ASYMMETRIC OF

HYDROGENATION

BISDIPHENYLPHOSPHINE V.

A.

Pavlov

and

E.

IN T H E COMPLEXES

I.

Klabunovskii

PRESENCE OF RHODIUM UDC 541.6:541.128:541.49:546.97:542.941.7

Information on the structure of bisdiphenylphosphine complexes of rhodium, which catalyze the asymmetric homogeneous hydrogenation of substituted ~-acylaminoacrylie acids (A) [i] and on the structure of the complex of the catalyst with a prochiral substrate (B) or the intermediate compound (C) [2-4] (Fig. 1), along with the results obtained in other studies [5-8], have provided a basis for formulating conclusions that are important in understanding the stereochemical mechanism of asymmetric hydrogenation, i) The planes of the phenyl rings of each phosphine group of the catalytic complex are orthogonal to each other [I, 3, 4, 8]. 2) The cationic and neutral catalytic complexes catalyze the formation of a product with the same configuration in both cases, and with similar values of the enantiomer excess (EE) [9, I0]. 3) The enantioselectivity of the catalyst is related to the eonformational rigidity of the diphosphine ligand [5, 7]. 4) When the catalytic complex interacts with H2, the diene part of its molecule is substituted [8]. 5) The stereoehemistry of asymmetric hydrogenation is determined in the stage of the coordination of the olefin to form a chelate complex [3, 11]. 6) The amide group of a-acetylaminoaerylie acids coordinates better than the carboxyl group in forming the chelate intermediate [3, 8, 12, 13]. 7) The amide coordinates the rhodium atom through the earbonyl oxygen atom, forming a relatively rigid ~-olefin, an O-carbonylrhodium ring [2, 3, 8, 11-14]. 8) H 2 (or D2) always undergoes cis-addition to the substrate [15, 16]. 9) The basic source of discrimination between the sides of the coordinated prochiraI olefin and the catalyst molecule is the chiral location of the phenyl groups around the phosphorus atoms [12]. The reaction conditions did not differ significantly among the experiments that led to the above conclusions. Moreover, these changes did not affect the configuration or EE of the product. For example, the solvent (alcohol or an alcohol/benzene mixture) [10, 17, 18], the mole ratio of substrate to catalyst (I 0-500) [I 7], the temperature (20-50~ [17], and the pressure (1-70 atrn) [i0] have practically no effect on the EE (to say nothing of the configuration) of the product within the intervals just mentioned. Therefore, the dependence of the product configuration on the catalyst structure can be studied by comparing the structures of the catalysts for the reactions under different conditions, within the indicated limits of variation. The chirality of the cation of [Rh(L)(1,5-COD)]BF4 (1) (in the scheme, the 1,5-eyelooetadiene, 1,5-COD, has been omitted for convenience of representation), where L is i, 2-bis [R-(o-methoxyphenyl)phenylphosphino]ethane [8] (Fig. 2), is due to the asymmetric phosphorus atoms. The chirality of the cation of [Rh(L)(NBD)]CIO 4 (II), where L is 2S, 3S-bis(diphenylphosphino)butane and NBD is norbornadiene [7], is due to the asymmetric carbon atoms. The chiral location of the aryl groups on the asymmetric phosphorus atoms, according to x-ray structure analysis (XSA), leads to a situation in which the five-membered chelate ring with achiral carbon atoms assumes a chiral folded conformation [8]. It can be assumed that the chiral 6 -conformation of the chelate ring with chiral carbon atoms of the complex (If), which has been established by x-ray structure analysis [7], leads to a ehiral location of the phenyl groups with achiral P atoms. It can be seen that the chiral position of the phenyl groups of the complexes (I) and (If) exists in enantiomeric ratios. We believe that such a correlation of the conformation of the chelate ring and the chirality of the position of the phenyl groups is general for the class of bisdiphenylphosphine complexes of rhodium. XS.4 of the complex (I) [8], the complex Ir(DIOP)(I, 5-COD)CI, where DIOP is 2S, 3S-O-isopropylidenedihydroxy-l,4-bis(diphenylphosphino)butane [I], and the intermediate compound based on the complex (II) [7], shows that in all cases, the planes of the quasiequatorial phenyl groups are perpendicular to the plane of the chelate ring, and the planes of the quasiaxial phenyl groups are orthogonal to the first. It can be assumed that this feature is also general for all Rh complexes of the indicated type.

N. D. Zelinskii Institute of Organic Chemistry, Academy of Sciences of the USSR, Moscow. T r a n s l a t e d f r o m Izvestiya Akademii Nauk SSSR, Seriya Khimieheskaya, No. 9, pp. 2015-2022, September, 1983. Original article submitted J a n u a r y 13, 1982. 1820

0568-5230/83/3209-1820507.50 9 1984 Plenum Publishing C o r p o r a t i o n

CHgR

H

-

-

"" C00H

MeCOHN

COOt{

CHaR

HO0

S

r

- / y" "vNRI~

P~

t~<..-,

{a)

.: COMe9 R

H -]+

~

(B)

' ~PN],,'

c--co(,Mol

(c)

Fig. 1+ Reaction of c~-acylaminoacrylic acid hydrogenation and s t r u c t u r e of catalytic complex (A) and intermediate c o m plexes (B) and (C).

(1) Fig. 2. P r o j e c t i o n s of cations of complexes (1) and (II) on equatorial plane. In analyzing the possible s t r u c t u r e s of the intermediate complex, taking into account all of the conclusions listed p r e v i o u s l y , we used the m o l e c u l a r model technique. * In all, four variants are possible in the c o o r d i n a tion of Z-c~ - a e e t y l a m i n o c i n n a m i c acid Z-(III) in the intermediate eomptex based on complex (I) (Fig. 3). The hydrogenation of Z-(III) through these four variants of the intermediate complex and subsequently through a s e m i hydrogenated i n t e r m e d i a t e compound [2, 3, 20] leads to optical i s o m e r s of N-aeetylphenylalanine (IV). In Fig. 4 we show the spatial a r r a n g e m e n t in variants 1 and 2 of the intermediate complex in the adjacent groups of the substrate of the catalytic complex for (I). It can be seen that variant 2 is m o r e hindered than variant 1 (this same conclusion is valid for variants 4 and 3). T h e r e f o r e , the reaction path through the intermediate complexes 1 and 3, leading to S-(IV), is m o r e favorable than through variants 2 and 4. The complex (II), from the standpoint of the location of the phenyt groups on the P a t o m s , is the antipode of complex (I) (see Fig. 2). In this e a s e , the distribution of the s t e r i c hindrances in variants 1-4 will be the opposite, and the reaction path through the i n t e r m e d i a t e complexes 2 and 4, leading to R-(IV), witl be p r e f e r r e d . These c o n f i ~ r a t i o n s have been obtained experimentally [7, 8]. In both c a s e s , the conformation of the f i v e - m e m b e r e d chelate ring of the catalytic complex d e t e r m i n e s the configuration of the product: The X-conformation leads to the S - p r o d u c t , and the 6 conformation leads to the R - p r o d u c t . The approach that we have proposed can be used to explain the known experimental data, not only those pertaining to the product configuration, but also to the EE. The m a x i m u m EE in the hydrogenation of Z-(III) was obtained in the case of complexes with ligands forming conformationMly rigid f i v e - m e m b e r e d chelate r i n g s . F o r example, in the hydrogenation of Z-(III) in the p r e s e n c e of (I), the (IV) is obtained with 94~0 EE (Table 1). Bulky substituents introduced into the earboxyl or amide group of the substrate do not affect the EE (Table 1), and this is consistent with the hypothesis that the reaction proceeds through the complexes 1-4. Actually, these substituents are oriented outside the intermediate complex (Fig. 3) and should not c r e a t e any s t e r i e hindrance in this complex. The EE changes very slightly when the NHCOMe group in Z-(III) is replaced by NHCOOEt or OCOMe, but it drops p r a c t i c a l l y to zero if the acylamino group is replaced by CH 3 (Table 1). On the other hand, the r e p l a c e m e n t of the COOH group by CH3 has almost no effect on the EE. $ This c o n f i r m s the coordination of Z-(III) by rhodium through the double bond and the acyiamino group or analogous group in the ease of a s y m m e t tie hydrogenation. 9 We used Buchi m o l e c u l a r m o d e l s . The length of the R h - P bond was taken as 2.3 A in a c c o r d a n c e with XSA data for [Rh(L) 2]C104, where L is 1,2-bis(diphenylphosphino)ethane [19]. r In this e a s e , the R-configuration cf the product f r o m the hydrogenation coincides with the S-configuration of (IV), since r e p l a c e m e n t of COOH by Me changes the o r d e r of seniority. 1821

H

Mg

U

8

I.I

i,ooo

/

H

1

,

/t OM0

.4"

Itz%;,

T' J "-i{

'~ t

2

Fig. 3 Fig. 4 Fig. 3. Sterically possible variants in the coordination of Z-acetylaminocinnamic acid in i n t e r mediate complex based on complex (I), and configurations of products. F i g . 4. F r a g m e n t s of i n t e r m e d i a t e c o m p l e x b a s e d on c o m p l e x (I) for v a r i a n t s 1 and 2.

Stable and very high values of the EE of (IV), N-AcAla, N-AeLeu, and N-AcTyr were obtained upon hydrogenation of their predeeessors in the presence of the complex (If) and [Rh(L)(NBD)]CIO4 (V), where L is R-1,2-bis(diphenylphosphino)propane (see Table 1). These experimental data provide support for the structure of the intermediate complex (Fig. 3) with the phenyl group of the substrate directed toward the outside of the complex and not creating any steric hindrance. In this case, the k-conformation [12] of the chelate ring of the complex (V) leads, as would be expected, to the S-product. The complexes [Rh(L)(NBD)]CIO4 (VI) and Rh(L)(NBD)CI (VII), where L is S-1-phenyl-l,2-bis(diphenylphosphino)ethane, must have the 6 -conformation of the chelate ring, since the CGH5 group should be located equatorially. Since the chlorine atom of neutral complexes of such a type is axially located [1, 11, 20], its position does not influence the distribution of the steric hindrances in varim~ts 1-4 of the intermediate complex. Therefore, variants 2 and 4 are still preferred (see Fig. 3), leading to R-(IV) in agreement with experiment [10] (see Table 1). The same as in the case of the complex (1), bulky substituents introduced into the carboxyl or amide group of the substrate do not affect the EE of the product from hydrogenation in the presence of the complexes (VI) and (VII) (Table 1); the 5 -conformation leads to the R-product. Complexes with the most rigid chelate ring, for example (VIII), are obtained on the basis of conformationally rigid ligands such as trans-l,2-bis(diphenylphosphino)-substituted small rings [18]. In accord with the hypothesis that the enantioseleotivity of the catalytic complex depends on the conformational rigidity of the chelating diphosphine ligand [7], the hydrogenation of Z-(Ill) in the presence of the complex (VIII) leads to a product with the maximum EE (Table 1). This complex has the X-conformation of the chelate ring, and in this case, consequently, the above-mentioned relationship between the conformation of the chelate ring and the configuration of the product is preserved. A catalytic system including O-isopropylidene-2,3-dihydroxy-1,4-bis(diphenylphosphino)butane (DIOP) has been thoroughly studied [17, 20]. The molecular models show that the complex 2R, 3R-DIOP/Rh(I) (IX) has a seven-membered chelate ring with a distorted 5 -conformation. Here the spatial interrelationships are the same as those of the bisdiphenylphosphine ligand and the coordinated substrate that are characteristic of variants 2 and 4 are preferred, leading to R-(IV) (Fig. 3). The complex with 2S, 3S-DIOP (X) has the opposite

1822

TABLE

t Su['str~te

ConformaComplex tion of chelate

(]) (l) (i) (1) (]) (1) (l) (i]) (H) (H) (H) (v) (v) (v) (v) (Vl) (v]) (Vl) (VH) (VII) (vH) ,(viii)

(Ix) (x) (x) (x) (x)

Q

,x

Rl\

H/C= t/'

R2

Ph Ph Ph Ph Ph Ph Ph Ph

NHCOMe NIICOMe NHCOPh NHC00Et OCOMe Me NHCOMe

H

5 6 }~

% Z 6 6 5 6 5

8 ~, 8 X

2~

), X

(Me) 2CH p-HOPh Ph H (Me) 2CH p-HOPh Ph Ph Ph Ph Ph Ph ph Ph Ph Ph Ph Ph

Product

C/it~ Nil' R3

NttCOMe NHCOMe NHCOMe NHCOMe NHCOMe NHCOMe NHCOMe NHCOMe NHCOMe NHCOMe NHcOPh NHCOMe NHCOMe NHCOPh NHCOMe NHCOMe INHCOMe NHCOMe NHCOPh NHCOPh

EE,%

COOH I ?4 COOMe I

C00H COOH Me COOH COOH COOH COOH COOH COOH COOH COOH COOH COOMe C00H COOH COOMe 1

configuRef, ration

~6 ~3 B9

~0

S S S S S

1

R

50 B9 9t 93 88 90 90 87 89 78 85 84 82 76 COOH 1 76 96 COOH 82 COOH 8t COOH C00Me 55 C00H 70 COOMe 37,5

R R R

R R S S S S

[8] [8] [S]

[81 [8] [8] [8] [7] [71

R

[71 [7] [t21 [t21 [t2] [t21 [101 [i0] [t0] [I01

R R

U0]

R R R

S R

D81 [q

S S S S

[20] [20] [20] [20]

% F i g . 5. F r a g m e n t s o f i n t e r m e d i a t e c o m p l e x b a s e d o n c o m p l e x (X) f o r variants 1 a n d 2.

aH

2

absolute configuration in comparison with the complex (IX); therefore, the chelate ring in this case must have a distorted ~-conformation. In the complex (X), the C6H 5 groups are located in opposite mirror-image positions on the P atoms; therefore, the distribution of steric hindrances in variants I and 4 will be the opposite. In this case, the reaction path through variants 1 and 3, leading to S-(IV), is preferred. And exactly these config~arations of the product were obtained in the hydrogenation of Z-(III) in the presence of these complexes (Table I). A comparison of the spatial positions of the carbonyl group of the substrate and the phosphine C6H~ group of the ligand for variants I and 2 on the basis of the complex (X) (Fig. 5) with the analogous picture for the c o m plex (1) {Fig. 4) shows that the favorable variant 1 for the eomplex (X) is sterically m o r e hindered than the corresponding variant for the complex (I). For the unfavorable variants 2, the opposite picture is observed (the s a m e is valid for 3 and 4). Therefore, the preference for the favorable variants over the unfavorable is less for the complex (X) than for the complex (I), and hence the E E of the hydrogenated product should be smaller. Thus, by the use of molecular models of the intermediate complex, w e can explain the small decrease of the E E of the product from the hydrogenation of Z-(III) in the presence of the complex (X) in comparison with the complex (1) (81% and 94%). It is obvious that the introduction of a bulky substituent into the carboxyl group should introduce additional steric hindrance into the favorable variant 1 for the complex (X), and that this will reduce its advantage and correspondingly will reduce the E E of the product (Table i). In all eases, the configuration of the product from the hydrogenation of Z-(HI) in the presence of the eomplexes (IX) and (X) corresponds to the conformation of the s e v e n - m e m b e r e d ehelate ring (S-X and R-5). The chelate ring of the eomp!exes with D I O P is formed by trans-bis(diphenylphosphinomethyl)dioxolane as the ligand. T h e c a r b o c y c l i c a n a l o g o f t h i s l i g a n d i s 1R, 2 R - t r a n s - l , 2 - b i s ( d i p h e n y l p h o s p h i n o m e t h y l ) c y c l o butane. I n t h e p r e s e n c e o f t h e c a t a l y t i c c o m p l e x b a s e d o n t h i s l i g a n d ( X I ) , Z - ( H I ) i s h y d r o g e n a t e d w i t h 86% E E o f R - ( I V ) [5]. H o w e v e r , i n t h e p r e s e n c e o f 1S, 2 S - t r a n s - l , 2 - b i s ( d i p h e n y l p h o s p h i n o m e t h y l ) e y c t o h e x a n e - r h o d i n m (I) ( X f I ) , Z - ( I I I ) i s h y d r o g e n a t e d t o f o r m S - ( I V ) w i t h a n E E o f o n l y 35% [6]. W e c a n a s s u m e t h a t t h e l o w e r E E

1823

b

I:I2y iH2 Ph2P

PPh2

y Ph2P

el PPh2

Me~NI Ph2P

N--Mel PPh2 (s) -.,- o > [(,~-so)-~(F+9o)]> o----(R)

Fig. 6 Fig. 7 Fig. 6. Structure of ligands of complexes (XII), (XIV), and (XVII), and configuration of product from the hydrogenation of Z-(Ill) in the presence of these complexes. Fig. 7. IV[aemonic rule of sectors. is related to eonformational lability of the chelate ring f o r m e d by the cyclohexanediphosphine. It is known that in the c a s e s of the complexes (XI) and (XII), t h e r e is also a relationship between the conformation and configuration that is analogous to that o b s e r v e d p r e v i o u s l y . The m o l e c u l a r models d e m o n s t r a t e the possibility that there may be two conformations of the chelate ring in the complex (XII) based on 2S, 4S-N-butyloxycarbonyl-4-diphenylphosphino-2-diphenylphosphinomethylpyrrolidone [21]. T h e r e f o r e , the experimentally o b s e r v e d , e x t r e m e l y small value of the EE of the product f r o m the hydrogenation of Z-(III) in the p r e s e n c e of this complex can be explained by the conformational equilib r i u m 5-~ k. However, the Z-(III) in the p r e s e n c e of Et3N is hydrogenated with EE 91% R-(IV) [21]. This unexpected result can be explained on the assumption that the Et3N affects the eonformational equilibrium, for example, by being coordinated along the axial position of the catalytic complex. In this c a s e , the k - c o n f o r m e r is s t e r i c a l l y hindered, and the 5 - c o n f o r m e r is p r e f e r r e d . Actually, in the p r e s e n c e of this complex, the p r o duet with the R-configuration is f o r m e d , evidence for the 5 - c o n f o r m a t i o n of the catalytic complex. The situation is the same in the case of the complex with 2R, 5R-bis(diphenylphosphinomethyl)-l,3-dioxolane [22]. In the p r e s e n c e of 1S, 2 S - t r a n s - 1 , 2 - b i s ( d i p h e n y l p h o s p h i n o x y ) c y c l o h e x a n e - r h o d i u m (I) (XIV), which has the h - c o n formation of the chelate ring, Z-(III) is hydrogenated with 68% R-(IV) [23]; i . e . , the EE is here twice as great as in the c a s e of the diphenylphosphinomethyl derivative of cyclohexane. P r o b a b l y , the chelate ring formed by the cyclohexanediphosphinite is conformationally m o r e rigid than in the case of the cyelohexanediphosphine. As shown by the m o l e c u l a r m o d e l s , the cyelohexane ring in m e t h y l - 2 , 3 - b i s (O-diphenylphosphino)-4,6-O-benzylid e n e - f l - D - g l u c o p y r a n o s i d e (XV) exists only in the c r o s s c o n f o r m a t i o n , since the CGH5 group must occupy the equatorial position. Consequently, the chelate ring of the complex (XVI) f o r m e d by this ligand must be c o n f o r mationally m o r e rigid than in the c a s e of the complex with the cyclohexanediphosphinite. And in fact, in the case of this complex, Z-(III) is hydrogenated to S-(IV) with EE 91% [24]. Ligands of related s t r u c t u r e differ in absolute configuration: 1 R , 2 R - t r a n s - 1 , 2 - b i s ( d i p h e n y l p h o s p h i n o methylamino)eyclohexane (XVIi) [25] has the opposite absolute configuration in c o m p a r i s o n with the two other ligands (Fig. 6). However, in the p r e s e n c e of any of the catalytic complexes with these ligands, Z-(III) is hydrogenated to form a product with the s a m e configuration in all c a s e s [5, 23, 25]. This contradiction can be explained by an analysis of the m o l e c u l a r models of the catalytic complex. Actually, of the two possible conformations of the complex (XVII) with the aminophosphine ligand, the h-conformation is m o r e highly s t r e s s e d than the 5 - c o n f o r m a t i o n . However, owing to the mutual r e p u l s i o n of the filled orbitals of the C1 atom and the unshared electron pair of the N a t o m , the 5 - c o n f o r m a t i o n is less advantageous. Consequently, the 1824

correlation between the conformation of the complex and the configuration of the product is preserved in the case of complexes with these three particular iigands, the same as for the ligands discussed previously. The conformation and the conformational rigidity of the chelate ring of the catalytic complex are important so far as they fix the chiral position of the phosphine phenyl groups. This location creates chiral steric hindrances to the re- or si-side of the substrate cooi~dinated in the intermediate complex, and it determines the config-dration of the product. We are proposing a mnemonic rule of sectors for determining the configuration of the product on the basis of the spatial position occupied by the phosphine phenyi groups in the catalytic complex (Fig. 7). Here we show a projection of the chelate ring formed by the diphosphine ligand of the catalytic complex (1) (see Fig. 7a) if viewed along the P-P axis (a view from the side shown by the arrow), such that the front 1~ atom occupies the center of the circle and the metal atom is located to the left on a horizontal line (Fig. 7b). The location of the front CcH 5 groups at the upper right and lower left or at the uppe~ - left and lower right sectors determines the S- or R-configuration of the product, respectively. The angles ~, p , and T are linked to the configuration of the product by simple relationships (Fig. 7). According to the magnitudes of T, the catalytic complexes with five- and seven-mernbered chelate rings are divided into three groups: T = 12~ 30~ and 38 ~ With the aim of determining the relationship between T and EE, we singled out the most enantioselective complexes from each group, with the exception of the maximally enantioselective complex (VIII), since owing to the high stress of the chelate ring formed by 2R,3Rbis(diphenylphosphino)norbornene, we were not able to measure the angle Y in the molecular models. In the presence of these complexes, Z-(Ill) is hydrogenated with an EE that is the closest to the theoretically possible value in the case of an ideal experiment. The dependence of EE on T for these complexes is linear:

EE (%) = 1 0 0 - T/2 Consequently, the EE values a r e the g r e a t e s t in the case of complexes with minimum values of T. But with T = 0, a chiral location of the phenyl groups is impossible. T h e r e f o r e , t h e r e are three consequences from this equation: 1) The absolute magnitude of the EE (100%) is unattainable. 2) The enantioseleetivity of a c a t a lyst will depend on the ehirality due to the orthogonal position of the planes of the CGH5 groups to a far g r e a t e r d e g r e e than on the chirality due to e q u a t o r i a l - a x i a l position of the phenyl groups. 3) The highest EE can be obtained for complexes with the lowest angle T at which chira]ity of the orthogonal position of the planes of the phenyl groups is p r e s e r v e d . CONCLUSIONS 1. The s t r u c t u r e of the catalytic complex and ~he product from the hydrogenation a r e linked 3y a conf o r m a t i o n - configuration relationship : ~-S, 5 - R . 2. The magnitude of the e n a n t i o m e r e x c e s s of the product is linearly related to the conformational angle 7 of the catalytic complex. LITERATURE 1. 2. 3. 4. 5. 6. 7. 8. 9.

10. 11. 12. 13. 14. 15. 16.

CITED

N . V . Kagan, P a r e Appl. C h e m . , 43, 401 (1975). A.S.C. Chan and J. Halpern, J. Am. Chem. S o c . , 102, 838 (1980). A.S.C. Chan, J. J. Pluth, and J . Halpern, J. Am. Chem. S o c . , 102, 5952 (1980). A.S.C. Chan, J. J. Pluth, and J. Halpern, Inorg. Chim. Acta, 37, L477 (1979). R. G l a s e r , M. Twaik, S. Geresh, and J . Blumenfeld, Tetrahedron L e t t . , 4635 (1977). R. G l a s e r , S. Geresh, J . Blumenfeld, and M. Twaik, T e t r a h e d r o n , 3__%4,2405 (1978). M . D . F r y s u k and B. Bosnich, J. Am. Chem. S o c . , 9__99,6262 (1977). B . D . Vineyard, W. S. Knowles, M. J. Sabacky, G. L. Bachman, and D. J. Weinkauff, J . $~n. Chem. S o e . , 9__99,5946 (1977). R. G l a s e r , S. Geresh, and J. Blumenfeld, J. Organomet. C h e m . , 112, 355 (1976). R. B. King, J . Bakos, C. D. Hoff, and L. Marko, J. Org. C h e m . , 444, 1729 (1979). J . M. Brown and P. A. Chaloner, Chem. Commun, p. 321 (1978). M. D. F r y s u k and B. Bosnieh, J . Am. Chem. S o c . , 100, 5491 (1978). J . M. Brown and P. A. Chaloner, J. Am. Chem. S o c . , 102, 3040 (1980). D. Sinou and H. B. Kagan, J. Organomet. C h e m . , 114, 325 (1976). C. D e t e l l i e r , G. Gelbard, and H. B. Kagan, J . Am. Chem. S o c . , 100, 7556 (1978). G. D e s c o t e s , D. Lafont, D. Sinou, J. M. Brown, P. A. Chaloner, and D. P a r k e r , Nouv. J. C h i m . , 5, 167 (1981).

1825

17. 18. 19. 20. 21. 22. 23. 24. 25.

H . B . Kagan and T. P. Dang, J. Am. Chem. S o c . , 9_44, 6429 (1972). H. B r u n n e r and W. P i e r o n e z y k , Angew. C h e m . , 1_88, 620 (1979). M . C . Hall, B. T. Kilbourn, and K. A. T a y l o r , J. Chem. See. A, 2539 (1970). G. Gelbard, H. B. Kagan, and R. Stern, T e t r a h e d r o n , 32, 233 (1976). K. Aehiwa, d. Am, Chem. S o c . , 9_.~8, 8265 (1976). G. D e s c o t e s , D. Lafont, and D. Sinou, J. Organomet. C h e m . , !50 , C14 {1978). M. Tanaka and I. Ogata, J. Chem. S o c . , Chem. C o m m u n . , 735 (1975). Y. Sugi and W. 1R. Cullen, Chem. L e t t . , 39 (1979). K. H~uaki, K. Kashiwabara, and J . F~tjita, Chem. L e t t . , 489 (1978).

ENANTIOSELECTIVE ACETOACETATE

HYDROGENATION

ON C O P P E R -

NICKEL

INFLUENCE

O F p H ON M O D I F I C A T I O N

NATURE

THE

V.

OF V.

OF

ETHYL

CATALYSTS: AND

THE

MODIFIER

Chernysheva

and

E.

I.

UDC

Klabunovskii

541. 128.1

It was shown in [1, 2] that Cu-Ni catalysts (CNC) of variable composition, modified with (+)-tartaric acid, manifest enantioselectivity in the hydrogenation of ethyl acetoacetate (EAA) and acetylacetone. However, the ini]uence of the pH of the modifying solution and {he nature of the modifier on the optical yield had not been investigated. Here we are reporting on an investigation of CNC of variable composition modified with tartaric acid and amino acids, in the hydrogenation of EAA, as influenced by variations of pH of modification and the catalyst composition. EXPERIMENTAL The starting materials were as follows: freshly distilled EAA, bp 63~ (12 ram); amino acids and tartaric acid with values of [c~]D corresponding to those reported in the literature. The CNC was obtained by reducing a mixture of oxides in a stream of N2-H 2 (i0:I) for 2 h at 180~ and 3 h at 280~ following [3]. The CNC was modified with 0.5% solutions of the acids in a l-h treatment at 50~ using 200 ml of the modifier solution per gram of CNC (the modification with the amino acids was performed at 50~ and at the pH of the isoelectric point of the particular amino acid). The following conditions were used in autoclave hydrogenation: for the CNC-tartaric acid, 5 ml E~CA without solvent, 0.5 g CNC, 100~ i00 atm, hydrogenation time 3.5 h; for the CNC-amino acid, 5 ml EAA, 0.25 g CNC, 130~ i00 arm, 3.5 h. The hydrogenated products were analyzed by GLC and p o l a r i m e t r y . The LKhM-8MD c h r o m a t o g r a p h had a k a t h a r o m e t e r detector and a glass column packed with 10% PEG M 15,000 on Chromatone N-AW. The p o l a r i m e t e r was a H i l g e r - W a t t s i n s t r u ment (D line of Na). The optical yield was calculated, a s s u m i n g for optically pu:,e ethyl ~ -hydroxybutyrate (EHB) [~ ]~ = 24 ~ At conversions below 7%, a c o r r e c t i o n was app]ied for the change of [~ ]~ with the con-

centration of EHB in the c a t a l y s a t e . DISCUSSION

OF

RESULTS

The reaction rate v, as determined from the conversion achieved in 3.5 h (Fig. I, curve I), and the optical yield p (Fig. i, curve 2) vary antibatically with the pH of the solution of (+)-tartaric acid used "co modify the catalyst with the 50:50 Cu:Ni composition, the reaction rate v reaching quite high levels at pH 3 and 12 (v = 0.26, conversion 70%), approaching the effectiveness of the unmodified catalyst (v = 0.36). This is explained by the low degree of adsorption of the tartaric acid on the catalyst surface, as indicated by the low values of p at these levels of pH. With pH > 4, the optical yield tends to increase, but the pH 6-8.5 region has not yet been investigated experimentally. As can be seen from Fig. 1 (curves 3-5), in the case of Ni catalysts, the optimal pH is in the 5-9 range. For the Cu catalysts, the pH range is somewhat lower, and a maximum in the

N. D. Zelinskii Institute of Organic Chemistry, Academy of Sciences of the USSR, Moscow. T r a n s l a t e d from Izvestiya Akademii Nauk SSSR, Seriya Khimicheskaya, No. 9, pp. 2023-2026, September, 1983. Original a r t i c l e submitted D e c e m b e r 6, 1982.

1826

0568-5230/83/3209-1826507.50 9 1984 Plenum Publishing C o r p o r a t i o n