Influence of the nature of chiral auxiliaries on the diastereoselective hydrogenation of ortho-substituted benzoic acid derivatives

Article abstract of DOI:10.1016/S0957-4166(00)00112-9

The diastereoselective hydrogenation of o-toluic acid or o-methoxy benzoic acid covalently bound to different chiral auxiliaries was performed on Rh and Ru supported catalysts. The cis-isomers were formed predominantly, with a diastereoselectivity largely

Full text of DOI:10.1016/S0957-4166(00)00112-9

Tetrahedron: Asymmetry 11 (2000) 1809±1818
In¯uence of the nature of chiral auxiliaries on the
diastereoselective hydrogenation of ortho-substituted
benzoic acid derivatives
Michele Besson,* Pierre Gallezot, Samuel Neto and Catherine Pinel
Institut de Recherches sur la Catalyse-CNRS, 2, Avenue Albert Einstein, 69626 Villeurbanne Cedex, France
Received 2 March 2000; accepted 23 March 2000
Abstract
The diastereoselective hydrogenation of o-toluic acid or o-methoxy benzoic acid covalently bound to
di€erent chiral auxiliaries was performed on Rh and Ru supported catalysts. The cis-isomers were formed
predominantly, with a diastereoselectivity largely in¯uenced by the structure of the chiral inductor and the
steric hindrance brought for the preferential adsorption of one face of the aromatic substrate. The e€ect of
the functional group on the proline auxiliary (alcohol or ester groups susceptible to modify the anchoring
of the aromatic substrate) was weak. Hydrogenolysis occurred rather extensively with the methoxy benzoic
acid and constituted the most important hydrogenation pathway on Rh/C. The presence of the CˆO group
in the pyroglutamic acid methyl ester is a determining factor for obtaining good diastereoselectivity.
# 2000 Elsevier Science Ltd. All rights reserved.
1. Introduction
Selective synthesis of optically active alicyclic compounds is of interest to the pharmaceutical
and agrochemical industries, because these compounds are constituents of biologically active
molecules.1 One attractive way to access these molecules is the asymmetric catalytic hydrogenation
of the corresponding prochiral aromatic compound. Among the possible di€erent methods to
achieve this, the diastereoselective hydrogenation of disubstituted aromatic rings over hetero-
geneous catalysts has attracted some attention in recent years.2 7 In this strategy, the substrate is
temporarily covalently attached to a chiral auxiliary, before the hydrogenation over supported
metallic catalysts to the pairs of cis- and trans-isomers. The bulkiness of the chiral auxiliary causes
the molecule to approach the surface of the catalyst, with the face which produces the least steric
hindrance toward the catalyst surface.8
* Corresponding author. Fax: (+33) (0)4 72 44 53 99; e-mail: mbesson@catalyse.univ-lyon1.fr
0957-4166/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(00)00112-9

1810
M. Besson et al. / Tetrahedron: Asymmetry 11 (2000) 1809±1818
We have reported recently2,3 on the diastereoselective hydrogenation of o-toluic acid, chosen as
a model of disubstituted aromatic compounds. We have shown that, using supported metals, the
hydrogenation yielded the corresponding cis-cyclohexane derivatives with high yield and
enantiomeric excess. The diastereoselectivity of the reaction was largely in¯uenced by the nature
of the chiral auxiliary. Whereas d.e.s up to 68% were achieved using proline esters as the chiral
auxiliary,2 improvement up to 95% was observed with pyroglutamic acid esters.3 The importance
of the aromatic compound-auxiliary moiety was also found to be an important factor by
Honig et al.5 in the hydrogenation of vanillic acid. After modi®cation of either the carboxylic
acid function with (S)-proline or (^)-menthol, or the hydroxyl group with N-acetyl-(S)-proline,
only low diastereoisomeric excesses for the cis-products were obtained (<6%). By making the
structure more rigid, with the attachment of the chiral auxiliary to both the hydroxy and
methoxy groups, an increase in d.e. to 25% over an Ru/C catalyst was found. In the group
of Prins,6,7 (S)-proline-modi®ed anthranilic acid was hydrogenated on Rh and Ru catalysts with
moderate selectivity to the cis-isomers, but with high diastereoselectivity (up to 96%), probably
due to the rigidity of the system. On the other hand, hydrogenation of (S)-proline-modi®ed
o-toluidine yielded the cis-isomers (cis to trans ratio of 4±5) with d.e.=41±57% on di€erent
rhodium catalysts.
Considering the dramatic in¯uence of the nature of the chiral auxiliary on the diastereo-
selectivity, we tested other chiral inductors for the hydrogenation of o-toluic acid. The use of
prolinol allowed us to study the in¯uence of the ester group compared to that of alcohol in the
chiral auxiliary. (S)-(2-Pyrrolidinylmethyl)pyrrolidine is a very bulky chiral ligand which should
modify the selectivity. Finally, (R)-pantolactone has often been used in diastereoselective
reactions because of the easy esteri®cation of the alcohol function.9 The in¯uence of the ortho-
substitutent group in the benzoic substrate was also studied (methoxy group in place of methyl
group).
2. Results and discussion
2.1. Synthesis of substrates 1 and identi®cation of hydrogenation products
The substrates 1c, e, f (Table 1) were synthesized in good yields (65±87%) from o-toluoyl
chloride and the corresponding auxiliaries according to Scheme 1. (S)-N-(2-Methoxybenzoyl)-
prolinol 1d was commercially available (Aldrich) and used without further puri®cation.
Scheme 1. (i) R*H, NEt3, CHCl3; (ii) catalyst, EtOH, 5 MPa hydrogen, room temperature

M. Besson et al. / Tetrahedron: Asymmetry 11 (2000) 1809±1818
1811
Table 1
In¯uence of the nature of the chiral auxiliary on the diastereoselectivity and the initial reaction rate of
the hydrogenation
Typically, the asymmetric hydrogenation of the aromatic substrates 1 was performed in ethanol
at room temperature and under 5 MPa hydrogen pressure. Rhodium and ruthenium catalysts,
known as the most active catalysts for hydrogenation of aromatic substrates and able to catalyze
these reactions at mild process conditions, were used.

1812
M. Besson et al. / Tetrahedron: Asymmetry 11 (2000) 1809±1818
The hydrogenation resulted in the formation of four cyclohexane diastereoisomers, the pair of
cis-isomers 3 and 30 and the pair of trans-isomers 4 and 40. All hydrogenation reactions of the
di€erent substrates yielded predominantly the two cis-diastereoisomers with a cis to trans ratio
higher than 7.5. The diastereoisomeric excess was then reported for the cis-isomers and de®ned as
d.e.=(%3^%30)/(%3+%30)Â100; usually, the d.e. did not vary with the conversion of 1. During
the course of the reaction variable amounts of cyclohexene intermediate 2 were also detected. The
position of the double bond in 2 was veri®ed by DEPT 13C NMR spectroscopy for the proline
methyl ester auxiliary.2 The concentration of 2 went through a maximum during the reaction, the
amount of which depended on the substrate. In some cases (R=OMe), hydrogenation was
accompanied by extensive hydrogenolysis and signi®cant amounts of the hydrogenolysis±
hydrogenation compound 5 were detected.
The main results achieved in the hydrogenation of these substrates, with di€erent metal sup-
ported catalysts, are reported in Table 1. For comparison, data corresponding to the substrates
1a and b with the proline and pyroglutamic acid auxiliaries described in previous studies are also
mentioned.2 4
2.2. Hydrogenation of (S)-N-(2-methylbenzoyl)-prolinol 1c
Hydrogenation of (S)-N-(2-methylbenzoyl)-prolinol 1c resulted in the formation of ®ve com-
pounds analyzed by gas chromatography and identi®ed by GC±MS as 2, 3 and 30, 4 and 40. The
absolute con®guration of the hydrogenated diastereoisomers was not identi®ed directly, but
attributed by analogy with the hydrogenated products of (S)-N-(2-methylbenzoyl)proline methyl
ester 1a. The comparison of the initial reaction rates (Table 1) shows that the replacement of the
methyl ester group in the proline auxiliary by a CH2OH group resulted in a reaction which pro-
ceeded more slowly. The initial reaction rates are 2 mol h^1 molRh and 9.5 mol h^1 molRu for
hydrogenation of 1c, compared to 15 mol h^1 molRh and 33 mol h^1 molRu for hydrogenation of
1a. A better availability of the oxygen lone pairs of the proline ester for stronger interactions
between the substrate and the metallic surface may explain these di€erences.
Fig. 1 gives the composition of the reaction medium versus time, when the reaction was per-
formed on Ru/C. The pathway was very close to that observed with substrate 1a.2 The aromatic
ring was mainly hydrogenated to the two cis-isomers with a low and constant d.e.=10% in favor
of diastereoisomer 3. The cyclohexenic compound accumulated and reached a maximum yield of
47%. When the amount of unreacted 1c approached zero, the accumulated intermediate was
hydrogenated slowly to give the cis-isomers as main products. However, the diastereoselectivity
of hydrogenation was higher than that resulting from the hydrogenation of the parent aromatic
compound, as indicated by the slight increase of d.e. up to 19% at nearly complete conversion of 2.
Very similar results were obtained on Rh/C catalyst, but at a lower rate. Upon hydrogenation
of intermediate 2 (maximum yield 35%), the d.e. value, initially 9%, increased to 17% in favor
of 3.
If we compare the e€ect of the substituent at the stereogenic carbon on the pyrrolidine group,
replacing an ester group by an alcoholic functional group does not seem to exert any supple-
mentary orientating e€ect.
The in¯uence of the addition of an amine, which was found to improve the diastereoselectivity
in hydrogenation of 1a and b, was not studied in detail, as this caused a signi®cant decrease in the
activity.

M. Besson et al. / Tetrahedron: Asymmetry 11 (2000) 1809±1818
1813
Figure 1. Distribution of products versus time for hydrogenation of 1c over Ru/C. Reaction conditions: 2.27 mmol
substrate, 0.06 mmol Ru, 130 ml ethanol, 25ꢀC, 5 MPa hydrogen. * 1c, ~ 2, ^ 3, ^ 30, ! 4+40
2.3. Hydrogenation of (S)-N-(2-methoxybenzoyl)-prolinol 1d
Catalytic hydrogenation of commercial (S)-N-(2-methoxybenzoyl)-prolinol 1d over di€erent
catalysts resulted in four hydrogenated products. They were identi®ed as the two cis-diastereo-
isomers 3 and 30, the cyclohexenic intermediate 2 and (S)-N-cyclohexane carbonyl prolinol, the
saturated compound 5 in which the methoxy group has been eliminated. This latter compound
was identi®ed by GC±MS and the reference was also synthesized from cyclohexane carboxylic
acid chloride and (S)-prolinol. No trans-isomers were observed.
The initial reaction rate was apparently much faster on Rh/C than on Rh/Al2O3 or Ru/C (for
^1
instance, 12 mol h^1 mol on Rh/C compared to 2 mol h^1 mol on Rh/Al2O3, Table 1).
^1
Rh
Rh
However, these initial reaction rates were measured from the disappearance of the aromatic
substrate, and one must take into account that 1d is converted either by hydrogenation or by
hydrogenolysis of the methoxy group. The splitting of this group attached to the ring system is
possible by cleavage of the C±O bond adjacent to the ring.10,11 The proportion of hydrogenolysis
was found to be dependent on the catalyst. While hydrogenolysis became even a major part of the
reaction on Rh/C (52% of hydrogenolysis±hydrogenation product 5 detected), it occurred to a
lesser extent on Rh/Al2O3 or Ru/C (22% of 5). In all experiments, only small amounts of cyclo-
hexenic compound were formed (5%) and the d.e. measured on the cis-derivatives remained
moderate, in the range 28±33% (Table 1). However, one can note that these values are higher
than those observed with o-toluic acid grafted to the same chiral inductor (substrate 1c, d.e. in the
range 9±19%). The methoxy substituent, by steric e€ect and additional electronic interactions
with the metallic surface (haptophilicity), may contribute to the favorable adsorption of one face of
the aromatic substrate and enhance the diastereodi€erentiation. Such an interaction was already
proposed2 to explain the higher selectivity observed in hydrogenation of 1a on ruthenium catalysts,
whose anity for oxygen groups in the substrate is higher than rhodium. Also, this haptophilic

1814
M. Besson et al. / Tetrahedron: Asymmetry 11 (2000) 1809±1818
e€ect was suggested in a recent work12 dealing with hydrogenation of indane and tetraline sub-
strates monosubstituted with hydroxyl, amino and methyl groups.
As shown by the d.e. results (Table 1), the e€ects of support and metal are negligible for sub-
strate 1c. The addition of an amine to the Rh/C catalyst drastically suppressed the reaction.
2.4. Hydrogenation of (R)-pantolactone-methyl benzoate 1e
When (R)-pantolactone is bonded to the o-toluic acid, the stereogenic center on the auxiliary is
separated from the aromatic ring by an ester group instead of an amide group in the case of
proline methyl ester or prolinol modi®ed substrates. Hydrogenation of 1e led to the simultaneous
formation of mainly the cis-isomers, the trans-isomers and the semi-hydrogenated intermediate.
The identity of the products was established by GC±MS and by synthesis of the authentic samples.
However, determination of the absolute con®gurations was not performed.
In the absence of amine, the initial reaction rates varied greatly, depending on the nature of the
metal and the support. The catalyst activity followed the same order as that observed during
the hydrogenation of 1a,2 Ru/C being six times more active than Rh/C. A higher adsorption of
the carboxylate functional group of pantolactone on oxophilic ruthenium may explain the higher
activity of the ruthenium catalyst. The e€ect was lower on Rh/Al2O3 containing residual oxidized
rhodium species. One can also note that the initial reaction rates were higher than in the case of
1a or b, probably due to the presence of four oxygens which enhance the interactions between the
substrate and the metal. Fig. 2 shows the composition of the reaction medium as a function of
time for a reaction performed on Rh/Al2O3.
Figure 2. Distribution of products as a function of time for hydrogenation of 1e on Rh/Al2O3. Reaction conditions: 2.04
mmol substrate, 0.05 mmol Rh, 130 ml ethanol, 5 MPa hydrogen. * 1e, ~ 2, ^ cis 1, ^ cis 2, ! trans 1, ! trans 2
With this chiral auxiliary, the amount of the trans-derivatives was signi®cant and was in the
range of 8±12% at the end of the reaction. In contrast with other substrates, the semi-hydro-
genated compound 2 formed transiently (5±10%) did not accumulate and was hydrogenated

M. Besson et al. / Tetrahedron: Asymmetry 11 (2000) 1809±1818
1815
during the conversion of 1e. Concerning the diastereoselectivity in the cis-isomers, Rh/C is poorly
selective (5% d.e.), while slightly higher values were obtained on the other catalysts (ca. 20%).
The aromatic ring is bound to the pantolactone auxiliary via an ester function which is less rigid
than the amide present in the proline derivatives, and this could explain the low diastereoselectivities.
Previous experiments on 1a revealed that better diastereoselectivity could be achieved if the
Rh/Al2O3 catalyst was pre-treated at 100ꢀC or 300ꢀC under hydrogen ¯ow.13 However, using
these pre-treated catalysts in the reduction of 1e did not improve the diastereoselectivity of the
reaction (de=20±23%). Finally, amine addition decreased the reaction rate of the hydrogenation
of 1e, as already observed in the case of 1a or b. However, no improvement of the diastereo-
selectivity could be achieved.
2.5. Hydrogenation of (S)-N-(2-methylbenzoyl)-2-pyrrolidinylmethylpyrrolidine 1f
The structure of this chiral auxiliary is close to that of prolinol, where the hydroxyl group is
replaced by a second pyrrolidine group. Whatever the rhodium or ruthenium catalyst, no
hydrogenation products were detected, even when the reaction was conducted at 90ꢀC. This was
attributed to the too high steric hindrance of the auxiliary which inhibited the adsorption of the
substrate on the metallic surface.
3. Conclusion
The diastereoselective hydrogenation of o-toluic acid over Rh and Ru supported catalysts was
studied using di€erent chiral auxiliaries. Though the use of new chiral auxiliaries did not provide
as high a diastereoselectivity as previously obtained with pyroglutamic acid esters, some observations
can be summarized. The steric hindrance induced by the bonding of the di€erent chiral auxiliaries
facilitates the adsorption of one face of the aromatic substrate undergoing hydrogenation and
results in some stereoinduction. The importance of the nature of the chiral moiety on diastereo-
selective induction is con®rmed. Pyroglutamic acid esters are the most e€ective chiral auxiliaries
in this type of asymmetric hydrogenation, probably due to the ecient steric and electronic e€ects
introduced by the CˆO group. The use of prolinol or pantolactone as chiral auxiliaries for o-toluic
acid hydrogenation led to the formation of higher amounts of trans-derivatives compared to the
proline ester moiety (10±12% compared to less than 3%). The substitution of the methyl ester on
the proline by a CH2OH group resulted in a decrease in the reaction rate, but with little in¯uence
on the diastereoselectivity. (S)-(2-Pyrrolidinylmethyl)pyrrolidine, containing a second nitrogen
atom, was a too bulky ligand and the substrate could not be converted. The experiments performed
with (S)-N-(2-methoxybenzoyl)prolinol yielded the cis-cyclohexyl compounds with 33% d.e. indicating
that, in addition to steric e€ects of the substituent on the aromatic ring, this functional group can
further interact with the metallic surface. However, on these catalysts, hydrogenation was
accompanied by hydrogenolysis, which occurred extensively (52%) on Rh/C.
4. Experimental
Compounds 2e and 5, and the mixture of isomers (3e+30e+4e=40e) were synthesized to
authenticate the products formed during hydrogenation.

1816
M. Besson et al. / Tetrahedron: Asymmetry 11 (2000) 1809±1818
4.1. (S)-N-(2-Methylbenzoyl)-prolinol 1c
To a stirred solution of (S)-(+)-prolinol (2 g) in chloroform (15 ml) under argon at 0ꢀC were
added successively dropwise triethylamine (5.5 ml, 2 equiv.) and o-toluoyl chloride (2.8 ml, 1.1
equiv.) and stirring continued overnight at room temperature. The mixture was treated with
saturated NaHCO3 (2Â20 ml), NaCl (20 ml), dried (MgSO4) and concentrated to give 3.5 g of a
slightly yellow solid (81%), which was used for hydrogenation without further puri®cation.
1H NMR (CDCl3, 200 MHz): ꢀ (ppm) 7.0±7.2 (m, 4H); 5.1 (s, 1H); 4.3±4.5 (m, 1H); 3.8 (d, 5.5
Hz, 2H); 3.2 (t, 6 Hz, 2H); 2.3 (s, 3H); 1.7±2.2 (m, 4H); 13C NMR (CDCl3, 25 MHz): ꢀ (ppm)
171.89 (C); 137.06 (C); 133.32 (C); 130.27±128.80±125.75±125.14 (4 CH); 66.32 (CH2); 60.57
(CH); 49.58 (CH2); 28.23 (CH2); 24.30 (CH2); 18.61 (CH3). IR ®lm (cm^1): 3352; 3050; 2947±
24
D
2877; 1616; 1447; 1251±1176; MS: m/z 77, 90 (100%), 119, 188, 201, 219 (M); mp=76ꢀC, ‰ꢁŠ
^81.3 (c=1; CHCl3). Anal. calcd for C13H14NO2: C, 71.2; H, 7.8; N, 6.4. Found: C, 71.6; H, 7.6;
N, 5.8.
4.2. (S)-N-Cyclohexane carbonyl prolinol 5
To a stirred solution of (S)-prolinol (1.3 g) in chloroform (15 ml) under argon at 0ꢀC were
added successively dropwise triethylamine (3.3 ml, 2 equiv.) and cyclohexane carboxylic acid
chloride (1.7 g). The mixture was treated with saturated NaHCO3, NaCl, dried (MgSO4) and
concentrated to give 1.6 g of a yellow solid (66%).
1H NMR (CDCl3, 200 MHz): ꢀ (ppm) 5.3 (m, 1H); 3.9±4.3 (m, 1H); 3.3±3.7 (m, 2H); 1.0±2.5
(14H); 13C NMR (CDCl3, 25 MHz): ꢀ (ppm) 178.0 (C); 67.83 (CH2); 61.28 (CH); 47.99 (CH2);
43.31 (CH); 29.55±28.78±28.47±26.04±25.70 (5 CH2); 24.73 (CH2). IR ®lm (cm^1): 3393; 2935±
2872; 1611; 1437; MS: m/z 41, 55, 70 (100%), 83, 180, 193, 212 (M); mp=76ꢀC.
4.3. (R)-Pantolactone-o-toluoate 1e
To a stirred solution of (R)-pantolactone (12 g) in chloroform (60 ml) under argon at 0ꢀC were
added successively dropwise triethylamine (14.3 ml, 2 equiv.) and o-toluoyl chloride (13.2 ml, 1.1
equiv.) and stirring continued overnight under re¯ux. The solution was cooled to room tem-
perature and the organic layer was treated with saturated NaHCO3 (2Â20 ml), NaCl (20 ml),
dried (MgSO4) and concentrated to give a slightly yellow solid. The product was crystallized in
cyclohexane to give 20 g of white needles (87%).
1H NMR (CDCl3, 100 MHz): ꢀ (ppm) 8.0 (m, 1H); 7.2±7.4 (m, 3H); 5.6 (s, 1H); 4.1 (s, 2H); 2.6
(s, 3H); 1.2±1.3 (2s, 6H); 13C NMR (CDCl3, 25 MHz): ꢀ (ppm) 172.87 (C); 166.27 (C); 141.28 (C);
133.13±132.18±131.34 (3 CH); 128.27 (C); 126.20 (CH); 76.55 (CH); 75.63 (CH2); 40.71 (C); 22.23
(CH3); 23.35±20.47 (2 CH3). IR ®lm (cm^1): 3050; 2978±2958±2939; 1733±1740; 1256±1141±1099;
24
D
MS: m/z 77, 83, 90 (100%), 113, 119, 135, 248 (M); mp=65ꢀC, ‰ꢁŠ 0.6 (c=1; CHCl3). Anal.
calcd for C14H16O4: C, 67.7; H, 6.5. Found: C, 67.9; H, 6.5.
4.4. cis/trans (R)-N-(2-Methyl-cyclohexane carbonyl)-pantolactone (3e+3e0+4e+40e)
To a solution of commercial mixture (cis/trans) of 2-methyl-cyclohexane carboxylic acid (2 g,
14 mmol) in CHCl3 (10 ml) was added dropwise SOCl2 (4 ml, 4 equiv.) at 0ꢀC under argon, the
stirring was maintained overnight at room temperature and the solvents were removed. The yellow

M. Besson et al. / Tetrahedron: Asymmetry 11 (2000) 1809±1818
1817
oil was used without puri®cation in the following step. In a three-neck round-bottomed ¯ask was
placed (R)-pantolactone (1.66 g, 12.7 mmol) in CHCl3 (30 ml). Triethylamine (1.5 ml, 1.1 equiv.)
followed by crude acid chloride (2.3 g) were added dropwise and the mixture was stirred over-
night under re¯ux. The solution was washed with NaHCO3 (2Â10 ml) then NaCl (2Â10 ml), dried
(MgSO4) and concentrated under vacuum to give a brown oil. Puri®cation on silica gave 540 mg
of a white solid (17%), which was a mixture of the four stereoisomers.
Selected data of the main compound: 1H NMR (CDCl3, 100 MHz): ꢀ (ppm) 5.35 (s, 1H); 4.0 (s,
2H); 2.6±2.7 (m, 1H); 2.0±2.4 (s, 1H); 0.9±1.8 (m, 8H); 1.1 (s, 6H); 0.93 (d, 4.3 Hz, 3H); 13C NMR
(CDCl3, 25 MHz): ꢀ (ppm) 174.03±172.72 (2C); 76.38 (CH2); 74.68 (CH); 46.11 (CH); 40.39 (C);
31.87 (CH2); 31.42 (CH); 24.53 (CH2); 23.28±20.27 (2 CH3); 21.67±21.62 (2 CH2); 15.56 (CH3);
MS: m/z 96 (100%), 125, 131, 157, 172, 186, 254 (M).
4.5. (R)-N-(2-Methyl-1-cyclohex-2-ene carbonyl)-pantolactone 2e
To a solution of 2-methyl-1-cyclohex-2-ene carboxylic acid4 (100 mg, 7 mmol) in CHCl3 (2 ml)
was added SOCl2 (0.2 ml, 4 equiv.) at 0ꢀC under argon. The solution was stirred overnight and
the solvents were removed under vacuum. The crude product was used in the following step
without puri®cation. In a three-neck round-bottomed ¯ask was placed (R)-pantolactone (246 mg,
1.89 mmol) in CH2Cl2 (6 ml). Triethylamine (0.3 ml, 3 equiv.) followed by the crude acid chloride
(300 mg) were added and the mixture was stirred for one night at 60ꢀC. The solution was washed
with NaHCO3 (2Â10 ml) then NaCl (2Â20 ml), dried (MgSO4) and concentrated under vacuum
to yield 300 mg of a slightly brown oil (63%).
1H NMR (CDCl3, 100 MHz): ꢀ (ppm) 5.4 (s, 1H); 4.0 (s, 2H); 2±2.3 (m, 4H); 1.5±1.7 (m, 4H);
1.2 (s, 3H); 13C NMR (CDCl3, 25 MHz): ꢀ (ppm) 173.16 (C); 167.29 (C); 150.70 (C); 123.08 (C);
76.52 (CH2); 74.80 (CH); 40.50 (C); 34.54±26.47±22.50±22.39 (4 CH2); 22.70 (CH3). IR ®lm (cm^1):
2935±2863±2824; 1792±1721; 1634; 1224±1203±1097; MS: m/z 94 (100%), 99, 113, 122, 139, 163,
22
194, 252 (M); ‰ꢁŠ ^3.9 (c=0.51; CHCl3).
D
4.6. (S)-N-(2-Methylbenzoyl)-(2-pyrrolidinylmethyl)-pyrrolidine 1f
To a stirred solution of (S)-(pyrrolidinemethyl)pyrrolidine (3 g) in chloroform (30 ml) under
argon at 0ꢀC were added successively dropwise triethylamine (3 ml, 1.1 equiv.) and o-toluoyl
chloride (3.3 g, 1.1 equiv.) and stirring continued overnight at room temperature. The mixture
was treated with saturated NaHCO3 (2Â20 ml), NaCl (20 ml), dried (MgSO4) and concentrated
to give a brown oil which was distilled under vacuum to give 3.8 g of a slightly yellow oil (65%).
1H NMR (CDCl3, 100 MHz): ꢀ (ppm) 7.1±7.3 (m, 4H); 4.3 (m, 1H); 3.47 (m, 2H); 2.8 (m, 2H);
2.5 (m, 4H); 2.1 (s, 3H); 1.4±2.0 (m, 8H); 13C NMR (CDCl3, 25 MHz): ꢀ (ppm) 169.33 (C); 137.30
(C); 133.13 (C); 129.76±128.11±125.22±124.97 (4 CH); 57.13 (CH2); 55.05 (CH); 53.78 (CH2);
48.03±44.51 (2 CH2); 28.48±23.61±23.00±22.76 (4 CH2); 18.30 (CH3). IR ®lm (cm^1): 3050; 2966;
2876; 2788; 1783; 1723; 1632; 1411; 1256; MS: m/z 84 (100%), 91, 110, 119, 137, 188, 202, 272
(M); bp=160ꢀC/2.7Â10^2 mbar.
4.7. Catalysts
The catalysts used in hydrogenation were 3.6% Rh/C (Aldrich 20,616±4), 3.7% Rh/Al2O3
(Aldrich 37,971±9) and 5% Ru/C (Aldrich 28,147±6). High-resolution electron microscopy

1818
M. Besson et al. / Tetrahedron: Asymmetry 11 (2000) 1809±1818
showed that most of the rhodium and ruthenium particles in the catalysts were in the size range
1±4 nm and homogeneously distributed inside the grains. The ruthenium catalysts were pre-treated
at 300ꢀC under hydrogen atmosphere and transferred in the reactor without exposure to air.
4.8. Hydrogenation experiments
The hydrogenation of the substrates was carried out in a stainless steel autoclave equipped with
a magnetically driven turbine stirrer under 50 MPa and at room temperature. Standard experi-
ments used 2.25 mmol of substrate dissolved in 130 ml ethanol. In order to improve the selectiv-
ity, some experiments were carried out with the addition of EDCA (ethyldicyclohexylamine). The
conversions and selectivities were determined from gas chromatography analyses (GC) which
were performed using a Shimadzu GC14A apparatus using a J&W DB1701 column. The
products of the reaction were identi®ed by GC±MS analysis using a Fisons GC 8000 apparatus
and by synthesis of the references. The absolute con®guration of the diastereoisomers was not
established.
References
1. The Merck Index, 12th, Merck Research Lab., Eds. Div. of Merck and Co, Inc., 1996.
2. Besson, M.; Blanc, B.; Champelet, M.; Gallezot, P.; Nasar, K.; Pinel, C. J. Catal. 1997, 170, 254±264.
3. Besson, M.; Gallezot, P.; Neto, S.; Pinel, C. Chem. Commun. 1998, 1431±1432.
4. Besson, M.; Delbecq, F.; Gallezot, P.; Neto, S.; Pinel, C. ChemistryÐA European Journal 2000, 6, 949±958.
5. Exl, C.; Ferstl, E.; Konig H.; Rogi-Kohlenprath, R. Chirality 1995, 7, 211±218.
6. Ranade, V. S.; Consoglio, G.; Prins, R. Catal. Lett. 1999, 58, 71±74.
7. Ranade, V. S.; Prins, R. J. Catal. 1999, 185, 479±486.
8. Besson, M. C.; Pinel, C. Topics in Catalysis 1998, 5, 25±38.
Á
Â
9. Seyden-Penne, J. In Synthese et Catalyse Asymetriques; Intereditions/CNRS Editions: Paris, 1994.
10. Rylander, P. In Catalytic Hydrogenation in Organic Synthesis; Academic Press: New York, 1979; pp. 205±207.
11. Augustine, R. L. In Heterogeneous Catalysis for the Synthetic Chemist; Marcel Dekker: New York, 1996.
12. Ranade, V. S.; Consiglio, G.; Prins, R. J. Org. Chem. 1999, 64, 8862±8867.
13. Besson, M.; Gallezot, P.; Neto, S.; Pinel, C. Studies in Surface Science and Catalysis: Heterogeneous Catalysis and
Fine Chemicals IV; Blaser, H. U.; Baiker, A.; Prins, R., Eds.; Elsevier: Amsterdam, 1997; Vol. 108, pp. 215±222.