Pressure and temperature control of product chirality in asymmetric photochemistry. Enantiodifferentiating photoisomerization of cyclooctene sensitized by chiral benzenepolycarboxylates

Article abstract of DOI:10.1021/ja981929a

Pressure effects upon asymmetric photosensitization have been investigated for the first time in the enantiodifferentiating Z-E photoisomerization of cyclooctene (1), sensitized by chiral aromatic esters (2-7). The product's enantiomeric excess (ee) and E/Z ratio were critical functions of the applied pressure, exhibiting an unprecedented switching of the product chirality. Depending upon the chiral sensitizer employed, the differential activation volume (ΔΔV(+) varies widely from -3.7 to +5.6 cm³ mol^-1, which is unexpectedly large for an enantiodifferentiation in the excited state. However, the ΔΔV(+) values obtained do not correlate with the differential activation enthalpy (ΔΔH(+)) or entropy (ΔΔS(+)) obtained from temperature-dependence studies, indicating that pressure and temperature function as independent perturbants for the photoenantio- differentiation process. Further investigations on the pressure dependence of ee at low temperatures enable us to construct the first three-dimensional diagram that correlates the product's ee with pressure and temperature changes. The combined effects of temperature and pressure provide us with a versatile tool for the multidimensional control of asymmetric photochemical reactions, in which we can switch and/or enhance the product chirality at more readily accessible temperatures and pressures, without using antipodal sensitizers.

Full text of DOI:10.1021/ja981929a

J. Am. Chem. Soc. 1998, 120, 10687-10696
10687
Pressure and Temperature Control of Product Chirality in Asymmetric
Photochemistry. Enantiodifferentiating Photoisomerization of
Cyclooctene Sensitized by Chiral Benzenepolycarboxylates
Yoshihisa Inoue,* Eiji Matsushima, and Takehiko Wada
Contribution from the Inoue Photochirogenesis Project, ERATO, JST, 4-6-3 Kamishinden,
Toyonaka 565-0085, Japan, and Department of Molecular Chemistry, Faculty of Engineering,
Osaka UniVersity, 2-1 Yamada-oka, Suita 565-0871, Japan
ReceiVed June 2, 1998
Abstract: Pressure effects upon asymmetric photosensitization have been investigated for the first time in the
enantiodifferentiating Z-E photoisomerization of cyclooctene (1), sensitized by chiral aromatic esters (2-7).
The product’s enantiomeric excess (ee) and E/Z ratio were critical functions of the applied pressure, exhibiting
an unprecedented switching of the product chirality. Depending upon the chiral sensitizer employed, the
q
3
-1
differential activation volume (∆∆V ) varies widely from -3.7 to +5.6 cm mol , which is unexpectedly
q
large for an enantiodifferentiation in the excited state. However, the ∆∆V values obtained do not correlate
q
q
with the differential activation enthalpy (∆∆H ) or entropy (∆∆S ) obtained from temperature-dependence
studies, indicating that pressure and temperature function as independent perturbants for the photoenantio-
differentiation process. Further investigations on the pressure dependence of ee at low temperatures enable us
to construct the first three-dimensional diagram that correlates the product’s ee with pressure and temperature
changes. The combined effects of temperature and pressure provide us with a versatile tool for the
multidimensional control of asymmetric photochemical reactions, in which we can switch and/or enhance the
product chirality at more readily accessible temperatures and pressures, without using antipodal sensitizers.
Introduction
This conflicts with the wide-spread belief that higher op’s are
generally obtained at lower temperatures. Eyring treatments
of the relative rate constants for the production of (S)- and (R)-
Asymmetric photosensitization with a catalytic amount of
optically active sensitizer, leading to the formation of an excited
state complex (exciplex) and ultimately to chiral products, has
fascinated (photo)chemists from both a mechanistic and syn-
thetic point of view over the last three decades since the first
attempt by Hammond and Cole in 1965.3 Despite the consider-
able efforts devoted to studies of asymmetric photosensitizations
with a variety of substrates and chiral sensitizers, the optical
purities (op) or enantiomeric excesses (ee) reported have not
exceeded 10% until recently.2
The greatest inherent advantage of photochemical asymmetric
reactions over their thermal counterparts is the virtually
unrestricted temperature range that is available. However, this
unique benefit, which is specific to excited-state reactions, had
not been positively utilized or even examined in most (asym-
1
E reveal that this unusual temperature switching behavior is
entropic in origin. Unequal entropies of activation for the
enantiodifferentiating step are responsible for this unusual
temperature dependence. Hence, temperature can be used as a
convenient, but powerful tool for product chirality control, and
both enantiomers can be prepared not by using antipodal
sensitizers, but merely by changing the irradiation temperature.
These unprecedented results and interpretations open a new
concept in the entropic control of asymmetric photochemistry,
and we have been prompted to study the effect of pressure as
an alternative tool for controlling the weak interactions that
discriminate product chirality in the excited state.
1,2
The effects of hydrostatic pressure upon chemical and
biological reactions and equilibria have been extensively
investigated, and considerable amounts of data on the activation
2
metric) photochemical reactions for a long period of time. Only
recently we have demonstrated that changing the reaction
temperature drastically affects the product’s opsas well as the
photostationary E/Z ratiosobtained in the enantiodifferentiating
Z-E photoisomerization of (Z)-cyclooctene (1Z) sensitized by
optically active (poly)alkyl benzene(poly)carboxylates.
several cases, the chirality of the photoproduct (E)-cyclooctene
1
0-20
and reaction volumes have been compiled.
Apart from
these ground-state reactions and equilibria, several photophysical
(6) Inoue, Y.; Yamasaki, N.; Yokoyama, T.; Tai, A. J. Org. Chem. 1993,
58, 1011.
4
,5
In
(7) Tsuneishi, H.; Hakushi, T.; Tai, A.; Inoue, Y. J. Chem. Soc., Perkin
Trans. 2 1995, 2057.
4-9
(1E) can switch simply by changing the irradiation temperature.
(8) Tsuneishi, H.; Hakushi, T.; Inoue, Y. J. Chem. Soc., Perkin Trans.
1996, 1601.
2
*
Address correspondence to this author at Osaka University.
(9) Inoue, Y.; Tsuneishi, H.; Hakushi, T.; Tai, A. J. Am. Chem. Soc.
1997, 119, 472.
(10) For earlier works, see: (a) Asano, T.; le Noble, W. J. Chem. ReV.
1978, 78, 407. (b) van Eldik, R.; Asano, T.; le Noble, W. J. Chem. ReV.
1989, 89, 549.
(
(
(
(
1) Rau, H. Chem. ReV. 1983, 83, 535.
2) Inoue, Y. Chem. ReV. 1992, 92, 441.
3) Hammond, G. S.; Cole, R. S. J. Am. Chem. Soc. 1965, 87, 3256.
4) Inoue, Y.; Yokoyama, T.; Yamasaki, N.; Tai, A. J. Am. Chem. Soc.
1
989, 111, 6480.
5) Inoue, Y.; Yamasaki, N.; Yokoyama, T.; Tai, A. J. Org. Chem. 1992,
7, 1332.
(11) Ringo, M. C.; Evans, C. E. J. Phys. Chem. B 1997, 101, 5525.
(12) Mark o´ , I. E.; Giles, P. R.; Hindley, N. J. Tetrahedron 1997, 53,
1015.
(
5
1
0.1021/ja981929a CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/06/1998

10688 J. Am. Chem. Soc., Vol. 120, No. 41, 1998
Inoue et al.
and photochemical processes have recently been subjected to
pressure-dependence studies.10,21-47 Thus, the effects of ap-
plied pressure and accompanying viscosity change upon
Scheme 1
21-32
33-37
fluorescence,
stitution,
excimer/exciplex formation,
photosub-
1
0,38
34,39-43
photoisomerization,
and photocycloaddi-
tion4
4-47
have been studied in considerable detail. In the
44-47
photocycloaddition reactions,
the product’s regio- or dias-
tereoselectivity has been shown to be more or less dependent
upon the applied pressure. However, no such efforts have been
devoted to enantiodifferentiating photochemical reactions, partly
as a result of the fairly low ee’s reported for such asymmetric
processes at ambient temperature and pressure.2
We now wish to report our results concerning the first pres-
sure-dependence study on the enantiodifferentiating Z-E photo-
isomerization of cyclooctene, sensitized by optically active ben-
zene(poly)carboxylates at ambient temperature. We have further
extended our study to the pressure-dependent behavior of this
enantiodifferentiating photoreaction at lower temperatures to
-1
construct a three-dimensional P-T -ee diagram that represents
(13) Fantin, G.; Fogagnolo, M.; Guerzoni, M. E.; Lanciotti, R.; Medici,
A.; Pedrini, P.; Rossi, D. Tetrahedron: Asymm. 1996, 7, 2879.
(
(
(
14) Jenner, G. Tetrahedron Lett. 1995, 36, 233.
15) Smith, E. T. J. Am. Chem. Soc. 1995, 117, 6717.
16) Diedrich, M. K.; Hochstrate, D.; Kl a¨ rner, F.-G.; Zimny, B. Angew.
Chem., Int. Ed. Engl. 1994, 33, 1079.
17) Takeshita, H.; Liu, J.-F.; Kato, N. J. Chem. Soc., Perkin Trans. 1
994, 1433.
18) Kuhlmann, B.; Arnett, E. M.; Siskin, M. J. Org. Chem. 1994, 59,
377.
19) Doering, W. v. E.; Birladeanu, L.; Sarma, K.; Teles, J. H.; Kl a¨ rner,
F.-G.; Gehrke, J.-S. J. Am. Chem. Soc. 1994, 116, 4289.
the product’s ee as a function of temperature (T) and pressure
(P). We have also established a general method for the mul-
tidimensional control of asymmetric photochemical reactions
by the application of several independent external variables such
as T, P, etc.
(
1
(
5
(
(
(
20) Asano, T.; Furuta, H.; Sumi, H. J. Am. Chem. Soc. 1994, 116, 5545.
21) Hara, K.; Bulgarevich, D. S.; Kajimoto, O. J. Chem. Phys. 1996,
Results and Discussion
1
1
5
04, 9431.
Under comparable conditions except for the applied pressure,
which ranged from 0.1 (atmospheric) to 400 MPa, the enan-
tiodifferentiating photosensitized isomerizations of 1Z (5 mM)
to chiral 1E were performed in pentane at 25 °C, and
subsequently at lower temperatures, using an unfiltered high-
pressure mercury arc as the light source (>250 nm). The chiral
sensitizers employed were benzene(poly)carboxylic esters 2-7
(22) Lang, J. M.; Dreger, Z. A.; Drickamer, H. G. Chem. Phys. Lett.
995, 243, 78.
(
23) Tanaka, F.; Okamoto, M.; Hirayama, S. J. Phys. Chem. 1995, 99,
25.
24) Foguel, D.; Chaloub, R. M.; Silva, J. L.; Crofts, A. R.; Weber, G.
Biophys. J. 1992, 63, 1613.
(
(
25) Hara, K.; Rettig, W. J. Phys. Chem. 1992, 96, 8307.
(26) Hirayama, S.; Yasuda, H.; Okamoto, M.; Tanaka, F. J. Phys. Chem.
1
991, 95, 2971.
(
0.2-1 mM) of optically active alcohols, i.e., (-)-menthol (a),
-)-borneol (b), (S)-(+)-2-butanol (c), (R)-(-)-2-hexanol (d),
(27) Crane, D. R.; Ford, P. C. J. Am. Chem. Soc. 1991, 113, 8510.
(28) Crane, D. R.; Ford, P. C. J. Am. Chem. Soc. 1990, 112, 6871.
(29) Lueck, H.; Windsor, M. W.; Rettig, W. J. Phys. Chem. 1990, 94,
(
and (R)-(-)-2-octanol (e), shown in Scheme 1.
4
550.
In most of these experiments the course of the photoisomer-
ization process was not monitored, and only data at the initial,
intermediate, and final (or photostationary) stages were collected
at each pressure and temperature. This is because the ee values
obtained in this enantiodifferentiating photoisomerization have
been demonstrated to be practically independent of conversion,
(30) Schmidt, R.; Janssen, W.; Brauer, H.-D. J. Phys. Chem. 1989, 93,
4
66.
(
(
31) Chong, P. L.-G. Biochemistry 1988, 27, 399.
32) Turro, N. J.; Okamoto, M.; Kuo, P.-L. J. Phys. Chem. 1987, 91,
1
819.
(
(
(
33) Schneider, S.; J a¨ ger, W. J. Phys. Chem. 1996, 100, 8118.
34) Hara, K. Trends Phys. Chem. 1995, 5, 1.
35) Okamoto, M. ReV. High-Pressure Sci. Technol. 1993, 2, 119 (in
5
at least under atmospheric pressure. The E/Z ratio, conversion,
Japanese).
chemical yield, and ee of 1E produced are listed in Table 1.
The ee values are preceded by the sign of product’s optical
rotation at 589 nm, thus enabling immediate identification of
the enantiomer produced.
It is noted that, even at pressures as high as 200 or 400 MPa,
the main course of the photosensitization is the geometrical
isomerization of 1Z to highly strained 1E in good to excellent
chemical yields based on consumed 1Z. Side reactions, such
as the cycloaddition of 1 to the sensitizer or the polymerization
of 1, do not accompany this process.
(36) Okamoto, M.; Tanaka, F.; Teranishi, H. J. Phys. Chem. 1986, 90,
1
055.
(
37) Viriot, M. L.; Guillard, R.; Kauffmann, I.; Andre, J. C.; Siest, G.
Biochim. Biophys. Acta 1983, 733, 34.
38) Wieland, S.; Reddy, K. B.; van Eldik, R. Organometallics 1990, 9,
802.
(
1
(
(
(
39) Hara, K.; Ito, N.; Kajimoto, O. J. Phys. Chem. A 1997, 101, 2240.
40) Hara, K. Kagaku Kogyo 1996, 47, 1 (in Japanese).
41) Schroeder, J.; Schwarzer, D.; Troe, J.; V o¨ hringer, P. Chem. Phys.
Lett. 1994, 218, 43.
42) Schroeder, J.; Troe, J.; V o¨ hringer, P. Chem. Phys. Lett. 1993, 203,
55.
43) Hara, K. ReV. High-Pressure Sci. Technol. 1993, 2, 110 (in
Japanese).
(
2
(
Photoisomerization and Enantiodifferentiation Mecha-
nisms. It has recently been demonstrated that the enantiodif-
ferentiating photoisomerization of 1 sensitized by chiral ben-
zenepolycarboxylates proceeds via a singlet mechanism involving
an exciplex intermediate between excited chiral sensitizer and
(44) Iwaoka, T.; Katagiri, N.; Sato, M.; Kaneko, C. Chem. Pharm. Bull.
1
992, 40, 2319.
(
(
45) Buback, M.; B u¨ nger, J.; Tietze, L. F. Chem. Ber. 1992, 125, 2577.
46) Heritage, T. W. Ph.D. Thesis, University of Reading, 1991; Diss.
Abstr. Intl. 1992, 53, 847-B.
47) Chung, W.-S.; Turro, N. J.; Mertes, J.; Mattay, J. J. Org. Chem.
989, 54, 4881.
4
,5
1
.
The intervention of an exciplex was confirmed by the
(
1
efficient quenching of sensitizer fluorescence, accompanied by

Pressure Effect in Asymmetric Photochemistry
J. Am. Chem. Soc., Vol. 120, No. 41, 1998 10689
Table 1. Pressure Effects upon Enantiodifferentiating Photoisomerization of Cyclooctene (1Z) Sensitized by Chiral (Poly)alkyl
a
Benzene(poly)carboxylates (2-7) in Pentane at 25 °C
pressure/ irradiation
%
%
pressure/ irradiation
%
%
sensitizer
MPa
time/min 1E/1Z conversion yield % ee sensitizer
MPa
time/min 1E/1Z conversion yield % ee
2
2
2
3
3
3
4
4
a
b
e
0.1
10
0.028
0.082
0.12
0.016
0.085
0.15
0.031
0.083
0.15
0.057
0.12
5.8
11
18
5.8
10
17
b
9.2
19
4.7
11
18
4.9
13
18
2.6
12
20
6.8
10
12
7.2
9.7
18
4.6
11
18
8.1
11
17
2.4
3.1
14
7.3
7.3
17
b
b
-2.5
-2.5
-2.3
-1.8
-2.5
b
-2.4
-1.9
+0.1
+0.5
+0.1
-1.1
-0.6
-1.2
b
+0.1
-0.5
+1.1
+1.2
+0.2
+0.5
-0.1
-0.5
-0.5
+1.1
-0.8
b
4e
5a
5b
5e
6a
0.1
10
30
60
10
30
60
10
30
60
10
30
60
10
30
60
10
30
60
10
30
60
10
30
60
10
30
60
10
30
60
10
30
60
10
30
60
10
30
60
10
30
60
10
30
60
60
60
10
30
60
10
30
60
10
30
60
10
30
60
10
60
10
30
60
10
30
60
10
60
10
30
60
0.10
0.23
0.30
0.10
0.24
0.34
0.091
0.24
0.32
0.23
0.32
0.39
0.25
0.37
0.40
0.24
0.39
0.43
0.18
0.26
0.30
0.22
0.39
0.41
0.20
0.40
0.47
0.19
0.29
0.31
0.22
0.38
0.44
0.24
0.44
0.51
0.12
0.19
0.21
0.10
0.26
0.32
0.11
0.31
0.36
0.020
0.050
0.35
0.30
0.31
0.19
0.33
0.34
0.19
0.36
0.43
0.14
0.20
0.19
0.19
0.33
0.23
0.39
0.44
0.15
0.20
0.22
0.19
0.34
0.23
0.34
0.45
7.0
20
27
9.7
22
29
b
18
22
9.1
19
24
-1.3
-1.1
-1.1
-1.2
-1.0
-0.9
-1.2
-1.2
-1.0
-6.0
-6.6
-6.2
-7.0
-7.4
-7.0
-7.5
-7.4
-7.7
-3.9
-4.9
-4.9
-5.0
-4.9
-4.7
-3.9
-5.1
-4.9
-2.4
-3.4
-3.1
-3.0
-2.7
-2.7
-2.7
-2.9
-2.6
-9.3
-10.7
-11.2
-1.9
-1.5
-3.6
b
+6.6
+5.7
+12.4
+17.9
+14.7
+16.1
+14.2
+15.0
+14.4
+14.3
+13.4
+11.9
+13.0
+2.1
+2.3
+2.9
+0.6
+0.1
-0.4
-0.3
-0.6
-2.5
-2.5
-2.7
+0.8
-1.1
+1.9
+2.3
+1.3
3
6
0
0
7.3
9.7
b
7.6
13
b
7.5
12
b
11
13
4.8
8.5
14
b
9.2
14
4.8
9.0
14
b
9.3
14
b
7.6
16
b
5.1
b
b
b
6.8
b
9.4
9.0
b
5.5 +11.9
7.1 +10.6
1
00
10
100
200
0.1
3
6
0
0
2
00
10
9.4
8.2
3
6
0
0
22
29
25
37
36
33
38
45
23
43
53
19
29
33
22
35
40
19
36
44
17
27
32
22
34
41
24
37
45
15
28
38
20
34
45
21
40
41
b
18
22
17
21
25
17
23
22
18
22
20
14
18
20
17
25
25
16
26
26
16
21
21
17
25
26
19
27
28
10
14
13
0.1
10
3
6
0
0
0.16
1
00
00
10
0.050
0.098
0.17
0.038
0.10
0.18
0.051
0.10
100
200
0.1
3
6
0
0
2
10
3
6
0
0
0.1
10
3
6
0
0
0.16
0.046
0.10
1
00
00
10
100
200
0.1
3
6
0
0
0.17
2
10
0.033
0.085
0.19
3
6
0
0
a
b
e
0.1
10
0.028
0.058
0.055
0.018
0.075
0.080
0.019
0.10
3
6
0
0
+6.3
+4.9
b
+4.4
+3.3
b
+2.0
+1.6
+11.0
1
00
00
10
100
200
0.1
3
6
0
0
2
10
3
6
0
0
0.11
0.1
10
0.031
0.059
0.079
0.047
0.074
0.095
0.043
0.099
0.10
0.028
0.068
0.068
0.033
0.071
0.091
0.038
0.10
0.13
0.10
0.24
0.33
0.12
0.29
0.34
0.11
4.9
6.2
10
12
17
22
11
19
22
3
6
0
0
1
00
00
10
4.2
6.1
7.5
b
8.1
8.0
b
6.3
6.2
b
6.4
8.1
3.7
9.1
b
100
200
8.3
17
18
3
6
0
0
+8.2
+7.2
+5.2
+5.0
+4.7
b
2
10
8.8
3
6
0
0
18
21
24
23
24
22
19
15
22
20
14
21
23
11
14
13
14
19
16
23
23
12
14
14
15
20
16
23
23
0.1
10
2.3
300
400
0.1
3
6
0
0
7.5
8.4
3.2
8.9
11
4.5
9.7
15
7.0
20
30
6.6
21
32
+2.9
+3.2
+3.1
+3.7
+2.3
+1.9
+1.9
+1.6
-3.6
-3.9
-4.3
-3.5
-3.8
-4.0
-3.5
-4.1
-4.5
b
b
6b
32
29
39
21
35
41
25
40
46
22
30
35
25
42
29
41
48
19
28
35
19
41
28
38
49
1
1
00
10
3
6
1
3
6
0
0
0
0
0
100
200
0.1
11
a
0.1
10
b
19
23
b
23
23
b
22
28
9.5
18
24
9.6
18
21
5.7
3
6
0
0
6c
00
10
3
6
1
3
6
0
0
0
0
0
100
200
6.8
0.26
0.36
0.11
0.23
0.33
0.11
0.22
0.31
14
21
11
23
29
13
19
33
10
20
33
b
0.1
10
3
6
0
0
-0.3
-0.4
b
-0.4
+0.4
b
6d
0.1
1
2
00
00
10
3
6
0
0
100
200
10
0.06
0.20
0.32
3
6
0
0
16
21
-0.4
-0.4

1
0690 J. Am. Chem. Soc., Vol. 120, No. 41, 1998
Table 1 (Continued)
pressure/ irradiation
Inoue et al.
%
%
pressure/ irradiation
%
%
sensitizer
MPa
time/min 1E/1Z conversion yield % ee sensitizer
MPa
time/min 1E/1Z conversion yield % ee
6
e
0.1
10
0.15
0.21
0.22
0.19
0.31
0.35
0.23
0.38
0.46
0.012
0.026
0.034
b
19
26
31
22
31
37
24
35
43
b
13
15
15
15
21
22
17
25
26
b
-3.0
-3.2
-3.8
0.0
-0.3
0.0
+2.1
+2.1
+2.1
-14.1
-10.9
7b
0.1
30
60
10
30
60
10
30
60
10
0.091
0.10
10
30
60
10
30
0.029
0.034
0.023
0.040
0.062
0.029
0.057
0.073
0.055
7.7
4.9
7.5
b
6.2
13
5.3
8.4
15
5.2
8.4
9.1
8.0
14
19
7.3
b
+7.1
3
6
0
0
3.2 +5.0
b
3.8 +18.5
5.4 +15.8
b
100
200
+17.3
1
00
10
3
6
0
0
+30.0
2
00
10
5.2 +29.4
6.2 +26.8
5.2 +5.0
3
6
0
0
7e
0.1
30
60
7
ac
0.1
30
+4.3
30
60
b
+3.4
+3.7
+1.4
+1.3
+1.9
60
2.4
b
11
+4.2
7.8
13
5
0
60
10
5.1
0.1
5.2
2.6
9.3
3.2 -15.0
-17.0
4.2 -17.7
-20.9
5.1 -21.4
100
0.085
0.15
0.19
0.10
0.21
0.26
1
00
b
60
0.045
0.014
0.056
15
b
17
20
1
50
10
b
200
60
18
24
6
0
a
Irradiation (>280 nm) conducted at 25 °C in a pressurized vessel with a sapphire window, using an unfiltered high-pressure mercury arc; [1Z]
b
)
5 mM, [sensitizer] ) 1 mM, unless noted otherwise. Value not determined. c [Sensitizer] ) 0.5 mM.
Scheme 2
kqS ) kqR in Scheme 2) and unequivocally show that the
rotational relaxation within the exciplex intermediate is the only
enantiodifferentiating process in this photosensitized isomer-
ization (i.e., kS * kR in Scheme 2), at least at room temperature
and atmospheric pressure.
Table 1 shows that the product’s ee varies significantly with
the electronic and stereochemical structure of sensistizer
employed, and also with the applied pressure, but is not
appreciably influenced by irradiation period (within experimental
error; (0.5-1.0% ee) at atmospheric as well as at much higher
pressures. It is concluded therefore that the photosensitization
and enantiodifferentiation mechanisms established at atmo-
spheric pressure can be extended to the photosensitized isomer-
ization at high pressures.
Pressure Effect upon Photoisomerization. According to
the reversible photoisomerization mechanism shown in Scheme
2
, the E/Z ratio at the photostationary state (pss) is given as a
product of the excitation ratio, kqZ/kqE, and the decay ratio, kE/
kZ; i.e., (E/Z)pss ) (kqZ/kqE)(kE/kZ), where kqE ) (kqR + kqS)/2.
Since no enantiodifferentiation was observed to occur during
the quenching process, the quenching rate constant (kqE) for
racemic 1E is virtually equal to kqS and kqR in the present case;
5
the concomitant development of exciplex emission at longer
wavelengths. The mechanism of the enantiodifferentiating pho-
tosensitized isomerization is illustrated in Scheme 2. The pho-
tosensitized geometrical isomerization has been shown to be
reversible, affording the same E/Z ratio at the photostationary
i.e., kqE ) kqS ) kqR * kqZ.
To elucidate the effects of pressure upon photosensitized
enantiodifferentiation, only three fixed irradiation periods were
employed, which are appropriate to check the initial, intermedi-
ate, and final stages of the photoisomerization. As can be seen
from Table 1, apparent pss’s are established in several cases
after irradiation for 60 min. We therefore discuss the efficiency
of photoisomerization by assuming that these E/Z ratios, (E/
5,48
state irrespective of the initial isomer composition employed.
However, the recycling paths, connecting 1Z, (R)-1E, and (S)-
E on the top and the bottom, are omitted in the scheme for
1
the sake of simplicity. The rate constants kqZ, kqR, and kqS re-
Z) , are equal or very close to the real pss values.
pss
1
spectively refer to the quenching of excited sensitizer ( S*) by
From the mechanistic point of view, it should be emphasized
that the observed pressure effect on the (E/Z)pss ratio depends
critically upon the sensitizer employed, and that the (E/Z)pss
ratios obtained are enhanced at high pressures without exception.
The first observation clearly rules out the possibility of pressure-
dependent decay ratios, indicating that the quenching step must
be responsible for the pressure-dependent (E/Z)pss ratios. Since
the (E/Z)pss or kqZ/kqE ratios increase with increasing pressure,
the activation volume for quenching by 1Z is inferred to be
smaller than that for 1E in all cases examined, facilitating the
quenching by 1Z and the production of 1E at high pressures.
1
Z, (R)-1E, and (S)-1E, while kR and kS refer to the rotational
1
relaxation of 1Z to twisted singlets (R)- and (S)- p within the
exciplex, and kZ and kE to the subsequent decay of p to 1Z
1
and 1E.
It has been further revealed that the product’s ee’s are
independent of conversion from the initial stage to the ultimate
photostationary state, and an attempted kinetic resolution starting
from racemic 1E gives only negligible ee’s during the early
stages of photoisomerization.5 These observations clearly rule
out any possible kinetic resolution at the quenching step (i.e.,
Interestingly, the sensitizers used may be classified into two
categories on the basis of the magnitude of the pressure effect.
(
48) Inoue, Y.; Takamuku, S.; Kunitomi, Y.; Sakurai, H. J. Chem. Soc.,
Perkin Trans. 2 1980, 1672.

Pressure Effect in Asymmetric Photochemistry
J. Am. Chem. Soc., Vol. 120, No. 41, 1998 10691
The increments in (E/Z)pss induced by increasing the pressure
from 0.1 to 200 MPa are less than 20% of the original values
in the majority of cases which employ benzoates 2, isophthalates
the solvent viscosity does not appreciably affect the product’s
ee, at least at atmospheric pressure. For instance, in the
5
photosensitizations of 1Z with 2a and 5a, practically identical
54
4
, and terephthalates 5 as sensitizers. On the other hand, the
op’s of 1E are given in pentane (η 0.225 mPa s) and in
54
(
E/Z)pss ratio is enhanced 100% or more at 200 MPa in the
cyclohexane (η 0.898) at 25 °C. Similarly, upon sensitizations
with 6a, the op values of 1E produced at 25 °C are -9.4, -8.8,
photosensitizations with phthalates 3, 1,2,4,5-benzenetetracar-
boxylates 6, and benzenehexacarboxylates 7. Thus, (E/Z)pss
ratios as high as 0.4-0.5, which are difficult to attain at ambient
5
4
54
and -8.7% in pentane (η 0.225), heptane (η 0.397), and
5
4
decane (η 0.861), respectively, and those obtained at -87 °C
are -28.9 and -30.7% in pentane (η 0.645) and heptane (η
1.22), respectively. We may conclude that the product’s ee
is governed not by the viscosity change caused by increasing
pressure, but through the direct effect of pressure upon the
enantiodifferentiating process in the photoisomerization sensi-
tized by chiral benzenepolycarboxylates.
4
8-51
55
temperatures as a result of the high strain energy of 1E,
5
5
are quite easily obtained at high pressure. Incidentally, the same
categorization of sensitizer has been proposed on the basis of
the large temperature dependence of the product’s ee obtained
in the enantiodifferentiating photosensitizations by the ortho
4
,5
esters 3, 6, and 7.
It is likely that each of the pressure and
temperature effects observed are entropy-related, and originate
from the conformational flexibility of mutually interacting ortho
substituents in 3, 6, and 7. In this context, it is interesting to
note that the pressure effect upon (E/Z)pss appears to be
particularly exaggerated when more substituted sensitizers with
less bulky ester groups such as 6c-e and 7e are used, probably
as a result of their greater compressibility arising from their
conformational flexibility. This rationalization may be justified
Activation Volume. To discuss the pressure effect upon the
product’s ee more quantitatively, it is necessary to calculate the
differential volume of activation for the enantiodifferentiating
photosensitized isomerization.
Rate constants are known to depend on temperature and
pressure. At a given temperature (T), the effect of pressure (P)
upon rate constant (k) is described by eq 1:
q
-
10
by the much higher compressibility of hexane (17.09 × 10
(∂ln k/∂P) ) -∆V /RT
(1)
T
2
-1 52
-10
2
-1 53
m N ) than that of cyclohexane (11.40 × 10
m N )
q
determined at 25 °C and 0.1 MPa.
where ∆V stands for the activation volume and R the gas
3
-1
-1
Pressure Effect upon Enantiodifferentiation. From the
pressure dependence of product’s ee, a more rigorous distinction
between the chiral sensitizers can be made between ortho and
nonortho benzene(poly)carboxylates. As shown in Table 1,
negligible pressure effects are found in the ee’s obtained upon
photosensitizations with all benzoates 2, isophthalates 4, and
terephthalates 5. In contrast, the use of ortho benzenepolycar-
boxylates (3, 6, and 7) as chiral sensitizers under comparable
conditions leads to drastically pressure-dependent ee’s, although
the outcome can either be positive or negative.
constant (8.314 cm MPa K mol ). Applying this general
equation to each enantiomeric photoisomerization process, which
affords (S)- and (R)-1E in the present case, we obtain:
q
S
(∂ln k
/∂P)
) -∆V
/RT
(2)
(3)
S
T
q
(∂ln k /∂P) ) -∆V /RT
R
T
R
Following on, the relationship between the relative rate constant
k /k and pressure is expressed by eq 4.
S
R
Interestingly, the product chirality is switched at the equipodal
pressure (P0) simply by changing the pressure applied upon
sensitization with benzeneteteracarboxylates (6a,c,d). In an
extreme case, the chiral sensitizer 6a gave (R)-1E, with an 11%
ee at atmospheric pressure, but the antipodal (S)-1E with an
q
[∂ln(k /k )/∂P] ) -∆∆V S-R/RT
(4)
S
R
T
Assuming a constant activation volume over the entire pressure
range employed, the relative rate constant is expressed as a linear
function of pressure (eq 5), in which the differential activation
1
8% ee at 400 MPa.
q
volume ∆∆V S-R is calculated from the proportionality coef-
Although the effects of pressure upon enantiodifferentiating
ficient.
photochemical reactions have not yet been explored and, as a
result, no such observations are found in the literature, an
unusual temperature switching of the product’s chirality has been
reported to occur in the same enantiodifferentiating photosen-
q
ln(k /k ) ) -(∆∆V S-R/RT)P + C
(5)
S
R
4
,5
S R
In eq 5, the integration constant C is equal to ln(k /k ) at
sitization system.
The temperature-switching phenomenon,
P ) 0.
also observed only with the ortho benzenepolycarboxylate
sensitizers, has been attributed to a significant contribution by
the entropy factor, arising from global conformational changes
of the mutually interacting ortho ester moieties, which are
believed to synchronize with the enantiodifferentiating rotational
relaxation of the double bond of 1 within the exciplex
Using eq 5, we plotted the logarithm of the relative rate
constant obtained experimentally from the ee value (i.e. kS/kR
(100 + % ee)/(100 - % ee)) as a function of the applied
pressure. In all cases, good straight lines were obtained over
the pressure range examined, as exemplified in Figure 1. The
differential activation volumes (∆∆V S-R) obtained from the
plots are listed in Table 2, along with the differential enthalpy
∆∆H S-R) and entropy (∆∆S S-R) of activation calculated from
the data obtained in the temperature-dependence studies previ-
)
q
5
intermediate. Judging from the similar tendencies shared by
the temperature and pressure effects, it is obvious that the
mutually interacting ortho ester groups play a crucial role in
determining the product’s ee.
The unprecedented pressure dependence of the product’s ee
cannot be attributed to increased viscosity at high pressure, since
q
q
(
5
ously reported.
From the linear P vs ln k plots obtained, it may be concluded
that the hydrostatic pressure causes no alterations to the
enantiodifferentiation mechanism, or to the activation volumes
(
(
(
49) Swenton, J. S. J. Org. Chem. 1969, 34, 3217.
50) Deyrup, J. A.; Betkouski, M. J. Org. Chem. 1972, 37, 3561.
51) Yamasaki, N.; Yokoyama, T.; Inoue, Y.; Tai, A. J. Photochem.
(54) Riddick, J. A.; Bunger, W. B. Organic SolVents, 3rd ed.; Wiley:
New York, 1970.
Photobiol., A: Chem. 1989, 48, 465.
(
(
52) Blinowska, A.; Brostow, W. J. Chem. Thermodyn. 1975, 7, 787.
53) Holder, G. A.; Walley, E. Trans. Faraday Soc. 1962, 58, 2095.
(55) Extrapolated viscosity at -87 °C, using the Bingham equation and
the parameters listed in ref 54 (p 31).

10692 J. Am. Chem. Soc., Vol. 120, No. 41, 1998
Inoue et al.
Table 3. Temperature and Pressure Effects upon
Enantiodifferentiating Photoisomerization of Cyclooctene (1Z)
Sensitized by (-)-Menthyl, (-)-Bornyl, and/or (R)-1-Methylheptyl
Pyromellitate (6a) and Mellitates (7a,b,e) in Pentanea
temp/ pressure/ irradiation 1E/
%
%
sensitizer °C
MPa
time/min
1Z conversion yield % ee
6
a
7
0.1
10
0.16
0.28
0.35
0.47
0.17
0.35
0.55
0.14
0.30
0.42
0.44
0.44
b
18
29
36
42
18
32
46
15
29
40
39
38
13
20
22
27
14
24
30
12
21
25
27
27
b
b
b
b
b
b
b
-3.8
-4.4
3
6
0
0
-6.1
1
2
00
00
60
10
+7.7
+16.7
+16.8
+16.5
+1.9
3
6
0
0
-9
0.1
10
3
6
0
0
+2.7
+1.6
2
5
5
0
60
60
10
+6.1
+9.4
7a
5
0.1
0.7
-14.4
-13.4
-12.3
-17.9
-12.9
-16.7
-17.1
3
6
0
0
b
3.6
2.0
1.8
3.8
2.5
3.8
2.8
1.0
b
b
1.6
b
b
b
S R
Figure 1. Pressure dependence of the relative rate constant k /k in
0.018
b
0.023
b
0.020
0.032
b
0.014
b
the enantiodifferentiating photoisomerization of (Z)-cyclooctene 1Z
sensitized by (-)-menthyl (a) and (-)-bornyl (b) benzoates 2a,b (O),
phthalates 3a,b (b), isophthalate 4a,b (4), terephthalate 5a,b (0),
benzenetetracarboxylates 6a,b (2), and benzenehexacarboxylates 7a,b
50
10
60
1
00
10
3
6
0
0
(9) in pentane at 25 °C.
3.1 -15.3
-6
0.1
30
b
b
b
b
b
b
b
b
b
-11.9
-12.7
-12.4
-10.4
-12.8
-12.1
b
Table 2. Activation Parameters at 25 °C Determined from
Temperature and Pressure Dependence of Enantiomeric Excess in
Enantiodifferentiating Photoisomerization of Cyclooctene (1Z)
Sensitized by Chiral Benzenepolycarboxylates (2-7) in Pentane
60
2
5
5
10
60
0.014
b
b
0
10
30
60
10
30
60
10
60
10
30
60
10
q
a
∆∆SqS-Ra/
q
∆
∆H S-R
/
∆∆V S-R
/
-
1
-1
-1
3
-1
0.016
0.008
0.026
0.048
0.008
0.043
0.009
0.026
0.049
b
0.015
0.030
b
0.026
b
0.017
0.029
0.034
0.091
0.139
0.032
0.153
0.033
0.100
0.173
0.022
0.066
0.116
0.017
0.111
0.023
0.068
0.132
sensitizer
kcal mol
cal mol
K
cm mol
7b
5
0.1
0.7
3.2
5.2
b
+16.2
+16.5
2
2
2
3
3
3
4
4
4
5
5
5
6
6
6
6
6
7
7
7
a
b
e
a
b
e
a
b
e
a
b
e
a
b
c
d
e
a
b
e
0.014
b
b
-0.19
-0.50
b
+0.1
b
b
-0.039
b
b
-0.51
-1.38
b
+0.1
b
b
-0.13
+0.16
+0.27
+0.83
+1.48
+0.39
+0.07
-0.03
-0.02
+0.36
+0.01
-0.12
-3.71
+0.29
+0.87
-0.99
-1.44
+3.50
-5.56
+0.56
4.5 +14.0
b
4.0 +18.3
b
b
4.6 +24.0
b
b
3.0 +20.8
b
2.4 +23.1
b
b
2.8 +25.3
b
50
100
+24.2
6.8
0.4
1.6
5.6
1.5
1.4
2.4
1.2
5.2
2.2
1.7
3.6
5.1
10.6
15.7
5.2
17.6
5.7
12.8
19.4
6.2
9.8
15.6
3.5
12.3
6.0
11.2
16.0
+27.6
+27.3
-9
0.1
+24.5
+21.3
3
6
0
0
+0.092
b
b
+0.077
b
b
2
5
5
10
+25.7
60
0
10
+28.7
+27.8
-0.77
-0.61
+0.29
-0.54
-0.54
-0.96
-0.86
-1.13
-3.00
-1.55
+1.01
-1.89
-1.93
-3.85
-2.60
-3.48
3
6
0
0
7e
4
0.1
0
10
+15.6
3
6
0
0
8.2 +14.4
11.7 +16.5
5
10
b
+16.0
60
12.6 +14.4
b
1
00
10
30
b
a
Reference 5. b Value not available.
8.7 +12.4
60
13.9 +12.0
-
10
0.1
10
b
+20.8
for these sensitizers over the pressure range employed, since
any change in mechanism caused by pressure would lead to
30
5.9 +21.7
60
10
9.8 +18.2
1
0
q
25
b
+16.1
nonlinearity. Naturally, the magnitude of ∆∆V is strongly
dependent upon the sensitizer employed. As listed in Table 2,
the nonortho sensitizers 2, 4, and 5 afford only small or
60
9.8 +17.9
b
6.0 +16.9
11.1 +15.9
50
10
b
30
q
3
-1
negligible |∆∆V | values (,1.0 cm mol ), whereas much
60
q
3
-1
greater |∆∆V | values (e1.5, 3.4, and 5.5 cm mol ) are
obtained with the ortho sensitizers (3, 6, and 7, respectively).
Although we could not find any appropriate data for
comparison in the literature, differential activation volumes of
a
Irradiation (>280 nm) conducted in a thermostated pressurized cell
with a sapphire window, using an unfiltered high-pressure mercury arc;
[1Z] ) 5 mM, [sensitizer] ) 1 mM. b Value not determined.
3
-1
up to 5.5 cm mol
obtained in the present study are
ment of 1,1-diphenyl-2-methyl-2-anisyloxirane in the presence
1
0b,57
unexpectedly large for the Z-E photoisomerization of such a
of (-)-menthol.
The pressure effects upon regio- or
q
simple cycloalkene in view of the small actual ∆V values of
diastereoselectivity have also been examined in some thermal
and photochemical reactions. Thus, both the thermal Diels-
3
-1
-
0.6 to +9.3 cm mol which are reported for thermal Z-E
56
isomerizations of more bulky indigo derivatives. A very small
∆V of -0.8 cm mol was reported for the Lewis acid-
(
56) Sueishi, Y.; Ohtani, K.; Nishimura, N. Bull. Chem. Soc. Jpn. 1985,
q
3
-1
∆
5
8, 810.
catalyzed enantiodifferentiating Wagner-Meerwein rearrange-
(57) Kraemer, H. P.; Plieninger, H. Tetrahedron 1978, 34, 891.

Pressure Effect in Asymmetric Photochemistry
J. Am. Chem. Soc., Vol. 120, No. 41, 1998 10693
Figure 2. Temperature and pressure dependencies of the relative rate constant k
/k
S R
in the enantiodifferentiating photoisomerization of (Z)-cyclooctene
1
Z sensitized by (-)-menthyl benzeneteteracarboxylates 6a (a) and (-)-menthyl (b), (-)-bornyl (c), and (S)-1-methylpropyl (d) benzenehexacar-
boxylates 7a,b,e in pentane.
Alder reactions of various dienes and alkenes10b and the
Table 4. Differential Activation Volume (∆∆V S-R)a for
Enantiodifferentiating Photoisomerization of Cyclooctene Sensitized
by Chiral Aromatic Esters in Pentane
q
photochemical [2+2] cycloadditions of cyclopentenones with
4
5
q
alkenes also give relatively small ∆∆V values of 0.5-4.5
3
-1
temperature/°C
and 0-2.0 cm mol , respectively. In the [4+2] photocy-
clodimerization of 1,3-cyclohexadiene, the ∆∆V value for endo/
q
sensitizer
25
7
4
5
-6
-9
-10
exo selectivity depends substantially upon the solvent and the
substrate concentration employed, affording small values of
6
7
7b
7e
a
a
-3.4 -5.2
+1.9
-6.8
+1.1 -2.6
-4.8
3
-1
1
.2-2.4 cm mol in acetonitrile, but unusually large values
-5.5
-4.1
3
-1
+0.6
+2.1
+2.1
of 9-12 cm mol in benzene as a result of the involvement
4
7
of different types of solvated ion pairs in the two solvents. It
is therefore inferred that, in the absence of any active participa-
tion of solvent molecules in pentane solution in the present
a
In cm mol-1.
3
even in a qualitative sense. This can be seen if a comparison
of the data for 3b, 6a, and 7a,b,e is made. The inconsistent
tendencies of ∆∆S and ∆∆V unequivocally show that the
temperature and pressure effects cannot be directly correlated
with one another, but instead behave as independent perturbing
factors on the enantiodifferentiation process.
q
study, the large absolute ∆∆V S-R values which are obtained
only for ortho sensitizers reflect possible differences in the
degree of conformational changes upon enantiodifferentiating
‡
‡
1
1
rotational relaxation of 1Z to (S)- p or (R)- p within the exciplex
intermediate, as shown in Scheme 2.
q
It is difficult to anticipate the sign and magnitude of the ∆∆V
q
We therefore attempted the determination of ∆∆V values at
value, as is the case for the ∆∆Hq and ∆∆Sq values.5,6
lower temperatures. Only a limited temperature range (down
to -10 °C) is available in the high-pressure experiments as a
result of pressure leakage caused by the inevitable contraction
of the pressure vessel. However, the high-pressure irradiations
at varying temperature not only reveal details and correlation
between the temperature and pressure effects upon this enan-
tiodifferentiating photosensitization, but also provide us with a
powerful, versatile tool for the multidimensional control of
enantioselectivity in photochemical reactions.
Temperature Dependence of Activation Volume. The
enantiodifferentiating photoisomerizations of 1Z sensitized by
a representative group of ortho esters 6a and 7a,b,e were
performed in pentane at two temperatures lower than 10 °C
under pressures ranging from 0.1 to 200 MPa. The results are
listed in Table 3. To facilitate a quantitative discussion, the
effects of pressure at varying temperatures, the ln(kS/kR) values
obtained at each temperature are plotted against the pressure,
affording straight lines for all sensitizers employed (see Figure
Nonetheless, the sensitizers possessing more bulky substituents
q
generally afford greater ∆∆V values, as demonstrated by the
series of sensitizers 6c-e with increasing chain length. Similar
q
q
general trends have been observed in the ∆∆H and ∆∆S values
reported for the same enantiodifferentiating photoisomerization
at varying temperatures.5 In this context, it would be interesting
to examine the general validity of the proposed linear relation-
q
q
ship between ∆S and ∆V , which was originally claimed for
5
8-60
some thermal inorganic reactions
and solvent-solute
interactions.6
1,62
However, it should be emphasized that, in our
case, the sign and magnitude of ∆∆V obtained with individual
sensitizer do not necessarily correspond with those of ∆∆S ,
q
q
(
(
58) Twigg, M. V. Inorg. Chim. Acta 1977, 24, L84.
59) Lawrance, G. A.; Suvachittanont, S. Inorg. Chim. Acta 1979, 32,
L13.
(
(
(
60) Palmer, D. A.; Kelm, H. Coord. Chem. ReV. 1981, 36, 89.
61) Phillips, J. C. J. Chem. Phys. 1984, 81, 478.
62) Phillips, J. C. J. Phys. Chem. 1985, 89, 3060.

10694 J. Am. Chem. Soc., Vol. 120, No. 41, 1998
Inoue et al.
Figure 3. Representative P vs T-1 vs ln(k
) diagrams. Schematic drawings of the ln(k
S
/k
R
S
/k
R
) value as a function of pressure and temperature: no
inversion of product chirality occurs in category 1; the product chirality is inverted either by temperature, affording an equipodal temperature (T
or by pressure (not shown, but essentially the same diagram can be drawn just by switching the axes) in category 2; the chirality is switched by both
pressure and temperature at the equipodal points P and T in category 3. In each category, all results should fall on the enantiodifferentiation plane
hatched), which intersects the P vs ln(k /k /k ) planes at lines A and B, while the vertical plane (gray) illustrates a typical
) and T- vs ln(k
0
),
0
0
1
(
S
R
S
R
q
pressure-dependence plot obtained at 25 °C, affording the -∆∆V /R value as a slope of the intersecting line C.
‡
2
). From these plots and eq 5, we can calculate the ∆∆V at
the lightly shaded enantiodifferentiation plane. The dependence
of ee upon temperature at 0 MPa gives the intersection B, of
slope -∆∆H /R, and intercept ∆∆S /R on the T vs ln(kS/kR)
plane.
each temperature for these sensitizers, the values of which are
listed in Table 4.
As can be seen from Figure 2 and Table 4, the ∆∆V values
obtained are highly dependent upon temperature, even over a
relatively narrow range. The temperature dependence of ∆∆V
is also a critical function of the sensitizer structure. Whereas
sensitizers 6a and 7e give greater absolute ∆∆V values at lower
temperatures, 7b shows the opposite behavior, and interestingly
the sign of ∆∆V for 7a is switched within this narrow
temperature range. The apparently puzzling dependency of ee
upon both pressure (as measured by ∆∆V ) and temperature
as measured by ∆∆H or ∆∆S ) can be graphically represented
and globally understood by a three-dimensional diagram depict-
ing the relationships between ee, T, and P.
The P vs T-1 vs ee Diagram: Correlated Temperature
and Pressure Effects upon ee. The ln(kS/kR) values have been
demonstrated experimentally to be proportional to the pressure
q
q
-1
q
Category 1 is illustrated in Figure 3a. The solid and dashed
bold lines (A and B), at which the enantiodifferentiation plane
q
-
1
intersects the P vs ln(kS/kR) and T
vs ln(kS/kR) planes,
q
respectively, afford very small slopes and intercepts which have
the same sign, and no inversion of product chirality is seen, or
is predicted at any pressure or temperature. Most benzoate,
isophthalate, and terephthalate sensitizers (2, 4, and 5), which
q
q
q
q
q
give very small ∆∆V , ∆∆H , and ∆∆S , fall in this category.
Thus, upon sensitization with these chiral benzene(di)carboxy-
lates, the product’s ee is not significantly improved by increasing
pressure or by decreasing temperature.
q
q
(
More interestingly, photosensitizations with ortho esters 3,
6
2
, and 7 belong to one of the latter two categories. In category
examples, the slope and intercept of line A possess the same
sign. However, the slope and intercept of line B have opposite
signs, affording the equipodal line D at the intersection of the
(
(
P) at a constant temperature, and also to the inverse temperature
T ) at a constant pressure. This allows us to construct a three-
-1
-1
-
1
enantiodifferentiation plane with the P vs T plane. Switching
dimensional P vs T vs ee diagram shown in Figure 3. The
observed behavior of ee as a function of pressure and temper-
ature, although apparently complicated and quite specific to each
sensitizer, can be classified into three categories based on the
presence/absence of an intrinsic equipodal temperature (T0) or
equipodal pressure (P0). In Figure 3, the ln(kS/kR) value is
expressed as a function of pressure (P) and inverse temperature
-
1
of the axes P and T in category 2 leads to an apparently
different diagram, which might be classified as an independent
category 2′ since the product’s chirality is inverted by pressure,
not by temperature.
Unfortunately the temperature and pressure effects do not
work cooperatively to enhance the product’s ee in category 2
(or 2′). As can be seen from Figure 3b, the applied pressure
-1
(
T ) giving the lightly shaded plane which crosses the P vs
ln(kS/kR) plane at intersection A (solid line) in the foreground
q
reduces the temperature dependence of ee or ∆∆H and 100%
-
1
-1
and the T vs ln(kS/kR) plane at intersection B (dashed line) in
the background. All enantiodifferentiating eVents in the pho-
tosensitizations in this study should occur on this enantiodif-
ferentiation plane. Our experiments, which considered the
pressure dependence of ee at, for example, 25 °C, are performed
on the vertical plane (dark shadow) in Figure 3. These give
ee could be attained when either P or T is infinite. The
occurrence of the equipodal line D with a positive slope in the
-1
P vs T plane simply means that the ee value will never become
zero at temperatures higher than T0.
The difference between categories 2 and 3 lies only in the
signs of slope and intercept, but the consequences of these
differences are crucial to obtain very high ee’s at reasonable
q
the -∆∆V /RT value as the slope of line C which intersects

Pressure Effect in Asymmetric Photochemistry
J. Am. Chem. Soc., Vol. 120, No. 41, 1998 10695
Table 5. Enantiomeric Excesses (%ee) Observed or Extrapolated at Several Pressures and Temperatures in the Photosensitization of
Cyclooctene (1Z) with (-)-Tetramenthyl 1,2,4,5-Benzenetetracarboxylate (6a)
pressure/MPa
temp/°C
0.1
25
50
100
200
300
400
1000
1500
a
a
a
2
5
7
9
-11.2
-6.1
+1.6
-3.6
+7.7
+5.7
+16.5
+12.4
+17.9 a
(+38.1)
(+57.1)a
(+56.6)a
(+76.8)
(+92.9)
(+98.3)
a
(+79.5)
-
+6.1
+9.4
(+45.5)a
(+92.1)a
a
Extrapolated value.
temperatures and pressures. In category 3, the slope and
intercept of lines A and B possess opposite signs, affording the
equipodal line D with a negative slope. The equipodal line
connecting T0 and P0 not only indicates that the ee value never
becomes zero at temperatures lower than T0 and at pressures
greater than P0, but also demonstrates that temperature and
pressure work cooperatively to significantly enhance the prod-
uct’s ee. Despite the limited number of chiral sensitizers
examined, such a fortuitous situation is realized in photosen-
sitizations with (-)-menthyl benzenetetracarboxylate 6a as the
sensitizer. As can be seen from Table 4, 6a affords fairly large
q
3
-1
|
∆∆V | values that are enhanced from 3.4 to 6.8 cm mol
simply as a result of lowering the temperature from 25 to -9
C. Thus, the product’s ee is anticipated to increase dramatically
°
in particular at low temperatures, ultimately giving almost
completely pure (S)-(+)-1E at -9 °C under 1500 MPa, as shown
in Table 5. These conditions are realistically attainable and
experiments along this line are currently planned, not only to
obtain the very high ee’s, but also to examine the linearity of
q
the P vs ln(kS/kR) plot, i.e., the pressure dependence of ∆∆V
at much higher pressures.
Conclusions
Figure 4. Calibration plot of the enantiomeric excess (% ee)
determined by chiral gas chromatography against the optical purity (%
op) determined by polarimeter using the same samples of (+)-rich,
(-)-rich, and racemic 1E, exhibiting an excellent correlation, with unit
slope and negligible intercept.
The present studies clearly indicate that, in addition to
temperature, external pressure can be used as a practical tool
for controling product’s ee in asymmetric photochemical
reactions. The combined effects of temperature and pressure
provide us with a versatile and powerful multidimensional
control that allows us to switch the product’s chirality more
easily than by the manipulation of either these factors on their
own. Although further examinations of a wide variety of
asymmetric (photo)reaction systems may be needed to generalize
these unprecedented chirality-switching phenomena caused by
changing temperature and/or pressure, it is evident that the
product chirality in photochemical, and probably thermal,
asymmetric syntheses is governed not only by the enthalpic
factor, symbolically represented by steric hindrance, but also
by the entropic factor, typically illustrated by structural freedom
in the transition state. Finally, we wish to emphasize that the
most crucial role is played by the entropy factor in photochemi-
cal and thermal processes that are governed by weak interactions
such as electrostatic, hydrogen-bonding, van der Waals, and/or
π-π stacking interactions.63 In such systems, the reaction itself
is mostly enthalpy-driven, while the selectivity is entropy-
controlled. It is under these conditions that external perturbants
such as temperature and pressure become effective tools to
control the weakly interacting crucial processes.
on a Shimadzu GC6A or GC14B instrument with a packed or chiral
capillary column. The temperature of the injection port was kept below
1
50 °C to avoid the thermal E-Z isomerization of (E)-cyclooctene 1E
in the irradiated solutions.
Materials. The pentane solvent was shaken repeatedly with
concentrated sulfuric acid until the acid layer no longer turned yellow,
then washed with water, neutralized with sodium carbonate powder,
and fractionally distilled. (Z)-Cyclooctene 1Z (Nakarai Tesque) was
purified by treatment with 20% aqueous silver nitrate solution, followed
by fractional distillation. The purified sample of 1Z contained 3-4%
cyclooctane as the sole detectable impurity, but was completely free
of the (E)-isomer and 1,3- and 1,5-cyclooctadienes.
Optically active alkyl benzoates 2, phthalates 3, isophthalates 4,
terephthalates 5, 1,2,4,5-benzenetetracarboxylates 6, and benzenehexa-
carboxylates 7 were prepared by mixing the corresponding acid
chlorides and optically active alcohols in pyridine, and purified by
fractional distillation or repeated recrystallization from ethanol, ac-
cording to the method reported previously.5
Photolysis. All irradiations were conducted in a pressure vessel
HKP-921208 designed and manufactured by Hikari Koatsu Co., which
was equipped with a sapphire window (5 mm i.d.) for external irra-
diation, and also with a coolant circulation system in the body of the
3
reactor. A solution (11 cm ) of (Z)-cyclooctene (5 mM), containing
Experimental Section
optically active sensitizer (0.2-1.0 mM) and cycloheptane added as
an internal standard, was placed in the vessel, and pressurized up to
Instruments. Specific rotations were determined on a Perkin-Elmer
model 243B polarimeter with a temperature-controlled 10 cm cell. Gas
chromatographic (GC) analyses of photolyzed samples were performed
4
00 MPa with a high-pressure pump KP5B, thermostated at a constant
temperature ((0.1 °C) between +25 and -10 °C by circulating water
or water-methanol coolant through the reactor body. The solution
was then irradiated for a given period of time with a 250-W high-
pressure mercury arc (Ushio UI-501C). The collimated incident beam
(
63) Inoue, Y.; Wada, T. AdV. Supramol. Chem. 1997, 4, 55.
(64) Cope, A. C.; Ganellin, C. R.; Johnson, H. W. J.; Van Auken, T. V.;
Winkler, H. J. S. J. Am. Chem. Soc. 1963, 85, 3276.

10696 J. Am. Chem. Soc., Vol. 120, No. 41, 1998
Inoue et al.
from the lamp housing was focused with a quartz lens (f 16.5 cm)
which was placed in front of the sapphire window, allowing an efficient
irradiation.
onstrated that the ee values determined by chiral GC were highly
reproducible within (0.5% ee and in excellent agreement with the
optical purities calculated from the specific rotations of the same
samples, as long as the integrated area of each peak falls within a certain
range (20 000-200 000 µV s) on a Shimadzu C-R6A integrator, as
shown in Figure 4.
The conversion, chemical yield, and E/Z ratio of photolyzed samples
were determined by GC on a 3-m packed column of 25% â,â′-
oxydipropionitrile at 65 °C. (E)-Isomer 1E produced was selectively
3
extracted from the irradiated solution with a 2 cm portion of 20%
5
,62
Acknowledgment. We are grateful to Asahi Glass Founda-
tion and Nagase Science and Technology Foundation for their
support in the undertaking of this work. We thank Dr. Simon
Everitt at the Inoue Photochirogenesis Project for corrections
and improvements arising from a preliminary reading of this
manuscript.
aqueous silver nitrate at <5 °C.
The aqueous extract was washed
3
with two 1 cm portions of pentane and then added to concentrated
ammonium hydroxide (1 cm ) at 0 °C to liberate a chemically pure
3
sample of 1E (>99%), which was in turn extracted with a small portion
3
(
0.1-0.5 cm ) of pentane. The product’s ee in the pentane extract was
determined by chiral capillary GC over a Supelco â-DEX 120 column
30 m × 0.25 mm i.d.) at 60 °C, which gave a satisfactory enantiomeric
separation of (R)- and (S)-1E. In separate experiments, it was dem-
(
JA981929A