Asymmetric photolysis of 2-phenylcycloalkanones with circularly polarized light: A kinetic model for magnetic field effects

Article abstract of DOI:10.1246/bcsj.75.1223

Magnetic field effects (MFE) on asymmetric photolysis involving a biradical intermediate have been investigated on a kinetic model where an intersystem crossing (ISC) of the intermediate is taken into account. Changes in the enantiomeric excesses (ee) of chiral substances with circularly polarized light (CPL) irradiation have been simulated, and the necessary conditions for observing the MFE were obtained. The asymmetric photolysis of racemic 2-phenylcycloalkanones (2-PCAs) with CPL has been carried out in both the presence and absence of a magnetic field. Since the anisotropy g factors of 2-PCAs are considerably large, the CPL-induced ee are achieved to a few percent after 90% decomposition though the MFE are not observed. The photolysis mechanism of 2-PCAs in an air-saturated solution has been also clarified. Triplet acyl-benzyl biradicals, formed via photochemical α-cleavage of 2-PCAs, react with O₂ dissolved in the solution and result in the formation of acetophenone and alkenoic acids. The bimolecular reactions are diffusion-controlled and the rates are comparable to those of ISC to the singlet biradicals for all 2-PCAs. The recombination yields of the biradicals are sufficiently large. However, the biradical ISC rate shows little magnetic field dependence, which explains the absence of the MFE in this asymmetric photolysis.

Full text of DOI:10.1246/bcsj.75.1223

Bull. Chem.
Soc. Jpn.
75
6
© 2002 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 75, 1223–1233 (2002) 1223
The Chemical
Society of Ja-
pan
The Chemical
Society of Ja-
pan
GP
6
15
Asymmetric Photolysis of 2-Phenylcycloalkanones with Circularly
Polarized Light: A Kinetic Model for Magnetic Field Effects
Asymmetric Photolysis of 2-PCAs with CPL
S. Kohtani et al.
Shigeru Kohtani,* Masahide Sugiyama, Yoshihisa Fujiwara,† Yoshifumi Tanimoto,† and
2002
1223
1233
BCSJA8
0009-2673
01311
9
Ryoichi Nakagaki††
Graduate School of Natural Science and Technology, Kanazawa University, Takara-machi, Kanazawa 920-0934
†Graduate School of Science, Hiroshima University, Kagamiyama, Higashi-Hiroshima 739-8526
††Faculty of Pharmaceutical Sciences, Kanazawa University, Takara-machi Kanazawa 920-0934
4
2001
2
(Received September 4, 2001)
7
2002
Magnetic field effects (MFE) on asymmetric photolysis involving a biradical intermediate have been investigated
on a kinetic model where an intersystem crossing (ISC) of the intermediate is taken into account. Changes in the enanti-
omeric excesses (ee) of chiral substances with circularly polarized light (CPL) irradiation have been simulated, and the
necessary conditions for observing the MFE were obtained. The asymmetric photolysis of racemic 2-phenylcycloal-
kanones (2-PCAs) with CPL has been carried out in both the presence and absence of a magnetic field. Since the aniso-
tropy g factors of 2-PCAs are considerably large, the CPL-induced ee are achieved to a few percent after 90% decompo-
sition though the MFE are not observed. The photolysis mechanism of 2-PCAs in an air-saturated solution has been also
clarified. Triplet acyl–benzyl biradicals, formed via photochemical α-cleavage of 2-PCAs, react with O2 dissolved in the
solution and result in the formation of acetophenone and alkenoic acids. The bimolecular reactions are diffusion-con-
trolled and the rates are comparable to those of ISC to the singlet biradicals for all 2-PCAs. The recombination yields of
the biradicals are sufficiently large. However, the biradical ISC rate shows little magnetic field dependence, which ex-
plains the absence of the MFE in this asymmetric photolysis.
1.1.5
Absolute asymmetric synthesis is a methodology used for
asymmetric induction under physical forces, and has been at-
tracted much attention in relation to the origin of biomolecular
homochirality.1–3 Among the physical forces, circularly polar-
ized light (CPL) is the most extensively studied and well docu-
mented chiral source.1–5 Enantiomeric excess (ee) induced by
CPL irradiation is brought about through the preferential exci-
tation of one of enantiomers in a racemic mixture because of
circular dichroism of chiral substances. Kinetic studies of var-
ious type of asymmetric photoreactions with CPL have been
carried out both theoretically and experimentally.4–9 On the
contrary, the role of a magnetic field for absolute asymmetric
synthesis has been controversial for a long time.2,3 Recently,
Rikken and Raupach successfully demonstrated an enantiose-
lective photo-deracemization of chiral tris[ethanedioato(2−)-
O, Oꢀ]Cr(Ⅲ) complex by applying a static magnetic field either
parallel or anti-parallel to the propagation direction of light,
though the observed ee was extremely small (ca. 10−2% ee at
10 T) compared to that achieved by CPL irradiation.10
On the other hand, it has been known that a magnetic field
affects chemical reactions involving radical pair or biradical
intermediates where the rate of an intersystem crossing (ISC)
of the intermediates is influenced by a magnetic field.11–17
End-product yields in some photochemical reactions of bifunc-
tional chain molecules are also altered via a magnetic-field-de-
pendent spin-conversion process of the intermediates.18–20
These magnetic field effects (MFE) are closely related to spin
dynamics in which singlet-triplet state mixing is induced by
hyperfine coupling and electronic Zeeman mechanism.
There have been two notable studies concerning MFE on
asymmetric reactions and the radical pair mechanism. First,
Lem and Turro reported enantioselective radical pair recombi-
nation in the photolysis of meso- or dl-2,4-diphenylpentan-3-
one within a supercage of zeolites which was modified with a
chiral inductor (diethyl L-tartrate or ephedrine).21 They
claimed that when the carbon atom at the ketone is replaced by
13C, the enantioselectivity is enhanced by increasing the ISC
rate in an α-methylbenzyl and (α-methylbenzyl)acyl radical
pair generated in the supercage, though external MFE were not
observed. Second, Hegstrom and Kondepudi have predicted
dramatic MFE in a chirally autocatalytic reaction involving the
ISC process of radical pairs from a computational kinetic sim-
ulation.22 However, it seems to be difficult to demonstrate
their prediction because the chirally autocatalytic reaction is
presently a rare phenomenon.23,24
Our strategy to obtain large MFE in asymmetric reactions is
that the magnetic-field-dependent ISC process of radical pairs
or biradicals is incorporated into asymmetric photolysis with
CPL. While CPL is the only chiral source, the magnetic field
has no chirally asymmetric influence.2,3 However, the magnet-
ic field can change the ee of substances kinetically by affecting
the ISC rate. In this paper, we propose a kinetic model involv-
ing the ISC process, and discuss our simulated ee changes in
initially racemic substances with CPL irradiation in order to
understand the conditions for observing MFE. We then de-
scribe experimental attempts to observe the MFE in an asym-

1224 Bull. Chem. Soc. Jpn., 75, No. 6 (2002)
Asymmetric Photolysis of 2-PCAs with CPL
Scheme 1. Photolysis mechanism of 2-phenylcycloalkanones (2-PCAs, n = 6–12,15).
and/or isomerization reaction proceed upon left- or right-CPL
irradiation. Therefore, enantiomeric enrichment for another
enantiomer is achieved. The degrees of the preferential excita-
tion is defined by the Kuhn anisotropy g factor, (εL − εR)/ε =
∆ε/ε, where ε = (εL + εR)/2. This factor determines ee at a
certain extent of decomposition upon monochromatic CPL ir-
radiation.4–9
Scheme 2 illustrates a kinetic model for asymmetric photo-
lysis involving the ISC of radical pair or biradical intermedi-
ates. This reaction scheme is based on a photochemical reac-
tion of 2-PCAs in Scheme 1. The photocleavage of a chiral
compound (R and S enantiomer) takes place in the excited trip-
let state, thereby resulting in the formation of a triplet biradical
(3X). The initially prepared triplet biradical undergoes two
competitive processes, namely, an escape process to Product Y
and ISC to a singlet biradical (1X). Finally, recombination and
cage product formation from 1X take place to yield the starting
R and S enantiomers with equal rate constants (kc) and Product
Z, respectively. The formation rates of 3X from the starting R
and S enantiomer are different upon CPL irradiation as de-
scribed below:4,8
Scheme 2. Schematic diagram of asymmetric photolysis in-
volving ISC process of radical pairs or biradicals (3X: trip-
let, X: singlet). The intersystem crossing rate (kisc) de-
1
pends on magnetic field strengths.
metric photolysis of 2-phenylcycloalkanones (2-PCAs, n = 6–
8) induced by CPL irradiation. The reasons for selecting this
reaction system are as follows: (1) 2-PCAs are expected to
have a large Kuhn anisotropy g factor because of an asymmet-
ric carbon next to the carbonyl group in the cycloalkanone
ring.4,5 In general, chiral cyclic ketones, such as 2-PCAs, have
a large anisotropy g factor because the nπ* transition of carbo-
nyl group is electronically forbidden, but magnetically al-
lowed.25 (2) The photolysis of 2-PCAs involves the ISC pro-
cess of acyl-benzyl biradical as shown in Scheme 1, which has
been clarified by chemically induced dynamic nuclear polar-
ization (CIDNP),26–28 laser flash photolysis,29–33 and photo-
products analyses.34–38 (3) Magnetic-field-dependent CIDNP
and MFE on the ISC rates of acyl–benzyl biradicals are ob-
served in the photolysis of small (n = 6–8) and large (n = 10–
12) ring size 2-PCAs.28,31–33 We further investigated the pho-
tolysis mechanism of 2-PCAs in air-saturated solutions by us-
ing a GC/MS photoproduct analyses, transient absorption mea-
surements, and by examining the time evolution of the enanti-
omer concentrations for the starting species based on our ki-
netic simulation. The necessary conditions for observing large
MFE derived from the kinetic simulation are compared with
the experimental results obtained in this work. Finally, the
magnetic field control of photoinduced ee is discussed in some
detail.
1−10−A
(
)



IR = 1000I0η
IS = 1000I0η
[ ]
εR R
(1)



A
1−10−A
(
)


(2)
[ ]
εS S



A

where I0 is the incident light intensity (einstein/s•cm2). εR and
εS are the molar extinction coefficients for R and S enantiomers
toward left- or right-CPL at the irradiation wavelength, respec-
tively. A is absorbance (εR[R] + εS[S]), and η is the quantum
3
yield of X from the excited starting material. Kinetic equa-
tions for the concentration changes in R, S, 3X, and 1X are de-
scribed as follows (see Scehme 2):
d
[ ]
R
= −IR +k 1X
(3)
(4)
[
]
c
dt
Kinetic Simulation
d S
[ ]
In asymmetric photolyses with CPL, one of both enanti-
omers is preferentially excited and subsequent decomposition
= −IS +k 1X
[
]
c
dt

S. Kohtani et al.
Bull. Chem. Soc. Jpn., 75, No. 6 (2002) 1225
d
3 X
[
]
= IR +IS − k +k 3 X
(5)
(6)
(
)
[
]
isc
y
dt
1
d X
[
]
3
= kisc X − 2k +k 1X .
(
)
[
[
]
]
c
z
dt
Assuming that the concentration changes in 3X and 1X are neg-
ligible in a steady-state approximation, i.e. d[3X]/dt = 0 and
d[1X]/dt = 0, the differential equations for [R] and [S] are de-
rived as follows:
1−10−A
d
[
R
]

kisc

(
)






(7)
= −1000I0η
εR
[
R
]
−φ
A




dt
A
kisc +ky
1−10−A
d
[
S
]

kisc

(
)






= −1000I0η
εS
[
S
]
−φ
A
(8)


dt
A
kisc +ky


Fig. 2. The plots of ee versus logarithmic scales of kisc at 330
s. The rate constant kisc and the recombination yield φ are
treated as variable parameters; the φ values; (ꢀ) 0, (ꢁ) 0.1,
(ꢂ) 0.2, (×) 0.3, (ꢃ) 0.4, (ꢄ) 0.5. The other parameters
are set to the same values as in the case of Fig. 1.
where φ is the recombination yield from 1X to each enantiomer
( 0 < φ = kc / (2kc + kz) < 0.5).
The second terms in equations (7) and (8) containing ky, kisc,
and φ exhibit the magnetic field dependence if kisc is changed
by applying a magnetic field. When ky is zero, kisc in the sec-
ond terms is canceled. Therefore, the presence of the escape
process from 3X is indispensable for observing the MFE. Be-
sides, in the case of φ = 0, because the second terms vanish,
sentative values of kisc. Here, ky is set to be 1×107 s−1 and φ is
1
fixed at 0.5, which means that all X equally recombine to
form each enantiomer. It is apparent that the values of ee de-
pend on kisc and increase with decreasing kisc. Consequently,
large MFE are expected in this kinetic model.
1
large recombination yield from X to each enantiomer is also
necessary. In addition, it is the most important for obtaining
large ee that the g factor should be large enough, as mentioned
earlier.
Computer simulations of the concentration changes in enan-
tiomers were made by numerical integration of differential
equations (7) and (8) by the Runge–Kutta method.39 Here, the
values of εR and εS are fixed at 49 and 51 M−1 cm−1, respec-
tively, which corresponds to the g factor of 0.04. This value is
comparable to typical g values of chiral carbonyl compounds
reported in the literature.5
In order to obtain detailed conditions for the MFE, we eval-
uated ee as a function of kisc and φ. Fig. 2 shows plots of the ee
versus logarithmic values of kisc for each φ at 330 s. Note that
the variation in ee becomes larger with increasing φ. This
means that the large recombination yield is necessary for ob-
serving the MFE. In addition, it is noteworthy that the varia-
tion is significant in the range of log (kisc) = 6–7, which is
close to the escape rate (ky = 1×107 s−1). Therefore, the ISC
rate of a radical-pair or biradical should be nearly equal to the
rate of an escape process. Moreover, the ISC rate should be
changed over one order of magnitude upon applying a magnet-
ic field, even if the recombination yield is sufficiently large.
The simulation results of ee as a function of time are shown
in Fig. 1. Lines a–c indicate the increase in ee at three repre-
Experimental
Sample Preparations: 2-Phenylcyclohexanone (2-PCH, n =
6, Aldrich) and 2-phenylcycloheptanone (2-PCHep, n = 7, Ald-
rich) were used after purification by column chromatopraphy. 2-
Phenylcyclooctanone (2-PCO, n = 8) was prepared by the reac-
tion of an enolate ion of cyclooctanone (Tokyo Kasei Kogyo) with
diphenyliodonium trifluoromethanesulfonate (Tokyo Kasei Ko-
gyo) in the presence of copper(Ⅰ) cyanide (Katayama Kagaku), as
reported by Ryan and Stang.40 Purification of synthesized 2-PCO
was carried out by column chromatography (SiO2, 6% ethyl ace-
1
tate in hexane) to yield a colorless oil (6.4% yield, H-NMR
(CDCl3) δ 7.28 (m, 5H), 3.79 (dd, J = 12.0, 4.2 Hz, 1H), 2.65–
1.20 (m, 12H)). 2-Propanol of spectroscopy grade or HPLC grade
(Nacalai Tesque) was used as received. Degassed solutions were
made by several freeze-pump-thaw cycles. O2-saturated solutions
were prepared by bubling pure O2 gas for 20 min before experi-
ments.
Optical resolutions were conducted on a HPLC system (Shi-
madzu LC-6AD pump, a DGU-12A 2-line degasser, a CTO-10AS
column oven, and a SPD-10A UV-vis detector) equipped with a
normal-phase chiral column (Daicel Chiralcel OF for 2-PCH and
Fig. 1. Simulation results of the ee as a function of time.
The rate constant kisc is treated as variable parameter; (a)
1×105 s−1, (b) 2×106 s−1, (c) 1×107 s−1. I0 and η are as-
sumed to be 1×10−7 einsteins•cm−2 s−1 and unity, respec-
tively. Values of εR and εS are fixed at 49 and 51 M−1 cm−1
,
respectively. The rate constant ky is set to be 1×107 s−1
and the recombination yield φ is fixed at 0.5. Initial con-
centrations are 10 mM for both enantiomers.

1226 Bull. Chem. Soc. Jpn., 75, No. 6 (2002)
Asymmetric Photolysis of 2-PCAs with CPL
Daicel Chiralpak AS for 2-PCHep and 2-PCO, 4.6 mm i.d. × 250
mm). 2-propanol/hexane (1:9) at a flow rate of 0.5 ml/min was
used as an eluent. A chromatogram was monitored at 265 nm.
Spectroscopic Measurements: UV absorption spectra were
measured on a Hitachi U-3210 spectrophotometer. CD spectra
were obtained on a JASCO J-725S spectropolarimeter. Optical ro-
tations were measured using a Horiba SEPA-300 polarimeter with
a 10 cm path length.
Air-saturated sample solutions (4 mM) in a cylindrical quartz cell
(φ = 8 mm, 10 mm path length) were placed between the pole
pieces of an electromagnet (Tokin SEE-9G). The enantiomeric
excess and the extent of 2-PCAs photodecomposition were deter-
mined on a gas chromatograph (Shimadzu GC-14A) equipped
with a chiral capillary column (Supelco γ-DEX 225, 30 m × 0.25
mm i. d.). Helium carrier gas was used with a linear velocity of 30
cm/s. A sample solution (1 µL) was injected using a splitless
mode. The injector temperature was 250 °C and the FID tempera-
ture was 300 °C. The column temperature program began at 120
°C, was held for 2 min and then ramped at 2 °C/min to 180 °C. 1,
10-Decanediol or 1, 12-dodecanediol was used as an internal stan-
dard.
Transient Absorption Measurements: The
experimental
apparatus for transient-absorption measurements were similar to
those reported previously.12–16 In the case of a magnetic field
strength below 1 T,12,15,16 an excimer laser (Lambda Physik Com-
pEX 100, XeCl at 308 nm, fwhm ca. 15 ns, 100 mJ/pulse) was
used as the exciting light source and a xenon flash lamp (EG&G
FXQ-853) was used as the monitoring light source. Sample solu-
tions were placed between the pole pieces of an electromagnet
(Tokin SEE-9G). The magnetic field strength was controlled by a
current-regurated power supply. In the case of a high field
strength, a pulse magnet-laser flash photolysis apparatus was
used.13,14 A pulsed magnetic field (1–8 T, 2ms) was generated by
supplying an intense pulsed current from a condenser bank
(Nichikon, 5 kV, 50 kJ) to a home-made solenoid (9.7 mH and 0.7
Ω). The exciting and monitoring light sources were the fourth
harmonics (266 nm, fwhm 4–6 ns, 10 mJ/pulse) of an Nd:YAG la-
ser (Spectra Physics, GCR-11-1) and a Xe arc lamp (Ushio, 150
W), respectively. The laser pulse and the monitoring light were
introduced coaxially to a sample cell, and then light was passed
through the sample solution and then detected using optical fibers.
All of the measurements were carried out at 20–23 °C.
Photoproducts Analyses: O2-saturated 2-PCAs solutions (2
mM) in a cylindrical quartz cell (φ2.5 cm × 5 cm path length)
were irradiated with UV light from a 500-W high-pressure mercu-
ry lamp (Ushio) equipped with a glass filter (Toshiba UV-29) for
20 min for 2-PCH and 2-PCHep and 10 min for 2-PCO. After ir-
radiation, the solutions were concentrated to about 1/10 volume.
Photoproduct analyses were carried out on a multi-channel UV-
detected HPLC system (a Waters 510 pump, a Shimazdu CTO-
10AS column oven, a Otsuka Electronics MCPD 3600 multi-
channel photodetector, 2-propanol/hexane (1:9) eluent with a flow
rate of 0.5 mL/min) with a chiral column (Daicel Chiralcel OF)
Results
1. Optical and Chiroptical Properties:
The three 2-
PCAs were completely resolved using chiral HPLC methods.
The absolute configuration of the resolved 2-PCH was deter-
mined by comparing the sign of the optical rotation of our
sample solutions with reference data.41,42 As for 2-PCHep and
2-PCO, the sign of the optical rotation has not yet been corre-
lated with the absolute stereochemical configuration.
The absorption, CD, and anisotropy g factor spectra of 2-
PCAs are shown in Fig. 3. The absorption spectra (a–c) of
these compounds mainly consist of a broad nπ* band of carbo-
nyl group in 280–330 nm region and a vibronic ππ* band of
1
phenyl group (1A1g → B2u) around 260 nm. Intensity of the
forbidden nπ* transition increases with size of the cycloal-
kanone rings. This is due to an amplification of the nπ* transi-
tion probability by an interaction with the ππ* transition of
phenyl group caused by a ring distortion of 2-PCHep and 2-
PCO.43 Absorption bands (< 250 nm) which appeared for 2-
PCHep and 2-PCO can probably be ascribed to the second ππ*
1
transition of phenyl group (1A1g → B1u), which may be cou-
pled with the nπ* transition of the carbonyl group because CD
signals are observed in this region.43
Since the nπ* band of the carbonyl groups mainly contrib-
utes to the CD spectra of 2-PCAs (see spectra d–f in Fig. 3),
the anisotropy g factor calculated from CD and absorption
spectra indicates the maximum values around 300 nm. This is
the reason why we chose the excitation wavelength to be 308
nm for all 2-PCAs. The molar extinction coefficients (ε), CD
intensities (∆ε), and calculated g factors (∆ε/ε) at 308 nm are
summarized in Table 1. The g value of 2-PCH is the largest
because of the smallest nπ* transition intensity. Therefore, the
ee value of 2-PCH induced by CPL irradiation is expected to
be the largest of the three 2-PCAs.
2. Photoproducts in the Presence of Molecular Oxygen:
The photolysis mechanism of 2-PCAs is depicted in Scheme 1,
in which the Norrish type Ⅰ photocleavage of the excited triplet
2-PCAs yields the acyl–benzyl biradical. Four types of prod-
uct formation processes have been identified as (1) recombina-
tion to the starting 2-PCAs,26–28 (2) intramolecular dispropor-
tionation (hydrogen abstraction) to phenylalkenals,32,34-38 (3)
para-coupling to cyclophanes for large ring 2-PCAs (n = 10–
15),32,36,37 and (4) bimolecular radical scavenging processes
(escape process) with CCl4, CBrCl3 or O2.27,28 In these pro-
cesses, the escape process from the triplet acyl–benzyl biradi-
cal is indispensable for observing the MFE, as mentioned ear-
and
a GC/MS instrument (HP-5890GC/JEOL JMS AX5)
equipped with a chiral capillary column (Supelco γ-DEX 225, 30
m × 0.25 mm i. d.). Helium carrier gas was used with a linear ve-
locity of 30 cm/s. A sample solution (3 µL) was injected with a
split ratio of 20:1. The injector temperature was 250 °C, and the
column temperature program, which began at 100 °C, was held for
2 min and ramped at 3 °C/min to 190 °C. The mass spectrometer
was set to scan mass units of 20–350 for electron impact ioniza-
tion with a source temperature of 230 °C. Chromatograms were
obtained at a rate of 36 scan/min.
Photolyses with Linearly and Circularly Polarized Light:
The XeCl pulsed laser beam (Lambda Physik CompEX 100; fre-
quency, 10 Hz; fwhm, ca. 15 ns; 308 nm, 100 mJ/pulse) was first
converted into linearly polarized light (LPL) by passing through a
circular iris (φ = 10 mm) and a polarizer (Sigma Koki, glan laser
prism, 10 × 10 mm, 21 mm thick). LPL, left- and right-CPL were
obtained by passing the incident LPL through a phase-adjusted
Babinet-Soleil compensator (Oyodenko, φ = 8 mm) oriented with
angles of 0° for LPL, −45° for left-CPL, and +45° for right-CPL
with respect to the plane of polarization of the incident LPL. The
average output of the final polarized light was ca. 2 mJ/cm2•pulse.

S. Kohtani et al.
Bull. Chem. Soc. Jpn., 75, No. 6 (2002) 1227
Fig. 3. Absorption (a)–(c), CD (d)–(f), and g factor (g)–(i) spectra of 2-PCAs in 2-propanol in near UV region (240–340 nm). The
faster effluent enantiomers of 2-PCHep and 2-PCO on a chiral-capirally GC are indicated by broken lines (fast) in CD and g factor
spectra [(e), (f), (h) and (i)]. The slower ones for 2-PCHep and 2-PCO are also indicated by solid lines (slow) in the spectra [(e),
(f), (h) and (i)].
Table 1. Molar Extinction Coefficients (ε), Circular Dichro-
ism (∆ε), and Anisotropy g Factors of 2-PCAs at 308 nm
spectrum recorded for peak (c) on UV-detected HPLC indi-
cates that this component has a β-alkylstyrene chromophore.
Therefore, peak (c) is assigned to 6-phenyl-5-hexenal (Ph-
CHwCH(CH2)3CHO), which is mainly obtained in a degassed
solution.34–38 The peaks in the chromatograms of 2-PCHep
and 2-PCO were also identified in the same manner for 2-PCH.
Pairs of enantiomers for 2-PCHep and 2-PCO are also separat-
ed well. Other trace peaks in the three chromatograms have
not been identified. No peaks assigned to phenylalkanoic ac-
ids were detected in this work, whereas Doubleday reported
CIDNP signals due to these species, which may derived from a
ketene intermediate.26 The ketene was considered to be
formed through disproportionation (hydrogen abstraction from
the methylene group next to the acyl moiety by the phenylalkyl
radical).
Acetophenone and alkenoic acids were not detected in de-
gassed solutions at all. It is further worth noting that the peak
intensities of phenylalkenals (c) decreased with increasing O2
concentration, in contrast to increasing those of acetophenone
(b) and alkenoic acids (a). This opposite dependence on the O2
concentration can be explained by the reaction of the triplet bi-
radical with molecular oxygen dissolved in solutions, and
thereby the product yield of phenylalkenals is decreased with
2-PCAs
ε/M−1 cm−1
∆ε/M−1 cm−1
g factor
2-PCH
2-PCHep
2-PCO
36.6
117.2
249.1
1.1
2.3
4.2
0.030
0.020
0.017
lier. Therefore, we are especially interested in the O2 concen-
tration dependence on photoproduct formations which can be
investigated by a GC/MS analysis.
Figure 4 shows GC/MS total ion chromatograms of the pho-
toproducts in O2-saturated solutions. The retention times on
GC and the mass-to-charge ratio (m/z) of molecular and frag-
ment ions of the photoproducts and starting 2-PCAs are sum-
marized in Table 2. In the chromatogram of 2-PCH, two peaks
((R) and (S)) are assigned to the starting R and S enantiomers
of 2-PCH, indicating that a pair of enantiomer is well-separat-
ed on this chiral capillary GC. Peaks (a) and (b) are attributed
to 3-butenoic acid and acetophenone by a comparison with an
authentic sample. Peak (c) has a mass spectrum with a molec-
ular ion at m/z 174, which suggests that this component is due
to an isomer of the starting material. Additionally, absorption

1228 Bull. Chem. Soc. Jpn., 75, No. 6 (2002)
Asymmetric Photolysis of 2-PCAs with CPL
Fig. 5. Transient absorption decay profiles of 2-PCH moni-
tored at 320 nm in degassed (upper), air saturated (middle),
and O2 saturated (lower) solutions. The excitation light
source was a XeCl eximer laser (100 mJ/pulse, fwhm ca.
15 ns, 308 nm).
Fig. 4. Total ion chromatograms of photoproducts and start-
ing 2-PCAs in O2 saturated 2-propanol solutions after UV
irradiations (> 290 nm) of 20 min for 2-PCH and 2-
PCHep and 10 min for 2-PCO. See Table 2 for assignment
of peaks. Extents of sample decompositions were 54% for
2-PCH, 86% for 2-PCHep and 80% for 2-PCO. The chro-
matograms were obtained at the rate of 36 scan/min. The
detailed analytical conditions are described in the text.
in methanol and cyclohexanol reported by Doubleday and Tur-
ro et al.30,32 From previous studies,30–33 ISC is regarded as be-
ing the only deactivation process of the triplet biradicals in de-
gassed solutions. In addition, reverse ISC from singlet to trip-
let biradical is negligible because the rates of recombination
and disproportionation from the singlet biradical are much
larger than that of the ISC. Therefore, the reciprocal of τdeg
corresponds to the ISC rate constant (kisc) of the triplet biradi-
cal. In air-saturated solutions, however, because the escape re-
action with O2 is the second deactivation process of the triplet
biradical, the reciprocal of the lifetime (τair) is expressed as
1/τair = kisc + kesc[O2], where kesc is a bimolecular rate constant
and [O2] is the O2 concentration dissolved in the solution. As-
suming that [O2] is 2.2 mM in 2-propanol at 20 °C,44 the values
of kesc can be estimated. These kinetic parameters (kisc, kesc[O2]
and kesc) are summarized in Table 3. The values of kesc[O2] in
air-saturated solutions are nearly equal to those of kisc for all 2-
PCAs. This fact is very important for observing MFE in asym-
metric photolysis with CPL, as mentioned earlier. The values
of kesc are in the range of 109–1010 M−1 s−1, indicating that the
triplet biradical undergoes a diffusion-controlled reaction with
O2.
O2 concentration.
3. Biradical Lifetimes: In order to confirm the reaction
between the triplet biradical and O2, we measured transient ab-
sorption signals of the benzyl radical moiety (320 nm) on trip-
let acyl–benzyl biradicals generated upon 308 nm laser flash
photolysis. Figure 5 shows transient absorption decay curves
of 2-PCH in degassed, air- and O2-saturated solutions. Obvi-
ously the transient signal of decay component of the biradical
is quenched in the presence of molecular oxygen. The plateau
signal appeared for the air- and O2-satureated solutions are
presumably due to permanent photoproducts because its signal
remains on a time scale longer than millisecond. Similar re-
sults are obtained for 2-PCHep and 2-PCO.
These transient absorption singals were well-fitted with
Abs(t) = A + B exp (− t /τ), where Abs(t) is the transient ab-
sorption intensity, A and B are time-independent constants and
τ is the triplet biradical lifetime. The obtained lifetimes for de-
gassed (τdeg) and air-saturated (τair) solutions are listed in Table
3. The τdeg values for all 2-PCAs are in agreement with those
The lifetimes of the triplet biradicals at a field strength of
0.96 T (τMG) are listed in Table 3. It can be seen that there is
no differences between the values of τdeg (in the Earth magnet-

S. Kohtani et al.
Bull. Chem. Soc. Jpn., 75, No. 6 (2002) 1229
Table 2. Retention Time on Gas-Chromatography and Mass to Charge Ratio (m/z) of Molecular (M+) and
Fragment Ions on Mass Spectra of Photoproducts and 2-PCAs
2-PCAs
Peak Retention time m/z of molecular (M+)
2-PCAs and photoproducts
/ min(ꢀ)s(ꢁ)
and fragment ions
86 (M+)
a
4ꢀ 37ꢁ
3-butenoic acid
(CH2wCHCH2CO2H)
acetophenone
b
c
9ꢀ 27ꢁ
28ꢀ 45ꢁ
120 (M+), 105 (M+ − CH3)
2-PCH
174 (M+), 130 (M+ − CH3CHO) 6-phenyl-5-hexenal
(PhCHwCH(CH2)3CHO)
R
S
30ꢀ 14ꢁ
30ꢀ 41ꢁ
174 (M+)
174 (M+)
2-PCH, R enantiomer
2-PCH, S enantiomer
a
6ꢀ 17ꢁ
100 (M+)
4-pentenoic acid
(CH2wCH(CH2)2CO2H)
acetophenone
7-phenyl-6-hexenal
(PhCHwCH(CH2)4CHO)
2-PCHep
b
c
9ꢀ 27ꢁ
32ꢀ 44ꢁ
120 (M+), 105 (M+ − CH3)
188 (M+), 170 (M+ − H2O),
144 (M+ − CH3CHO)
188 (M+)
2-PCHep
d
d'
33ꢀ 20ꢁ
33ꢀ 40ꢁ
188 (M+)
2-PCHep
a
8ꢀ 19ꢁ
114 (M+),
5-hexenoic acid
(CH2wCH(CH2)3CO2H)
acetophenone
2-PCO
2-PCO
b
d
d'
c
9ꢀ 03ꢁ†
35ꢀ 54ꢁ
36ꢀ 12ꢁ
36ꢀ 30ꢁ
120 (M+), 105 (M+ − CH3)
202 (M+)
2-PCO
202 (M+)
202 (M+), 184 (M+ − H2O),
156 (M+ − CH3CHO)
7-phenyl-6-hexenal
(PhCHwCH(CH2)5CHO)
† The slight difference of retention time of acetophenone between 2-PCO and 2-PCH (2-PCHep) is
due to shortened length of the capillary column.
Table 3. Lifetimes of Triplet Acyl–Benzyl Biradicals and Other Kinetic Parameters
τair/ns
τ
deg/ns
τMFa)/ns
kiscb)/107 s−1 kesc[O2]c)/107 s−1 kesc/109M−1 s−1
2-PCH
2-PCHep
2-PCO
35
49
37
2
2
4
65
85
92
3
3
4
67
86
90
2
3
2
1.5
1.2
1.1
1.3
0.8
1.6
6.2
3.8
7.3
a) Magnetic field strength is 0.96 T.
b) kisc = 1 / τdeg
c) kesc[O2] = 1/τair − kisc, where [O2] is assumed to be 2.2 mM in 2-propanol at 20°C.44
.
ic field) and τMG within the experimental error. Although we
examined the triplet biradical lifetime in a higher field region
(1–8 T) for 2-PCH and 2-PCHep, the lifetimes remain un-
changed. This result clearly indicates that the ISC rates of
these biradicals are not affected by a static magnetic field be-
low 8 T.
4. Time Evolution of Enantiomer Concentrations: Ac-
cording to the results of our kinetic simulation, large recombi-
nation yield of the singlet biradical to each enantiomer is found
to be one of the necessary conditions to obtain large MFE.
This recombination reaction has already been demonstrated by
Doubleday’s CIDNP study.26 In addition, the rate constants for
recombination were estimated theoretically by de Kanter and
Kaptein45 based on Doubleday’s CIDNP data.26 In our ap-
proach, the recombination yield was evaluated by the follow-
ing procedures: (A) The time evolution of the enantiomer con-
centration of 2-PCAs, starting with the one optically pure
enantiomer, was monitored using chiral capillary GC. (B) The
observed time dependence was simulated by integrating differ-
ential Eqs. 7 and 8 based on the Runge–Kutta method, where
the recombination yield was treated as a fitting parameter.
Figure 6 shows the time evolution of the enantiomer con-
centrations of 2-PCAs with LPL irradiation. It is clear that as
the one enantiomer concentration decreases, the other enanti-
omer concentration increases to reach the maximum value, and
then decrease. Finally, both enantiomer concentrations coin-
cide asymptotically. This result clearly indicates that the race-
mization of 2-PCAs occurs via photoproduction and recombi-
nation of the biradicals.
The solid curves in Fig. 6 are fitting results, in which ky (=
kesc[O2]), kisc, εR, and εS in the Eqs. 7 and 8 are fixed to the val-
ues obtained from our absorption, CD, and transient absorption
measurements, as listed in Tables 1 and 3, whereas I0 η and φ
(recombination yield) are treated as variable parameters. From
the fitting procedures, sum of the recombination yield to each
enantiomer (2φ) were determined as listed in Table 4. 2φ was

1230 Bull. Chem. Soc. Jpn., 75, No. 6 (2002)
Asymmetric Photolysis of 2-PCAs with CPL
estimated to be more than 0.8 for all 2-PCAs, and thus ob-
tained large values. These values agree approximately with
those obtained from simulations for the CINDP data26 by de
Kanter and Kaptein,45 as listed in Table 4.
5. Asymmetric Photolysis in Air Saturated Solution:
After racemic 2-PCAs were photolyzed with CPL in both the
presence and absence of a magnetic field, the photoinduced ee
were determined by a chiral capillary GC. The values of ee af-
ter ca. 90% sample decomposition are summarized in Table 5.
For 2-PCH, R enantiomer enrichment is achieved to about 3%
ee by right-CPL irradiation and S enantiomer vice versa. This
value is consistent with our prediction based on a kinetic simu-
lation, as discussed later. 2-PCH has the largest ee values be-
cause of the largest g factor. However, within the experimental
error, there is no difference between the ee values in both the
presence and absence of the magnetic field. For 2-PCO, al-
though the ee values reach ca. 1–2% ee, no MFE on ee is ob-
served. The unbalanced ee between left- and right-CPL for 2-
PCHep is probably due to a peak overlap of the faster effluent
enantiomer with a small amount of a photoproduct because ca.
1% ee was observed even upon LPL irradiation. Taking into
account this value, the corrected ee values for 2-PCHep are ca.
1–2% ee for both right- and left-CPL, though the MFE was not
obtained.
Discussion
Our kinetic simulation indicates that four reaction condi-
tions are necessary for observing the MFE on asymmetric pho-
tolysis with CPL involving radical pair or biradical intermedi-
ates. These conditions are summarized as follows: (1) the
presence of an escape process is indispensable, and the rate
should be nearly equal to that of the intersystem crossing of
radical pair or biradicals; (2) the recombination yields from
singlet radical pair or biradical to each starting enantiomer
must be sufficiently large; and (3) the rate of intersystem cross-
ing of radical pairs or biradicals should be significantly
changed upon applying magnetic fields. In addition, (4) a
large Kuhn anisotropy g factor at the irradiation wavelength is
essential. These conditions are useful for exploring potential
reaction systems.
Our kinetic simulation also covers two types of asymmetric
photoreactions with CPL. One is an asymmetric photodestruc-
tion and the other is a photoderacemization.4–6,8 In the case of
kisc = 0 s−1 or φ = 0 but ky ≠ 0 s−1 in the Eqs. 7 and 8, the situ-
ation is reduced to the former type. We have confirmed that
the variations of ee as a function of the sample decomposition
simulated here precisely agree with predictions from the equa-
tion for asymmetric photodestruction with CPL, derived by
Kagan et al.6 and Nakamura et al.8 On the other hand, in the
limiting case where there are no photochemical side reactions,
i.e. ky = 0 s−1 and φ = 0.5, only the latter type of reaction
takes place. The ee change as a function of time is expected to
approach an asymptotic value predicted by the equation eepss
= g/2 in a photostationary-state.4,5
Fig. 6. Time evolution of 2-PCAs enantiomer concentration
started from optically pure enantiomer upon pulsed linear-
ly polarized light irradiation (308 nm, 2 mJ/pulse, fwhm
ca. 15 ns): (a) 2-PCH (ꢃ: R enantiomer, ꢅ: S enantiomer),
(b) 2-PCHep and (c) 2-PCO. Curve fittings were made by
numerical integration of equation (7) and (8) with parame-
ters as follows; (a) I0: 5×10−8 einsteins•cm−2 s−1, η: 0.30,
εR = εS: 36.6 M−1 cm−1, kisc: 1.5×107 s−1, ky : 1.3×107 s−1
,
φ: 0.4, (b) I0: 5×10−8 einsteins•cm−2 s−1, η: 0.18, εR = εS:
117.2 M−1 cm−1, kisc: 1.2×107 s−1 , ky : 0.8×107 s−1, φ:
0.4, (c) I0: 5×10−8 einsteins•cm−2 s−1, η: 0.24, εR = εS:
249.1 M−1 cm−1, kisc: 1.1×107 s−1 , ky :1.6×107 s−1, φ: 0.5.
Table 4. Comparison of the Recombination Yield of Acyl–
Benzyl Biradical (2φ) Obtained in Our Experiments and
Reference Values (Φ) Reported by de Kanter and
Kaptein45
2φ†
Φ‡
2-PCH
2-PCHep
2-PCO
0.8
0.8
1.0
0.72
0.80
0.91
The photochemistry of 2-PCAs in the presence of molecular
oxygen has been clarified, as shown in Scheme 3. This reac-
tion scheme is consistent with the kinetic model described in
Scheme 2. We have identified for the first time that the escape
products from small-ring 2-PCAs are acetophenones and alk-
†A sum of recombination yield to each enantiomer.
‡Calculated by the following equation; Φ = kcyc/(kcyc + kald
+ kket) where kcyc, kald and kket are rate constants for the for-
mation of parent 2-PCAs, alkenals and ketenes, respec-
tively, which are described in Table 1 in Ref. 45.

S. Kohtani et al.
Bull. Chem. Soc. Jpn., 75, No. 6 (2002) 1231
Table 5. Enantiomeric Excesses (ee) of 2-PCAs Induced by 308 nm CPL Irradiation in the Presence and Absence of a Magnetic
Field (0.76 T). Errors Correspond to Standard Deviations Derived from Five Independent Experiments. The Extent of Sample
Decompositions Indicated in Brackets Are in the Range of 84–92%
Left-CPL/% ee (decomp/%)
Right-CPL/% ee (decomp/%)
0 T 0.76 T
LPL/% ee (decomp/%)
2-PCAs
2-PCH†
2-PCHep‡
2-PCO‡
0 T
0.76 T
0 T
0.76 T
−2.6 0.2 (91)
+2.2 0.5 (89)
+1.7 0.8 (86)
−2.1 0.5 (92)
+2.9 0.4 (89)
+0.9 0.4 (91)
+3.4 0.2 (91) +3.3 0.4 (92)
−0.4 0.3 (90) −0.1 0.5 (88)
−1.6 0.5 (86) −1.0 0.7 (84)
+0.3 0.3 (92) +0.1 0.6 (92)
+1.5 0.5 (91) +1.1 0.2 (87)
+0.3 0.5 (87) +0.5 0.6 (88)
†The positive sign (+) indicates excess for R enantiomer of 2-PCH, ee = 100×([R] − [S])/([R] + [S]).
‡The positive sign (+) indicates excess for the faster effluent enantiomer on a chiral-capirally GC.
Scheme 3. Photolysis mechanism of 2-phenylcycloalkanones (2-PCAs, n = 6–8) in the presence of molecular oxygen in solution.
enoic acids. This photoreaction may consist of two steps of
photochemical reactions, as depicted in Scheme 4. First, the α
photocleavage of 2-PCAs (Norrish type Ⅰ) leads to the forma-
tion of triplet biradicals, and then results in the formation of
Compound (A), as described in the brackets by a reaction with
O2. Probably the bimolecular rate constant (kesc) obtained
from the transient absorption measurement indicates this reac-
tion rate constant. Second, Compound (A) can be excited by
the same UV light, and then subsequent intramolecular hydro-
gen abstraction (Norrish Type ꢂ) occurs to form acetophenone
and the corresponding alkenoic acids. This reaction mecha-
nism would also be supported from the facts that acetophenone
is also produced in the Norrish Type ꢂ reaction of alkyl phenyl
ketones, such as valerophenone and hexanophenone.46 The
subsequent Norrish Type ꢂ reaction may occur so efficiently
that Compound (A) is virtually unobservable in the present
GC/MS analysis.
The second noticeable point of our findings concerning the
photolysis of 2-PCAs is that the recombination yield from the
singlet biradical to the starting 2-PCAs can be estimated by fit-
ting procedures for the time evolution of the enantiomer con-
centration with LPL irradiation. Here, the recombination
yields are estimated to be more than 0.8 for all 2-PCAs. How-
ever, these values are regarded as being rough estimations.
Although CPL-induced asymmetric photolysis of 2-PCAs
(n = 6–8) was achieved to a few % ee the MFE on ee were not
observed. The absence of the MFE can be rationalized based
on four necessary conditions derived from our kinetic simula-
tion. First, the escape rates from the triplet biradical in air-sat-
urated solution are comparable to those of ISC for all 2-PCAs
as shown in Table 3. Thus the condition (1) is fulfilled. Fur-
ther, the recombination yields of the biradical are large enough
to satisfy the condition (2) as listed in Table 4. Second, the
anisotropy g factors of 2-PCAs listed in Table 1 are compara-
ble to the typical values for chiral carbonyl compounds as re-
ported previously.5 Additionally, more than 1% ee of 2-PCAs
after ca. 90% decomposition were obtained for all 2-PCAs (n
= 6–8) as expected. Consequently, conditions (1), (2), and (4)
are fulfilled in this 2-PCAs reaction system. However, the ISC
rates of acyl-benzyl biradicals (n = 6–8) were not affected by
the static magnetic field in our experiments (below 8 T).
Therefore, the condition (3) is apparently not satisfied. This is
the reason why we could not observe the MFE in this asym-
metric photolysis.
The ISC process of acyl-benzyl biradical has been investi-
gated in detail by Doubleday and Turro.26–28,30–33 They have
claimed that spin-orbit coupling (SOC), which has no magnet-
ic field dependence, plays a major role in the ISC process be-
cause of delocalization of the odd electron on the acyl oxygen.
In general, it is recognized that if radical pairs are separated by
1 nm or more, the singlet-triplet energy gap is decreased and
ISC is controlled by hyperfine coupling (HFC) induced nucle-
ar-spin interactions, which are dependent on the magnetic
fields and isotopes.11,17 However, since ISC of acyl–benzyl bi-
radicals is dominated by the field-independent SOC, MFE on
the ISC rate are diminished even for acyl–benzyl biradicals

1232 Bull. Chem. Soc. Jpn., 75, No. 6 (2002)
Asymmetric Photolysis of 2-PCAs with CPL
Fig. 7. Simulation results of the ee as a function of time. Pa-
rameters used here are as follows; Line A: I0: 5×10−8 ein-
steins•cm−2 s−1, η: 0.30, εR: 37.15 M−1 cm−1, εS: 36.05 M−1
cm−1, kisc: 1.5×107 s−1 , ky : 1.3×107 s−1, φ: 0.4. Initial
concentration for both R and S enantiomers is the same
value of 2 mM. Line B: kisc: 1.5×106 s−1, other parameters
are identical to those used for Line A.
two-fold increase in the simulated ee is predicted if kisc could
be decreased by one order of magnetude by applying a mag-
netic field. This difference may be distinguished in our chiral
capillary GC experiments (see Table 5). It is not unusual that
the ISC rate of biradical changes upon applying magnetic
fields of more than one order of magnitude in the case where it
is dominated by the HFC mechanism.12–15,17
In order to increase the MFE on asymmetric photolysis with
CPL for a series of cycloalkanones, we can impose a restric-
tion on the relative orientation of the acyl and alkyl radicals.
Some of biradical intermediates derived from bicyclic ketones
may not undergo a conformational change into an orientation
favorable for SOC-induced recombination in the triplet state,
as was pointed out by de Kanter and Kaptein.45 The HFC-in-
duced ISC may become dominant for the biradical with specif-
ic stereochemistry, which suppresses ISC due to the SOC
mechanism. Therefore, we still recognize the importance of
the photochemical α-cleavage of chiral cyclic ketones (Norrish
type Ⅰ), even if terminal acyl radicals are generated.
Scheme 4. An escape reaction mechanim of the triplet acyl–
benzyl biradicals with molecular oxygen.
with a long end-to-end distance (n = 10–12). According to
studies by Doubleday and Turro,31,32 the SOC contribution to
the ISC is estimated to be 76% for n = 12 and 88% for n = 11.
Moreover, the SOC interaction will exponentially increase
with decreasing radical pair distance so that contribution of
magnetic-field-dependent HFC to the ISC becomes much
smaller in short-chain biradicals. Consequently, ISC would be
completely dominated by SOC in the acyl–benzyl biradicals
derived from small rings (n = 6–8), and thus the MFE have not
been observed.
CPL-induced ee changes of 2-PCH were simulated based on
the assumption that kisc is decreased by one order of magnitude
upon applying a magnetic field. Figure 7 shows two simulated
curves. Line A indicates the actual ee changes based on our
experimental values, where kisc is set to be the actural value of
1.5×107 s−1. In this case, the values of ee and the sample de-
composition at 60 min reach to about 3% ee and 90%, respec-
tively. These values are in good agreement with the experi-
mental values. (see Table 5). Line B is for the case where kisc
is decreased by one order of magnitude (1.5×106 s−1), but all
other parameters are fixed. It is apparent that an approximate
The authors are indebted to Mr. Jun Kojima (The University
of Tokyo) who gave support in programming for the simula-
tion. This work was supported by Sasagawa Scientific Re-
search Grant from The Japan Science Society (No. 13-130).
References
1
2
W. A. Bonner, Orig. Life Evol. Biosphere, 21, 59 (1991).
M. Avalos, R. Babiano, P. Cintas, J. L. Jimenez, J. C.
Palacios, and L. D. Barron, Chem. Rev., 98, 2391 (1998).
B. L. Feringa, and R. A. van Delden, Angew. Chem. Int.
Ed., 38, 3418 (1999).
3
4
5
H. Rau, Chem. Rev., 83, 535 (1983).
Y. Inoue, Chem. Rev., 92 741 (1992), and references cited
therein.
6
G. Balavoine, A. Moradpour, and H. B. Kagan, J. Am.
Chem. Soc., 96, 5152 (1974).
7
H. Nishino, A. Nakamura, and Y. Inoue, J. Chem. Soc.,
Perkin Trans. 2, 2001, 1693.

S. Kohtani et al.
Bull. Chem. Soc. Jpn., 75, No. 6 (2002) 1233
8
A. Nakamura, H. Nishino, and Y. Inoue, J. Chem. Soc.,
28 K. C. Hwang, N. J. Turro, and C. Doubleday Jr., J. Am.
Chem. Soc., 113, 2850 (1991).
Perkin Trans. 2, 2001, 1701.
9
H. Nishino, A. Nakamura, H. Shitomi, H. Onuki, and Y.
29 D. Weir, and J. C. Scaiano, Chem. Phys. Lett., 118. 526
(1985).
30 M. B. Zimmt, C. Doubleday Jr., I. R. Gould, and N. J.
Turro, J. Am. Chem. Soc., 107, 6724 (1985).
31 M. B. Zimmt, C. Doubleday Jr., and N. J. Turro, J. Am.
Chem. Soc., 107, 6726 (1985).
32 C. Doubleday Jr., N. J. Turro, and J. F. Wang, Acc. Chem.
Res., 22, 199 (1989).
33 J. F. Wang, V. P. Rao, C. Doubleday Jr., and N. J. Turro, J.
Phys. Chem., 94, 1144 (1990).
34 A. A. Baum, Tetrahedron Lett., 1972, 1817.
35 P. J. Wagner and T. J. Stratton, Tetrahedron, 37, 3317
(1981)
36 X. Lei, C. Doubleday Jr., and N. J. Turro, Tetrahedron
Lett., 27, 467 (1986).
Inoue, J. Chem. Soc., Perkin Trans. 2, 2001, 1706.
10 G. L. J. A. Rikken and E. Raupach, Nature, 405, 932
(2000).
11 U. E. Steiner and T. Ulrich, Chem. Rev., 89, 51 (1989).
12 R. Nakagaki, O. Takahira, and K. Hiruta, Chem. Phys.
Lett., 233, 41 (1995).
13 R. Nakagaki, M. Yamaoka, O. Takahira, K. Hiruta, Y.
Fujiwara, and Y. Tanimoto, J. Phys. Chem. A., 101, 556 (1997).
14 Y. Fujiwara, T. Aoki, T. Haino, Y. Fukazawa, Y. Tanimoto,
R. Nakagaki, O. Takahira, and M. Okazaki, J. Phys. Chem. A.,
101, 6842 (1997).
15 R. Nakagaki, M. Yamaoka, and K. Mutai, Bull. Chem. Soc.
Jpn., 72, 347 (1999).
16 I. Suzuki, R. Tanaka, A. Yamaguchi, S. Maki, H. Misawa,
K. Tokumaru, R. Nakagaki, and H. Sakuragi, Bull. Chem. Soc.
Jpn., 72, 103 (1999).
37 X. Lei, C. Doubleday Jr., M. B. Zimmt, and N. J. Turro, J.
Am. Chem. Soc., 108, 2444 (1986).
17 H. Hayashi, Y. Sakaguchi, and M. Wakasa, Bull. Chem.
Soc. Jpn., 74, 773 (2001).
38 M. A. Miranda, E. F. Sanchis, and J. P. Prieto, J. Org.
Chem., 64, 3802 (1999).
18 R. Nakagaki, Y. Tsujimoto, and K. Mutai, Chem. Phys.
Lett., 244, 388 (1995).
19 R. Nakagaki and K. Mutai, Bull. Chem. Soc. Jpn., 69, 261
(1996).
20 Y. Konishi, P. Maruthamuthu, M. Okazaki, Y. Arakawa,
and R. Nakagaki, Bull. Chem. Soc. Jpn., 72, 27 (1999).
21 G. Lem and N. J. Turro, Chem. Commun., 2000, 293.
22 R. A. Hegstrom and D. K. Kondepudi, Chem. Phys. Lett.,
253, 322 (1996).
39 R. Hayano, and T. Takahashi, “Computer in Physics,”
Kyoritsu Shuppan (1992). Simulation source program was written
in the C programming language.
40 J. H. Ryan, and P. J. Stang, Tetrahedron Lett., 38, 5061
(1997).
41 G. Berti, B. Macchia, F. Macchia, and L. Monti, J. Chem.
Soc. C, 1971, 3371.
42 K. Ishihara, M. Kaneeda, and H. Yamamoto, J. Am. Chem.
Soc., 116, 11179 (1994).
23 K. Soai, T. Shibata, and I. Sato, Acc. Chem. Res., 33, 382
(2000).
24 M. Avalos, R. Babiano, and J. C. Palacios, Chem. Com-
mun., 2000, 887.
25 D. A. Lightner, in “Circular Dichroism,” 2nd ed., ed by N.
Berova, K. Nakanishi, and R. W. Woody, Wiley-VCH, New York
(2000), Chap. 10.
43 J. N.Murrel, in “The Theory of the Electronic Spectra of
Organic Molecules,” Chapman and Hall (1963), Chap. 8.
44 S. L. Murov, I. Carmichael, and G. L. Hug, in “Handbook
of Photochemistry,” 2nd. ed., Revised and Expanded, Marcel
Dekker (1993), Section 12–3.
45 F. J. J. de Kanter and R. Kaptein, J. Am. Chem. Soc., 104,
4759 (1982).
26 C. Doubleday Jr., Chem. Phys. Lett., 64. 67 (1979).
27 N. J. Turro, K. C. Hwang, V. P. Rao, and C. Doubleday Jr.,
J. Phys. Chem., 95, 1872 (1991).
46 N. J. Turro, in “Modern Molecular Photochemistry,”
Benjamin/Cummings, Menlo Park (1978), Chap. 10.