–7.40) (Fig. 3). The
H NMR signal of H appeared as approximately a double
5. (a) Itoh, T.; Asada, H.; Tobioka, K.; Kodera, Y.; Matsushima,
A.; Hiroto, M.; Nishimura, H.; Kamachi, T.; Okura, I.; Inada,
Y. Bioconjug. Chem. 2000, 11, 8–13; (b) Inoue, H.; Kubo, Y.;
Yoneyama, H. J. Chem. Soc. Faraday Trans. 1991, 87,
1
a
doublet (Fig. 2), which is also consistent with the cis
structure.
2
5
553–557; (c) Mandler, D.; Willner, I. J. Chem. Soc., Perkin
Trans. 2 1988, 997–1003.
2
4
1
2
6: Colorless solid; H NMR (300 MHz, CD OD) d 7.35–
3
7
.12 (5H, m), 6.17 (1H, t, J¼3.8 Hz), 3.68 (1H, br), 2.3–1.7
6. (a) Padwa, A.; Wetmore, S. I., Jr. J. Am. Chem. Soc. 1974, 96,
2414–2421; (b) Giezendanner, H.; M €a rky, M.; Jackson, B.;
Hansen, H.-J.; Schmid, H. Helv. Chim. Acta 1972, 55, 745–748.
7. (a) Sander, W. J. Org. Chem. 1989, 54, 4265–4267; (b)
Wheland, R.; Bartlett, P. D. J. Am. Chem. Soc. 1970, 92,
ꢁ
1
+
(6H, m); IR (film) 1700 (s), 758 (s), 697 (s) cm ; MS (EI )
m/z 202 (M , 64), 156 (100), 130 (69), 115 (57), 91 (97), 77
+
+
(
33); HRMS (EI ) calcd for C H O 202.0994, found
13 14 2
2
02.0993.
6
057–6058.
8
. (a) Chateauneuf €y , J. E.; Zhang, J.; Foote, J.; Brink, J.; Perkovic,
M. W. Adv. Environ. Res. 2002, 6, 487–493; (b) Nikolaitchik,
A. V.; Rodgers, M. A.; Neckers, D. C. J. Org. Chem. 1996,
OH
Ph
Ph
CO2H
27
Ph
CO H
CO H
2
2
6
1, 1065–1072; (c) Tagaya, H.; Onuki, M.; Tomioka, Y.;
Wada, Y.; Karasu, M.; Chiba, K. Bull. Chem. Soc. Jpn. 1990,
3, 3233–3237; (d) Minabe, M.; Isozumi, K.; Kawai, K.;
2
6
trans- and cis-28
6
Yoshida, M. Bull. Chem. Soc. Jpn. 1988, 61, 2063–2066; (e)
Ito, Y.; Uozu, Y.; Matsuura, T. J. Chem. Soc., Chem.
Commun. 1988, 562–564; (f) Tazuke, S.; Kazama, S.;
x
Each peak is finely split (J¼1–2 Hz).

3114
Y. Ito / Tetrahedron 63 (2007) 3108–3114
Kitamura, N. J. Org. Chem. 1986, 51, 4548–4553; (g) Toki, S.;
Hida, S.; Takamuku, S.; Sakurai, H. Nihon Kagaku Kaishi
984, 152–157.
15. K €o hler, G. J. Photochem. 1987, 38, 217–238.
16. Nakamura, S.; Kanamaru, N.; Nohara, S.; Nakamura, H.; Saito,
Y.; Tanaka, J.; Sumitani, M.; Nakashima, N.; Yoshihara, K.
Bull. Chem. Soc. Jpn. 1984, 57, 145–150.
17. Hashimoto, S.; Thomas, J. K. J. Phys. Chem. 1984, 88, 4044–
4049.
1
9
. Gennaro, A.; Isse, A. A.; Sav ꢀe ant, J.-M.; Severin, M.-G.;
Vianello, E. J. Am. Chem. Soc. 1996, 118, 7190–7196.
0. Mella, M.; Coppo, P.; Guizzardi, B.; Fagnoni, M.; Freccero,
1
1
M.; Albini, A. J. Org. Chem. 2001, 66, 6344–6352.
1. Photocyclization of 2,4,6-triisopropylbenzophenone in MeCN
18. Clayden, J.; Greeves, N.; Warren, S.; Wothers, P. Organic
Chemistry; Oxford University Press: Oxford, UK, 2001; pp
1114–1116.
19. (a) Sekiya, M.; Tomie, M.; Leonard, N. J. J. Org. Chem. 1968,
33, 318–322; (b) H €u nig, S. Chem. Ber. 1952, 85, 1056–1060.
20. Ito, Y.; Ida, H.; Matsuura, T. Tetrahedron Lett. 1978, 3119–
3120.
(
F
313¼0.67) was used as a standard. See (a) Ito, Y.; Umehara,
Y.; Hijiya, T.; Yamada, Y.; Matsuura, T. J. Am. Chem. Soc.
980, 102, 5917–5919; (b) Ito, Y.; Nishimura, H.; Umehara,
Y.; Yamada, Y.; Tone, M.; Matsuura, T. J. Am. Chem. Soc.
983, 105, 1590–1597; (c) Ito, Y.; Matsuura, T. J. Am. Chem.
1
1
Soc. 1983, 105, 5237–5244; (d) Ito, Y.; Matsuura, T.;
Fukuyama, K. Tetrahedron Lett. 1988, 29, 3087–3090.
2. Siegerman, H. Technique of Electroorganic Synthesis;
Weinberg, N. L., Ed.; Wiley-Interscience: New York, NY,
21. Seidell, A.; Linke, W. F. Solubilities of Inorganic and Metal
Organic Compounds, 4th ed.; D. Van Nostrand: Princeton,
NJ, 1958; Vol. 1, pp 459–494.
22. Sakakibara, T.; Nishimura, S.; Kimura, K.; Minato, I.; Odaira,
Y. J. Org. Chem. 1970, 35, 3884–3887.
1
1
975; Part II, p 667.
3. Koppenol, W. H.; Rush, J. D. J. Phys. Chem. 1987, 91, 4429–
430.
1
1
23. Bellassoued, M.; Couffignal, R.; Gaudemar, M. J. Organomet.
Chem. 1973, 61, 9–18.
4
4. (a) Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of
Photochemistry, 2nd ed.; Marcel Dekker: New York, NY,
24. Huang, F.; Browne, C. E.; Wu, W.-M.; Juhasz, A.; Ji, F.; Bodor,
N. Pharm. Res. 2003, 20, 1681–1689.
1
993; (b) Sundararajan, C.; Falvey, D. E. J. Org. Chem.
25. Turbanti, L.; Cerbai, G.; Ceccarelli, G. Arzneim.-Forsch. 1978,
28, 1249–1252.
2004, 69, 5547–5554.