Enantioselective epoxidation of trans-disubstituted alkenes by D2-symmetric chiral dioxoruthenium(VI) porphyrins

Article abstract of DOI:10.1039/a808776h

A series of D₂-symmetric chiral trans-dioxoruthenium(VI) porphyrins can effect enantioselective epoxidation of trans-β-methylstyrene in up to 70% ee, and 76% ee is attained for the oxidation of cinnamyl chloride; the facial selection for the tr

Full text of DOI:10.1039/a808776h

Enantioselective epoxidation of trans-disubstituted alkenes by D -symmetric
2
chiral dioxoruthenium(vi) porphyrins
Rui Zhang, Wing-Yiu Yu, Tat-Shing Lai and Chi-Ming Che*
Department of Chemistry, The University of Hong Kong, Pokfulam Road, Hong Kong. E-mail: cmche@hkucc.hku.hk
Received (in Cambridge, UK) 10th November 1998, Accepted 4th December 1998
A series of D
2
-symmetric chiral trans-dioxoruthenium(vi)
and 55% respectively. A paramagnetic bis(pyrazolato)ruth-
IV 1–3
porphyrins can effect enantioselective epoxidation of trans-
b-methylstyrene in up to 70% ee, and 76% ee is attained for
the oxidation of cinnamyl chloride; the facial selection for
the trans-alkenes epoxidation is explained by a ‘head-on
approach’ model.
enium(iv), [Ru (L *)(pz)
2
B
] 2 (meff = 2.9 m ), was isolated by
column chromatography at the end of the epoxidation reac-
tions.4 It is noteworthy that when styrene reacted with 1a
without Hpz; styrene oxide with only 40% ee was obtained and
II
1
[Ru (L *)(CO)] was isolated. To account for the ee discrep-
ancy, we propose that a ruthenium(ii) species produced by
further reduction of the putative Ru(iv) intermediate would
racemize the chiral styrene oxide by an epoxide ring opening
The design of new metal catalysts for highly enantioselective
epoxidation of unfunctionalized trans-disubstituted alkenes
remains a challenge in the field of asymmetric oxidation. A
1
5
pathway previously suggested by Groves et al. (Scheme 1).
clue to this problem would be to prepare reactive, and yet
isolable, chiral metal–oxo complexes which can act upon trans-
alkenes to give epoxides in better enantioselectivities than the
cis-counterparts.2 If such complexes can be obtained, the
structure–enantioselectivity relationship established by study-
ing their stoichiometric alkene epoxidations could assist future
design for better metal catalysts. Here, we show that a series of
More importantly, 1a reacted with trans-b-methylstyrene in
degassed benzene to give the trans-epoxide in 67% ee at room
temperature, and up to 70% ee was attained when performing
the reaction at 0 °C (entry 3). Notably, the oxidation of
cinnamyl chloride furnished the corresponding epoxide in 76%
ee and 70% yield (entry 4). Indeed, few reported metallo-
porphyrin-catalyzed asymmetric epoxidation systems are
trans-dioxoruthenium(vi) complexes with D
phyrins bifacially encumbered by four chiral threitol units (Fig.
) can react favorably with trans-alkenes in exceptional %ee.
2
symmetric por-
Table
1 Stoichiometric epoxidation of some aromatic alkenes by
1
VI
1
2
[Ru (L *)O ] (1a)
Also, our findings do not reconcile with the ‘side-on approach’
model proposed for the oxo-metalloporphyrin-mediated alkene
oxidations.
2 2
The D -symmetric porphyrins H
L1–3* and the [RuII-
L1–3*(CO)(EtOH)] precursors were prepared by the literature
3
methods. Treating the ruthenium(ii) carbonyl complexes with
m-chloroperoxybenzoic acid in dichloromethane for 5 min gave
VI 1–3
[
2
Ru (L *)O ] 1a–c, which were isolated as dark red–purple
crystalline solids (80% yield). Complexes 1a–c are diamagnetic
and stable in solid and in solution for hours at room temperature.
The oxidation state marker band at 1018 (1a), 1019 (b,c)
together with an intense asymmetric ONRuNO stretch at 818
2
1
(
1a), 821 (1b), 819 cm (1c) observed in their IR spectra are
consistent with a Ru(vi) formulation.
At room temperature, complex 1a in the presence of pyrazole
Hpz) oxidized styrene in a degassed benzene solution to give
(
styrene oxide in 64% yield and 62% ee (Table 1, entry 1). The
enantioselectivity was slightly improved (65% ee) when the
reaction was carried out at 0 °C, and the %ee is comparable to
the best reported value of 69% ee by employing the chiral Mn–
3
porphyrin catalysts. However, the other dioxoruthenium(vi)
derivatives bearing gem-diethyl (1b) and gem-cyclopentyl
groups (1c) at the threitol units afforded slightly lower ee of 60
3
Reaction conditions: to a degassed benzene solution (1 cm ) containing
alkene (1 mmol) and pyrazole (0.3 mmol) was added the dioxo-
ruthenium(vi) complex (0.015–0.03 mmol) under an argon atmosphere. The
reaction mixture was stirred at room temperature (or otherwise noted) for
1
2 h, and the aliquot was analyzed by gas chromatography for product
a
identification and quantification. Yields are based on the amount of the
oxidant used. b Enantiopurities (%ee) of the epoxides were determined by
GC equipped with a chiral capillary column (J & W Scientific cyclodex-B
or G-TA). Absolute configuration was determined by comparing with
c
authentic chiral samples. Benzaldehyde (27%) and phenylacetaldehyde
(
12%) were also detected. d Reaction was carried out without pyrazole.
Fig. 1
Chem. Commun., 1999, 409–410
409

Scheme 3
Scheme 1
atm O
7).
2
, 40 h) to give the trans-epoxide in 59% ee (turnover =
known to attain > 20% ee for the trans-b-methylstyrene
oxidation.3 We had previously reported that a chiral D
,6
4
-
We acknowledge support from The University of Hong Kong
and The Hong Kong Research Grants Council, ERB003 and
HKU7092/98P.
symmetric dioxoruthenium(vi) complex can effect epoxidation
of cis-b-methylstyrene in much higher enantioselectivity of
7
7
6% ee vs. 20% ee for the trans-isomer. In this work, when cis-
b-methylstyrene (entry 5) and 1,2-dihydronaphthalene (entry 6)
reacted with 1a, the cis-epoxides ( > 99% stereoretention) were
produced in only 40 and 20% ee, respectively. Oxidation of
trans- and cis-b-methylstyrene by complexes 1b and 1c also
resulted similar trans preference albeit in lower enantioselectiv-
ities, for instance when 1b was the oxidant, the trans- and cis-
epoxides with 50 and 28% ee resulted respectively.
Notes and references
†
Characterization data for the dioxoruthenium(vi) and bis(pyrazolato)-
VI
1
1
ruthenium(iv) porphyrin complexes: [Ru (L *)O
CDCl ): d 8.65 (d, J 4.7 Hz, 4H), 8.55 (d, J 4.7 Hz, 4H), 7.77 (t, J 6.5 Hz,
H), 7.22–7.38 (m, m-H overlapped with a solvent peak, 8H), 4.91 (d, J 10.4
2
1a: H NMR (300 MHz,
3
4
The asymmetric alkene epoxidations by 1a also exhibit
remarkable solvent dependence.8 The trans-b-methylstyrene
oxidation by 1a displays clean pseudo-first-order kinetics:
Hz, 4H), 4.62 (d, J 9.0 Hz, 4H), 4.43 (t, J 9.0 Hz, 4H), 4.22 (d, J 10.1 Hz,
4H), 3.76 (d, J 9.0 Hz, 4H), 2.60 (t, J 8.5 Hz, 4H), 0.77 (s, 12H), 20.78 (s,
3
21
21
2 2
12H). UV–VIS (CH Cl ) lmax/nm (log e/dm mol cm ): 442 (5.12), 536
+
+
+
(
3
4.07). FABMS m/z: 1379 (M , 18%), 1363 (M , 18%), 1363 (M 2 O,
-
d[Ru(vi)]/dt = kobs[Ru(vi)], where kobs = k
the second-order rate constant of the reaction. Upon changing
solvent from benzene to dichloromethane, the k value halves:
), 4.15 3 10 dm mol s21 (CH
Cl ) at
98 K. It is noted that the %ee of the trans-epoxide also
) vs. 32% ee
). By contrast, no solvent effect has been observed for
the analogous reactions of the D -chiral dioxoruthenium(vi)
porphyrin. Since the alkene oxidation is highly stereospecific
Table 1, entries 3 and 5), therefore, the %ee of the epoxide
2 2
[alkene] and k is
+
0%), 1347 (M 2 20, 100%).
VI 2 1
[Ru (L *)O
2 3
] 1b: H NMR (300 MHz, CDCl ): d 8.64 (d, J 4.7 Hz, 4H),
2
8
.53 (d, J 4.7 Hz, 4H), 7.74 (m, 4H), 7.37–7.30 (m, m-H overlapped with a
2
4
24
3
21
9
2
.04 3 10 (C
6
H
6
2
2
solvent peak, 8H), 4.98 (d, J 10.4 Hz, 4H), 4.60 (d, J 8.8 Hz, 4H), 4.44 (t,
J 8.9 Hz, 4H), 4.23 (d, J 10.4 Hz, 4H), 3.74 (d, J 8.8 Hz, 4H), 2.63 (t, J 8.7
Hz, 4H), 1.02 (m, 8H), 0.53 (t, J 7.2 Hz, 12H), 20.25 (m, 8H), 21.37 (t, J
decreases in a similar fashion: 67% ee (C
6 6
H
(CH Cl
2
2
7.3 Hz, 12H). UV–VIS (CH Cl ): l /nm (log e/dm
3
mol21 cm21) 443
2
2
max
+
+
4
(5.15), 534 (4.02). FABMS: m/z 1491 (M , 9%), 1475 (M 2 O, 20%), 1459
+
(M 2 2O, 100%).
VI
3
1
[
2 3
Ru (L *)O ] 1c: H NMR (300 MHz, CDCl ): d 8.67 (d, J 4.6 Hz, 4H),
(
8
.56 (d, J 4.6 Hz, 4H), 7.76 (m, 4H), 7.20–7.37 (m, m-H overlapped with a
solvent peak, 8H), 4.88 (d, J 9.3 Hz, 4H), 4.60 (d, J 9.4 Hz, 4H), 4.64 (t, J
.0 Hz, 4H), 4.24 (d, J 10.1 Hz, 4H), 3.77 (d, J 8.6 Hz, 4H), 2.57 (t, J 8.6
Hz, 4H), 0.83 (m, 12H), 0.60 (m, 12H), 0.32 (m, 4H), 21.16 (m, 4H). UV–
should be determined only by facial selectivity at the rate-
limiting association of the CNC bond with the RuNO group
9
(
Scheme 2).9 We suspect that polar solvent may aggregate
around the threitol units through dipole–dipole interaction
thereby affecting the facial approach of the alkene.
3
21
VIS (CH
FABMS m/z: 1483 (M , 8%), 1467 (M 2 O, 22%), 1451 (M 2 2O,
00%).
2 2
Cl ) lmax/nm (log e/dm mol cm ): 441 (5.03), 535(4.02).
+
+
+
1
IV
1
21
[
Ru (L *)(pz)
FABMS m/z: 1483 (M , 100%), 1415 (M 2 pz, 10%), 1347 (M 2 2 pz,
0%). Anal. Calc. for C78 19Ru·3H O: C, 60.89; H, 5.33; N, 7.28.
Found: C, 60.77; H, 5.31; N, 7.25%. UV–VIS (CHCl
) lmax/nm (loge/dm3
2
] 2a: IR(KBr): 1006 cm (oxidation state marker band).
+
+
+
2
76 8
H N O
2
3
2
1
21
mol cm ): 425 (5.09), 517 (4.04), 550 (sh).
Scheme 2
1
E. N. Jacobsen, in Comprehensive Organometallic Chemistry II, ed. G.
Wilkinson, F. G. A. Stone, E. W. Abel and L. S. Hegedus, Pergamon,
New York, 1995, vol. 12, ch. 11.1; For chiral ruthenium complexes
catalyzed asymmetric epoxidation of trans-alkenes, see: N. End and A.
Pfaltz, Chem. Commun., 1998, 589.
In literature, cis–trans stereoselectivity in the metallopor-
phyrin-catalyzed alkene epoxidations is generally explained by
the ‘side-on approach’.10 Yet, in this case, this model cannot
account for the observed trans preference. Here, we propose a
2 Z. Gross, S. Ini, M. Kapon and S. Cohen, Tetrahedron Lett., 1996, 37,
7325.
3 J. P. Collman, V. J. Lee, C. J. Kellen-Yuen, X. Zhang, J. A. Ibers and
J. I. Brauman, J. Am. Chem. Soc., 1995, 117, 692.
4
5
6
‘
head-on approach’ model in which the molecular plane of the
CNC bond lies perpendicular to the RuNO axis (Scheme 3).
Since the cis vs. trans stereoselectivity should arise from the
steric interaction between the incoming alkene and the por-
phyrin ligand, the ‘head-on approach’ of a trans-alkene
molecule to oxo–metalloporphyrins can also be feasible
according to previous studies.11
C.-J. Liu, W.-Y. Yu, S.-M. Peng, T. C.-W. Mak and C.-M. Che,
J. Chem. Soc., Dalton Trans., 1998, 1805.
J. T. Groves, K. T. Ahn and R. Quinn, J. Am. Chem. Soc., 1988, 110,
4
217.
The best reported ee is 83% for the oxidation of trans-b-methylstyrene
using chiral Cr–salen catalysts, see: C. Bousquet and D. C. Gilheany,
Tetrahedron Lett., 1995, 36, 7739.
The asymmetric epoxidation of trans-b-methylstyrene can
become catalytic using 2,6-dichloropyridine N-oxide
(
Cl
2
pyNO) or O
2
as terminal oxidant. Similarly the catalytic
7 T.-S. Lai, R. Zhang, K.-K. Cheung, H.-L. Kwong and C.-M. Che, Chem.
Commun., 1998, 1583; T.-S. Lai, H.-L. Kwong, R. Zhang and C.-M.
Che, J. Chem. Soc., Dalton Trans., 1998, 3559.
trans-alkene oxidation is more enantioselective than that of the
analogous cis-b-methylstyrene oxidation. For example, with
8
9
Z. Gross and S. Ini, J. Org. Chem., 1997, 62, 5514.
W. Zhang, N. H. Lee and E. N. Jacobson, J. Am. Chem. Soc., 1994, 116,
2
Cl pyNO (0.146 mmol) as terminal oxidant and benzene as
solvent, the trans-b-methylstyrene (1 mmol) oxidation fur-
nished the trans-epoxide in 50% ee (70% yield, turnover = 70);
whereas the oxidation of the cis analogue afforded the cis-
epoxide in only 7% ee (70% yield, turnover = 66) under the
same conditions. On the other hand, 1a can also effect aerobic
asymmetric trans-b-methylstyrene epoxidation in benzene (9
4
25; W.-H. Fung, W.-Y. Yu and C.-M. Che, J. Org. Chem., 1998, 63,
7
715.
1
1
0 J. T. Groves and R. S. Myers, J. Am. Chem. Soc., 1983, 105, 5791.
1 D. Ostovic and T. C. Bruice, Acc. Chem. Res., 1992, 25, 314.
Communication 8/08776H
410
Chem. Commun., 1999, 409–410