Solid-state specific and unidirectional photoisomerization of 3-substituted propyl to 1-substituted propyl cobaloxime complexes via 2-substituted propyl complexes

Article abstract of DOI:10.1246/cl.2001.1190

Solid state-specific (γ→β→α)photoisomerization of 3-cyanopropyl and 3-(methoxycarbonyl)propyl cobaloxime complexes were found to occur on visible-light irradiation. There exist considerable differences in the reaction rates, which have no relation to the

Full text of DOI:10.1246/cl.2001.1190

1
190
Chemistry Letters 2001
Solid-State Specific and Unidirectional Photoisomerization of 3-Substituted Propyl to
-Substituted Propyl Cobaloxime Complexes via 2-Substituted Propyl Complexes
1
Yoshiaki Ohgo,* Futoshi Kurashima, Yoshito Hiraga, Kanako Ishida, Naoko Takatsu, Yoshifusa Arai, and Seiji Takeuchi
Niigata College of Pharmacy, Kamishin’ei-cho, Niigata 950-2081
(Received August 3, 2001; CL-010745)
Solid state-specific (γ→β→α)photoisomerization of 3-
cyanopropyl and 3-(methoxycarbonyl)propyl cobaloxime com-
plexes were found to occur on visible-light irradiation. There
exist considerable differences in the reaction rates, which have
no relation to the basicity of the axial ligand. In the former
series of complexes, the rate constant (k ) for the second step
2
(
β→α) is much faster than that (k ) for the first step. However,
1
in the latter series of complexes, the rate constant (k ) for the
2
second step tends to decrease extremely, and thus, k become
2
smaller than k1.
Solid-state (solvent free) organic reactions have attracted
much attention not only from purely chemical interest but also
from the view point of so-called “green chemistry”.
ceeds rapidly and 93% of the starting material γ-1a was con-
verted to isomerized products within 2 hours. The time courses
of the ratios of complexes γ-, β-, and α-1a are shown in Figure
1 which indicates the reaction to be typically consecutive one.
From the first order rate plot of the ratio of the starting material
γ-1a of the early stage, the rate constant of the first step (γ to β
We previously reported that 2-substituted ethyl cobaloxime
complexes isomerized to 1-substituted ethyl complexes on visi-
1
–3
ble light irradiation in the solid state, and the asymmetric
2
,3
–3 –1
(
β→α) photoisomerization took place when chiral crystals
isomerization)(k1 = 2.0 × 10 s ) was obtained. The rate con-
7
were used. The above mentioned reactions do not occur in
solution, and we designate the reactions (that occur only in the
solid-state) as “solid-state specific” reactions.
stant of the second step (β to α), k2, was calculated to be 5.7 ×
–
3
–1
10
s
by using k1, initial concentration of γ-complex, and
concentration of β-complex at time t in the early stage.
Although solid-state reactions have been rapidly increasing
in recent years, reactions which occur only in the solid-state
are still extremely rare. It is necessary and important at this
stage to accumulate such novel examples and to reveal factors
controlling the reaction.
Other 3-cyanopropyl cobaloxime complexes (γ-1b – γ-1e)
were also shown to isomerize similarly. In the cases of these
complexes, the axial base was displaced by methyldiphenyl-
phosphine after the reaction, and the resulting sample was ana-
lyzed by HPLC to obtain k1 and k2 for each reaction. These
results are summarized in Table 1 (γ-1a – γ-1e). The reverse
reaction does not occur in the solid-state, and also the
(γ→β→α) isomerization does not occur in solution state.
Several 3-(methoxycarbonyl)propyl cobaloxime complexes
(γ-2c, γ-2e and γ-2n), and dimethylphenylphosphine-coordi-
nated 2-(methoxycarbonyl)propyl and 1-(methoxycarbonyl)-
4
In order to develop a novel and analogous system, we
examined whether photoisomerization of 3-substituted propyl
cobaloxime complexes could occur in the solid-state, and found
the titled reaction which carried quite interesting and unique
5
features. Very recently, Ohashi and his coworkers reported
that 4-cyanobutyl(pyridine)cobaloxime complex isomerized to
-cyanobutyl complex but occurrence of 1-cyanobutyl isomer is
8
2
propyl complexes (β-2n and α -2n) were also prepared.
not known. The report prompted us to publish our results
urgently.
Several 3-cyanopropylcobaloxime complexes (γ-1a – γ-1e)
coordinated with various type of axial ligand [methyl-
diphenylphosphine (a), 3-aminopyridine (b), 4-cyanopyridine
Photoreactions were carried out as mentioned above. The reac-
tion products were analyzed by HPLC using Dicel OD-H (sol-
vent system: hexane/ethanol (95/5)). From the time courses,
rate constants k1 and k2 of each reaction were obtained in a sim-
ilar manner as in the cyanopropyl series. The results are sum-
marized in Table 1 (γ-2c, γ-2e, and γ-2n).
(
c), propylamine (d), and aniline (e)], β- and α-cyanopropyl
isomers of γ-1a (β-1a and α-1a) were prepared and character-
ized by IR and H NMR spectra.
From Table 1, two quite interesting and unique features
can be seen: (1) The results show considerable differences in
the reaction rates, which have no relation to the basicity of the
axial ligand. (2) In the cyano-substituted series of complexes,
the reaction rate for the second step (β→α) is much faster than
that for the first step (γ→β), on the other hand, in the methoxy-
carbonyl-substituted series of complexes, the reaction rate for
the second step is slower than that for the first step.
1
6
A powdered sample of methyldiphenylphosphine-coordi-
nated 3-cyanopropyl cobaloxime complex (γ-1a) was sus-
pended in nujol and irradiated with a solar simulator (flux den-
2
sity: 100 mW/cm ) in the solid state for a set period of time at
room temperature, and the sample was analyzed by HPLC
using Dicel Chiralcel OD-H (solvent system: hexane/ethanol
(
93/7)) to obtain the ratio of starting material (γ-substituted
Since electronic effect of the axial ligand is shown to be
negligible on the rate of the Co–C bond photohomolysis of
alkyl cobaloximes in solution state in which intermolecular
complex), 2-cyanopropyl (β-substituted complex), and 1-
cyanopropyl (α-substituted complex). The isomerization pro-
Copyright © 2001 The Chemical Society of Japan

Chemistry Letters 2001
1191
References and Notes
1
2
3
Y. Ohgo and S. Takeuchi, J. Chem. Soc., Chem. Commun.,
985, 21.
Y. Ohgo, Y. Arai, M. Hagiwara, S. Takeuchi, H. Kogo, A.
Sekine, H. Uekusa, and Y. Ohashi, Chem. Lett., 1994, 715.
Y. Ohgo, M. Hagiwara, M. Shida, Y. Arai, and S. Takeuchi,
Mol. Cryst. Liq. Cryst., 277, 241(1996).
1
4
5
T. Tanaka and F. Toda, Chem. Rev., 100, 1025(2000).
C. Vithana, H. Uekusa, and Y. Ohashi, Bull. Chem. Soc. Jpn.,
7
4, 287–292 (2001).
H NMR (400 MHz, CDCl ) data for key compounds of
1
6
3
cyanopropyl series γ-1a; δ 7.37 (m, 10H, aromatic), 2.19 (t, 2H,
J 8.9 Hz, –CH –CN), 1.84 (d, 3H, J
= 8.1 Hz, P–CH ), 1.80
2
H–P
3
(
d, 12H, JH–P = 3.4 Hz, CH of dmgH(dmgH: dimethylglyoxi-
3
mato mono anion)), 1.64 (m, 2H, Co–CH –), 1.28 (m, 2H,
2
CoCH –CH –), β-1a; δ 7.38 (m, 10H, aromatic), 2.38 (ddd, 1H,
2
2
J = 4.4 Hz, J = 7.1 Hz, and J = 17.1 Hz, –CHH–CN), 2.04 (ddd,
H, J = 2.9 Hz, J = 9.0 Hz, and J = 17.1 Hz, –CHH–CN), 1.84
d, 3H, JH–P = 8.1 Hz, P–CH ), 1.81 (d, 6H, J = 3.7 Hz, CH3
1
(
3
H–P
of dmgH), 1.80 (d, 6H, JH–P = 3.7 Hz, CH of dmgH), 1.40 (m,
3
1
H, Co–CH–), 0.77 (dd, 3H, JH–P = 7.3 Hz and J = 7.3 Hz,
9
CoCH–CH ), and α-1a; δ 7.43 (m, 10H, aromatic), 2.20 (m, 1H,
Co–CH–), 1.89 (d, 3H, JH–P = 9.3 Hz, P–CH ), 1.88 (d, 6H, J
=
interactions are the same, the rate in the solid-state isomeriza-
tion should be controlled both by stability of alkyl radical due
3
3
H–P
3.2 Hz, CH of dmgH), 1.84 (d, 6H, J
= 3.2 Hz, CH of
1
0
3
H–P 3
to capto-dative effect (intramolecular electronic factor) and by
intermolecular (lattice-controlled) factor around the reactive
group.
dmgH), 1.10 (m, 1H, CoCH–CHH–CH ), 0.97 (t, 3H, J =
3
7
.0Hz, CoCHCH –CH ), 0.89 (m, 1H, CoCH–CHH–CH ).
2
3
3
7
8
k was calculated by applying the rate expression for consecu-
2
–k1t
–k2t
Here, the stabilities of the cyano- and methoxycarbonyl-
tive reaction, [β-1]t = [γ-1]0 k1(e – e )/(k2 – k1)where [β-1]t,
[γ-1] , k , and k are concentration of β-1 at time t, initial con-
centration of γ-1, rate constant of the first step, and that of the
1
1
substituted radicals are not so different in solution, so the
reaction rates, if compared in each step, will be roughly identi-
cal and also k is expected to be greater than k based on the sta-
bility of the radicals. Although there are considerable differ-
ences in intermolecular interactions (among 1a–1e) both at first
step and at second step in the cases of cyanopropyl complexes
as seen from considerable variation both in k and in k ), k is
greater than k in every case when k and k are compared in the
reaction of the same substrate. This implies that the difference
in intermolecular interactions) between first and second steps
is not so large that k remains to be greater than k . On the
0
1
2
second step, respectively.
2
1
1
H NMR (400 MHz, CDCl ) data for key compounds of
3
(methoxycarbonyl)propyl series γ-2n; δ 7.35 (m, 3H, aromatic),
7.10 (m, 2H, aromatic), 3.57 (s, 3H, –OCH ), 2.18 (t, 2H, J =
7.4 Hz, –CH2–COOCH3), 1.93 (d, 12H, JH–P = 3.4 Hz, CH3 of
dmgH), 1.58 (m, 2H, Co–CH –), 1.39 (d, 6H, J = 9.0 Hz,
P–(CH ) ), 1.25 (m, 2H, CoCH –CH –), β-2n; δ 7.34 (m, 3H,
3
(
2
H–P
1
2
2
3
2
2
2
1
1
2
aromatic), 7.09 (m, 2H, aromatic), 3.56 (s, 3H, –OCH ), 2.63
3
(
m, 1H, CoCH–CHH–), 1.94 (d, 12H, JH–P = 3.6 Hz, CH of
3
(
dmgH), 1.64 (m, 2H, Co–CH–CHH–), 1.37 (d, 6H, JH–P = 9.3
Hz, P–(CH ) ), 0.61 (dd, 3H, J = 6.6 Hz and J = 6.6 Hz,
2
1
3 2
H–P
other hand, in the cases of (methoxycarbonyl)propyl complex-
CoCH–CH3), and α-2n; δ 7.36 (m, 3H, aromatic), 7.11 (m, 2H,
aromatic), 3.44 (s, 3H, –OCH ), 1.99 (d, 6H, J
3
H–P
= 3.2 Hz, CH3
es, k became smaller than k in every substrate examined. This
2
1
of dmgH), 1.97 (d, 6H, JH–P = 3.2 Hz, CH of dmgH), 1.87 (m,
3
seems to imply that intermolecular interaction of the second
step become extremely greater than that of the first step.
However, extensive X-ray structural analyses are required
for the clear elucidation which will be dealt with in a full paper.
1
3
H, Co–CH–), 1.38 (d, 3H, JH–P = 10.0 Hz, P–CH ), 1.37 (d,
3
H, JH–P = 10.0 Hz, P–CH ), 1.20 (m, 2H, CoCH–CH –), 0.73
3
2
(t, 3H, J = 7.2 Hz, CoCHCH –CH ).
2 3
9
1
1
Y. Ohgo, H. Wada, C. Ohtera, M. Ikarashi, S. Baba, and S.
Takeuchi, Bull. Chem. Soc. Jpn., 64, 2656 (1991).
H. G. Viehe, Z. Janousek, R. Merenyi, and L. Stella, Acc.
Chem. Res., 18, 148 (1985), and references cited therein.
B. K. Bandlish, A. W. Garner, M. L. Hodges, and J. W.
Timberlake, J. Am.Chem. Soc., 97, 5856 (1975).
0
1
This work was partly supported by Grant-in-Aids for
Scientific Research from the Ministry of Education, Science,
Sports and Culture, Japan.