The role of protic solvent in asymmetric hydrogenation of methyl levulinate in the presence of a ruthenium-containing catalyst

Article abstract of DOI:10.1007/s11172-007-0090-4

A comparative study of asymmetric hydrogenation and deuteration of methyl levulinate catalyzed by the Ru^II-(S)-BINAP-HCl system (BINAP is 2,2′-bis(diphenylphosphino)-1,1′-binaphthyl) in MeOH and MeOD was carried out. The results obtained suggest an important role of the protic solvent in the formation of catalytically active ruthenium complexes.

Full text of DOI:10.1007/s11172-007-0090-4

552
Russian Chemical Bulletin, International Edition, Vol. 56, No. 3, pp. 552—554, March, 2007
The role of protic solvent in asymmetric hydrogenation
of methyl levulinate in the presence of a rutheniumꢀcontaining catalyst
E. V. Starodubtseva, O. V. Turova, M. G. Vinogradov,ꢀ L. S. Gorshkova, and V. A. Ferapontov
N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences,
4
7 Leninsky prosp., 119991 Moscow, Russian Federation.
Fax: +7 (495) 135 5328. Eꢀmail: ving@ioc.ac.ru
A comparative study of asymmetric hydrogenation and deuteration of methyl levulinate
II
catalyzed by the Ru —(S)ꢀBINAP—HCl system (BINAP is 2,2´ꢀbis(diphenylphosphino)ꢀ
1
,1´ꢀbinaphthyl) in MeOH and MeOD was carried out. The results obtained suggest an imporꢀ
tant role of the protic solvent in the formation of catalytically active ruthenium complexes.
Key words: asymmetric catalytic hydrogenation, deuteration, hydrogen heterolysis, H/D exꢀ
change, ruthenium complexes, 2,2´ꢀbis(diphenylphosphino)ꢀ1,1´ꢀbinaphthyl (BINAP),
levulinic acid ester, chiral γꢀvalerolactone.
1
Earlier, we have investigated asymmetric hydrogeꢀ
the reaction products were carried out as described in the preꢀ
ceding paper. Here the catalyst for the asymmetric hydrogenaꢀ
tion and deuteration of oxo ester 1 was prepared in situ without
isolating the intermediate complex [((S)ꢀBINAP)RuCl2].
1
nation of levulinic acid esters catalyzed by the
Ru —(S)ꢀBINAP—HCl system (BINAP is 2,2´ꢀbis(diꢀ
phenylphosphino)ꢀ1,1´ꢀbinaphthyl). The resulting
II
(
(
S)ꢀγꢀhydroxy esters underwent in situ transformation into
S)ꢀγꢀvalerolactone with up to 99% ee.
Results and Discussion
Here we studied the role of solvent in the aforemenꢀ
tioned reactions with methyl levulinate (1) as a model
substrate (Scheme 1).
Our experiments showed (Table 1) that the solvent
nature strongly influences the rate and enantioselectivity
of the catalytic hydrogenation of oxo ester 1.
The highest conversion and enantioselectivity were
reached when the hydrogenation was carried out in waꢀ
terꢀfree MeOH and EtOH (see Table 1, entries 1—3); the
rate of the catalytic reaction in MeOH was appreciably
Scheme 1
Table 1. Asymmetric catalytic hydrogenation of compound 1 in
protic and aprotic mediaa
Entry
Solvent
t/h
Conversion
ee (S)
of compound 1
%
R = H (a), D (b)
1
2
3
4
5
6
7
MeOH
3
3
5
5
5
100
90
100
38
<5
74
<5
99
99
99
92
—
67
—
b
EtOH
EtOH
b
Experimental
b
i
Pr OH
THF
THF—Н О (9 : 1)
Commercial (S)ꢀBINAP, (COD)Ru(H C(Me)C=CH2)2
c
2
9
5
2
(
COD is cyclooctaꢀ1,5ꢀdiene) (Acros), and D and MeOD (piꢀ
2
CH Cl
2
2
lot plant Prikladnaya Khimiya, St.ꢀPetersburg) were used.
Methyl levulinate was prepared from levulinic acid (Acros) acꢀ
cording to a standard procedure.2 Gaseous H2 and D2 were
purified by passing through columns with a nickel—chromium
catalyst and molecular sieves. All solvents were dehydrated beꢀ
fore use. Catalytic hydrogenation of oxo ester 1 and analysis of
a
II
Ru —(S)ꢀBINAP—HCl; [1]/[Ru] = 200, [HCl]/[Ru] = 10,
–1
[1] = 1.7 mol L , 60 atm (H ), 60 °C.
b
2
1
Previous data on the hydrogenation of ethyl levulinate in EtOH
i
and isopropyl levulinate in Pr OH.
c
At 75 °C.
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 3, pp. 531—533, March, 2007.
066ꢀ5285/07/5603ꢀ0552 © 2007 Springer Science+Business Media, Inc.
1

Rutheniumꢀcatalyzed hydrogenation
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 3, March, 2007
553
i
higher than in EtOH and Pr OH. In waterꢀfree aprotic
D (%)
solvents such as THF and CH Cl (entries 5, 7), the
2
2
1
00
4
2
reaction virtually did not occur under the same condiꢀ
tions. In aqueous THF (entry 6), the conversion of oxo
ester 1 hydrogenated at 75 °C for 9 h was 74%. However,
the reaction in this solvent was substantially less enanꢀ
tioselective (67% ee) than in alcohols, which is attributꢀ
able to a change in the composition of the catalytic sysꢀ
8
6
0
0
3
3
40
20
tem in the presence of water. Earlier, we have found that
water deactivates the ruthenium catalyst in the deuteraꢀ
tion of oxo ester 1 in aqueous 90% MeOH at 60 °C beꢀ
fore complete conversion of oxo ester 1 into lactone 2
1
(
at ~70% conversion).
1
2
3
t/h
The considerable acceleration of the reaction in protic
Fig. 2. Plots of the content of lactone 2b in the mixture of
compounds 2a and 2b (D) vs. the reduction time in the reaction
of methyl levulinate 1 with deuterium in MeOH (1), hydroꢀ
gen in MeOD (2), hydrogen in MeOH—MeOD (1 : 1) (3),
and deuterium in MeOD (4). The catalytic system is
solvents can be due to their role in the formation of cataꢀ
lytically active ruthenium complexes. To elucidate this
issue, we performed a comparative study of Ruꢀcatalyzed
hydrogenation and deuteration of oxo ester 1 in MeOH,
MeOD, and MeOH—MeOD.
(COD)Ru(H C(Me)C=CH ) —(S)ꢀBINAP—HCl; [1]
=
2
–
2 2
1
A comparison of the kinetic curves in Fig. 1 shows
that the catalytic hydrogenation with hydrogen in MeOH
1.7 mol
L
;
[1]
:
[Ru]
:
[HCl]
=
200 : 1 : 10;
6
0 atm D (H ); 40 °C.
2
2
(
curve 1) and with deuterium in MeOD (curve 2) have
very close initial rates. This suggests that the kinetic isotoꢀ
pic effect of this reaction is absent or insignificant.
At the same time, one can assume that the heterolyꢀ
sis4,5 of D (H ) is substantially facilitated in such a polar
solvent as methanol. This is indicated by the data on the
isotopic enrichment dynamics for lactone 2 during the
catalytic reduction of oxo ester 1 with deuterium in MeOH
and with hydrogen in MeOD (Fig. 2).
Fig. 2, curve 1), lactone 2 contained ~10% deuterium in
the γꢀposition 30 min after the reaction started (which
corresponds to a ~20% conversion of compound 1) and
~50% deuterium after 4 h (90% conversion). When D2
was replaced by H2 and MeOH was replaced by MeOD,
the dependence became opposite (see Fig. 2, curve 2): the
deuterium content of the reaction product decreased from
100 to ~50% over the same period of time. In the reducꢀ
tion of oxo ester 1 by hydrogen in MeOH—MeOD (1 : 1),
the contents of lactones 2a and 2b in the initial period of
the reaction were also equal (see Fig. 2, curve 3). As
expected, during the reduction of oxo ester 1 by deuteꢀ
rium in MeOD, the isotopic enrichment of product 2 in
the γꢀposition changed only slightly (see Fig. 2, curve 4).
Some decrease in it (from 100 to 90%) was due to the
formation of MeOH as the result of partial H/D exchange
between oxo ester 1 and MeOD.
2
2
The dependences obtained suggest that the change in
the isotopic composition of lactone 2 (ratio 2a/2b) is
determined by the isotopic composition of the solvent
(
ratio MeOH/MeOD), regardless of whether hydrogen or
deuterium is used as the reducing agent. For instance, in
the reduction of oxo ester 1 with deuterium in MeOH (see
C (%)
1
00
1
The data obtained (see Fig. 2) can be explained by
rutheniumꢀcatalyzed H/D exchange between the reducꢀ
2
6
8
6
4
2
0
0
0
0
ing agent (D or H ) and the solvent (MeOH or MeOD)
2
2
(
Scheme 2), which occurs in parallel with the hydrogenaꢀ
tion of oxo ester 1 and is much more rapid than the major
hydrogenation (deuteration) reaction.
Scheme 2
2
4
t/h
Fig. 1. Plots of the conversion (C) of methyl levulinate 1
vs. the hydrogenation time in MeOH (1) and the deuteraꢀ
tion time in MeOD (2). The catalytic system is
(
1
6
COD)Ru(H C(Me)C=CH ) —(S)ꢀBINAP—HCl; [1]
=
2
–
2 2
1
.7 mol
L
;
[1]
:
[Ru]
:
[HCl]
=
200 : 1 : 10;
0 atm H (D ); 40 °C.
2
2

554
Russ.Chem.Bull., Int.Ed., Vol. 56, No. 3, March, 2007
Starodubtseva et al.
In this case, a change in the MeOH/MeOD ratio durꢀ
References
ing the catalytic reduction should correspondingly change
the deuterium content of the catalytically active Ruꢀconꢀ
taining intermediates, which are in rapid equilibrium; in
turn, their isotopic composition determines the ratio of
lactones 2a and 2b. As noted above, the initial amounts of
lactones 2a and 2b are equal at MeOH : MeOD = 1 : 1
1
. E. V. Starodubtseva, O. V. Turova, M. G. Vinogradov, L. S.
Gorshkova, and V. A. Ferapontov, Izv. Akad. Nauk,
Ser. Khim., 2005, 2301 [Russ. Chem. Bull., Int. Ed., 2005,
5
4, 2374].
. A. Hernandez, M. Marcos, and H. Rapoport, J. Org. Chem.,
995, 60, 2683.
2
3
1
(
see Fig. 1, curve 3). This can be due to equal steadyꢀstate
. E. V. Starodubtseva, O. V. Turova, M. G. Vinogradov, and
V. A. Ferapontov, Abstrs VII Ross. konf. "Mekhanizmy
kataliticheskikh reaktsii" (SanktꢀPeterburg, 3—8 iyulya 2006 g.)
concentrations of hydrogenꢀ and deuteriumꢀcontaining
ruthenium complexes. This fact is consistent with the
absence of the kinetic isotopic effect in the catalytic reꢀ
(
VII Russian Conf. "Mechanisms of Catalytic Reactions"
7
,8
duction of compound 1 (see Fig. 1). Earlier, intermediꢀ
(St.ꢀPetersburg, July 3—8, 2006)), Novosibirsk, 2006, Vol. I,
ates of the types Ru(H ) and Ru—H in equilibrium have
ORꢀIIꢀ24, 244 (in Russian).
2
been thought to participate in Ruꢀcatalyzed hydrogenaꢀ
tion of strongly enolized 1,3ꢀdiketones.
4. D. M. Heinekey and W. J. Oldham, Jr., Chem. Rev., 1993,
3, 913.
9
5
6
. G. J. Kubas, Catal. Lett., 2005, 104, 79.
. A. C. Albeniz, D. M. Heinekey, and R. H. Crabtree, Inorg.
Chem., 1991, 30, 3632.
Based on the results obtained, one can assume that the
Ruꢀcatalyzed asymmetric hydrogenation of γꢀoxo esters
in protic organic solvents involves heterolysis of molecuꢀ
lar hydrogen; a protic solvent facilitates this process and
takes an active part in the formation of catalytically active
ruthenium complexes that are in rapid equilibrium.
7
8
. R. Noyori and T. Ohkuma, Angew. Chem., Int. Ed. Engl.,
2
001, 40, 41.
. A. Mezzetti, A. Tschumper, and G. Consiglio, J. Chem. Soc.,
Dalton Trans., 1995, 49.
This work was financially supported in part by the
Division of Chemistry and Materials Science of the Rusꢀ
sian Academy of Sciences (Basic Research Program
OKhNM No. 1).
Received July 24, 2006;
in revised form February 8, 2007