Phosphines versus phosphinites as ligands in the rhodium catalyzed asymmetric hydrogenation of imines: A systematic study

Article abstract of DOI:10.1016/S0957-4166(99)00431-0

The asymmetric hydrogenation of N-(1-phenylethylidene)benzylamine with a range of rhodium(I)-diphosphine and diphosphinite catalysts is studied. The reaction is strongly sensitive to the size of the metal chelate. Complexes based on five- and six-membered chelates or electron-rich alkylphosphines gave poor or moderate conversions. The reactivity of diphosphine catalysts could be increased by the addition of p-toluenesulfonic acid. Unexpectedly, Rh-complexes based on chiral diphosphinites and a diphosphite also rapidly converted the substrate to the desired amine. Highest efficiency was observed with a rhodium(I) complex with (R,R)-1,2-cyclohexanol-bisdiphenylphosphinite [(R,R)-bdpch] as chiral ligand. Without any additive complete hydrogenation of the imine was achieved within 5 h. The product was produced in an enantioselectivity of 71%.

Full text of DOI:10.1016/S0957-4166(99)00431-0

Pergamon
Tetrahedron: Asymmetry 10 (1999) 4009–4015
Phosphines versus phosphinites as ligands in the rhodium
catalyzed asymmetric hydrogenation of imines: a systematic
study
a,b
a
c
a
a,∗
Vitali I. Tararov, Renat Kadyrov, Thomas H. Riermeier, Jens Holz and Armin Börner
Institut für Organische Katalyseforschung an der Universität Rostock e.V., Buchbinderstr. 5/6, D-18055 Rostock, Germany
A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, Vavilova 28, 117813 Moscow,
a
b
Russian Federation
c
Aventis Research and Technologies GmbH Industriepark Höchst, G 830, D-65926 Frankfurt/Main, Germany
Received 3 September 1999; accepted 28 September 1999
Abstract
The asymmetric hydrogenation of N-(1-phenylethylidene)benzylamine with a range of rhodium(I)-diphosphine
and diphosphinite catalysts is studied. The reaction is strongly sensitive to the size of the metal chelate. Complexes
based on five- and six-membered chelates or electron-rich alkylphosphines gave poor or moderate conversions.
The reactivity of diphosphine catalysts could be increased by the addition of p-toluenesulfonic acid. Unexpec-
tedly, Rh-complexes based on chiral diphosphinites and a diphosphite also rapidly converted the substrate to
the desired amine. Highest efficiency was observed with a rhodium(I) complex with (R,R)-1,2-cyclohexanol-
bisdiphenylphosphinite [(R,R)-bdpch] as chiral ligand. Without any additive complete hydrogenation of the imine
was achieved within 5 h. The product was produced in an enantioselectivity of 71%. © 1999 Elsevier Science Ltd.
All rights reserved.
1. Introduction
In contrast to the high enantioselectivities reported in the hydrogenation of olefins and ketones,
1
only limited success has been achieved in the catalytic asymmetric hydrogenation of prochiral imines
2
although the reaction is of considerable industrial interest. Up to now a range of Rh-, Ir- and Ru-
3,4
complexes have been investigated in detail. As ligands chiral diphosphines were utilized preferentially.
Interestingly, most of those phosphines which are reputed to induce high enantioselectivity in the
Rh(I) or Ru(II) catalyzed hydrogenation of other prochiral substrates failed or gave poor results in
3,5
this particular reaction. In some instances Ir-catalysts were found to be superior. Unfortunately, the
∗
Corresponding author. E-mail: armin.boerner@ifok.uni-rostock.de
0957-4166/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00431-0

4010
V. I. Tararov et al. / Tetrahedron: Asymmetry 10 (1999) 4009–4015
6
7
5d
presence of additives like amines, imides or iodide in the solution is frequently required to avoid
deactivation of the catalyst and to achieve satisfactory yields and/or elevated degrees of enantioselectivity.
8
As an alternative the reduction of N-acylhydrazones was suggested. However, this approach requires
additional reaction steps and therefore a straightfoward method is highly desired.
It is interesting to note that besides diphosphines other chelating P-ligands were rarely employed. Only
9
recently has the application of a tridentate diphosphinite been reported. Amidophosphine–phosphinites
were suggested as ligands for the hydrogenation of cyclic iminium salts.10
Herein, we report on a systematic study on the asymmetric hydrogenation of imines based on Rh-
diphosphine, diphosphinite and diphosphite catalysts under varying conditions. As a test substrate N-(1-
phenylethylidene)benzylamine was chosen (Scheme 1).
Scheme 1.
2. Results and discussion
2.1. Nonasymmetric hydrogenation
In order to identify main requisites and conditions for the achievement of high conversion in the
hydrogenation of the imine, first achiral precatalysts with chelating diphosphines and diphosphin-
ites, respectively, were tested. The reaction was carried out at an initial hydrogen pressure of ca.
50 bar in methanol at room temperature. As ligands 1,2-bis(diphenylphosphino)ethane (1=dppe), 1,3-
bis(diphenylphosphino)propane (2=dppp), 1,4-bis(diphenylphosphino)butane (3=dppb), the correspon-
0
ding cyclohexyl-substituted diphosphines 4 and 5, the 1,1 -ferrocenyldiphosphines (Ph PCp) Fe 6 and
2
2
11
(
i-Pr PCp) Fe 7, respectively, and 1,2-ethandiol-bis-diphenylphosphinite (8=DPOE) were applied. The
2 2
results are summarized in Table 1.
As clearly illustrated five- and six-membered Rh(I)-chelates afforded the desired amine in poor or
modest yield (runs 1, 3). In contrast the more flexible seven-membered Rh(I)-chelate based on dppb as
ligand converted the imine within 2 h into the desired product (run 4). Surprisingly, catalysts based on
dialkylphosphines had inferior performance compared to those with diarylphosphines as ligands (runs 5,
10). It is remarkable that the addition of TsOH increased the rate of the reaction (runs 2, 7). Obviously,
the hydrogenation of imines is not subject to the same effects as the reduction of ketones, where electron-
rich phosphine ligands have been shown to be advantageous.12 This conclusion is supported by the result
obtained with the Rh(I)-complex of an electron-deficient ligand, such as the diphosphinite DPOE (run
11). This catalyst hydrogenated the imine rather effectively.
2.2. Asymmetric hydrogenation
Inspired by the results listed in Table 1, we next turned our attention to the asymmetric hy-
drogenation of N-(1-phenylethylidene)benzylamine with 7-membered Rh(I)-chelates of chiral bis-
1
3
14
15
diphenylphosphines such as (S,S)-bppm and (R,R)-DIOP (Table 2, runs 1, 3). Both catalysts
16
afforded the amine in only moderate yields. The enantiomeric excess observed was low. Interestingly
the application of the neutral precatalysts significantly increased the efficiency (runs 2, 4).

V. I. Tararov et al. / Tetrahedron: Asymmetry 10 (1999) 4009–4015
4011
Table 1
Hydrogenation of N-(1-phenylethylidene)benzylamine with achiral rhodium catalystsa
By modifications of the structure of DIOP the catalytic properties of the parent catalyst were significan-
tly changed. Thus, the introduction of a remote HO-group in the backbone of the DIOP-moiety (ligands
1
7
1
1
1 and 12) decreased the ee (runs 5, 6). It is noteworthy that the HO-groups in the catalyst with ligand
3 did not seriously diminish the yield (run 7) as observed in the hydrogenation of bidentate olefins.18
A superior reaction rate was observed when the benzoate group in ligand 14 (run 8) was replaced by the
19
sulfonatobenzoate moiety (ligand 15, run 9). This result corresponds well to observations made in the
aqueous two phase hydrogenation of imines with Rh(I)(mono-sulfonated bdpp) catalysts20 and to results
observed in the asymmetric hydrogenation of prochiral olefins with catalysts bearing chiral phosphines
19b
with low degrees of sulfonation.
As seen in Table 2 ‘active’ functional groups, e.g. the sulfonate group in the backbone of the catalyst,
can significantly improve the performance of the catalyst. Therefore, we investigated the influence of
different additives on the reaction. In Table 3 the effects of acids and bases upon the imine hydrogenation
with (R,R)-DIOP-complexes are summarized.
In comparison to the blank experiment (run 1) by addition of increasing amounts of p-toluenesulfonic
acid the degree of the conversion was improved (runs 2–5). Simultaneously, the formation of the (S)-
product was favored. This effect is likely to be due to the coordination of the tosylate anion (formed
by reaction of TsOH with the imine) on the rhodium center. Proof for this hypothesis came from run 6,
−
−
where a decrease of the ee was observed after replacement of the weakly coordinated BF4 by TsO in

4012
V. I. Tararov et al. / Tetrahedron: Asymmetry 10 (1999) 4009–4015
Table 2
Hydrogenation with Rh(I) catalysts based on chiral diphosphines as ligandsa
Table 3
a
Influence of additives on the efficiency of different catalysts with 10 [(R,R)-DIOP] as ligand

V. I. Tararov et al. / Tetrahedron: Asymmetry 10 (1999) 4009–4015
4013
the precatalyst. The addition of TsOH increased the tendency for the formation of the (S)-product (run
7). It is noteworthy that the presence of iodide or t-BuOK diminished the rate of the hydrogenation (runs
8–11). In the presence of a 12-fold amount of the alcoholate 41% ee of the (R)-product was obtained (run
10).
As described above, P-ligands like diphosphinites with diminished p-accepting properties can ad-
vantageously support the hydrogenation of our test imine. Thus, as listed in Table 4, by application of
Rh(I)-complexes of phosphinites like Ph-β-glup21 16, K −OH 17 or Propraphos 18 full conversion
+
22
23
β
of the substrate was achieved, although the enantioselectivities obtained were disappointingly low (runs
–3). Noteworthy is the high conversion observed with the catalyst based on the diphosphite 19 (run
). The best result afforded the cationic BF4 complex with (R,R)-bdpch 20 as ligand (run 5). The
24
1
−
25
4
desired amine was obtained in 71% ee. Inferior results were observed with the corresponding Rh(acac)
complex (run 6) and by application of the neutral chloro-precatalyst (run 7). As illustrated in Table 3,
additives in the reaction mixture may have a beneficial effect on the asymmetric hydrogenation of N-(1-
phenylethylidene)benzylamine with Rh-diphosphine catalyst. In contrast to these observations the Rh-
bdpch catalyst showed a different behavior. Thus, the addition of an excess of TsOH diminished the rate
of the hydrogenation (runs 8–10). However, the ee was not affected significantly. It is noteworthy that
the reaction is quite insensitive to an excess of benzyl amine (run 11). On the contrary, the addition of
t-BuOK or Bu NI diminished the overall efficiency of the catalyst (runs 12, 13). The decelerating effect
4
was more pronounced with the cationic than with the neutral complex (run 14). In nonpolar solvents like
toluene or methylene chloride the reaction was also less efficient (runs 15–17).
It is known that ees obtained in the asymmetric hydrogenation of N-(1-phenylethylidene)benzylamine
26
are sometimes poorly reproducible. We have faced the same problem with two freshly prepared batches
of the substrate. Thus, with [Rh(20)COD]BF as the precatalyst instead of 71% ee only 50 and 63%
4
ees were observed. Redistillation of the solvent MeOH, repeated recrystallizations of the substrate and
addition of water or acetophenone (fivefold molar excess to the precatalyst) did not improve the ee.
With these two batches the presence of BnNH gave 64 and 72% ee, respectively. Finally, we found
2
that distillation of the substrate in vacuo was important to reproduce the enantioselectivity of 71%
given in Table 4. It is noteworthy that in the hydrogenation of the pretreated substrate the addition of
BnNH had no effect on the enantioselectivity. The ee value of the product in the range of 70–73% was
2
reproducible and did not depend on the pressure (50 bar and 100 bar), the temperature (15 and 25°C) and
the concentration of the catalyst (1.0 and 2.5 mmol/L) as well as on the concentration of the substrate
(
0.5 and 0.1 mol/L). It was also found that at a molar substrate:[Rh(bdpch)COD]BF ratio of 500:1
4
more than 98% of the substrate was converted into the product within 5 h at room temperature. The
[
Rh(20)NBD]BF complex was as efficient as the [Rh(20)COD]BF precatalyst.
4
4
3. Conclusion
Several achiral and chiral diphosphine and diphosphinite Rh(I) complexes have been tested in the
hydrogenation of N-(1-phenylethylidene)benzylamine. The reaction with diphosphines as ligand was
strongly sensitive to the size of the chelate ring formed with the metal and the substituents on the phos-
phorus. In the presence of additives like TsOH the rate of the reaction could be increased. Surprisingly,
rhodium complexes of chiral diphosphinites showed an excellent reactivity. This is in contrast to results
obtained with the achiral ligands, where a 1,4-diphosphine catalyst was revealed to be more active than its
1,2-diphosphinite counterpart. With a cationic Rh(I)–(R,R)-bdpch complex up to 71% ee was achieved. It
is noteworthy that this result was obtained without any additive und under rather smooth conditions. Our

4014
V. I. Tararov et al. / Tetrahedron: Asymmetry 10 (1999) 4009–4015
Table 4
a
Hydrogenation with chiral Rh(I) catalysts based on electron poor P-ligands 16–20:
preliminary results make diphosphinite Rh(I)-complexes27 attractive catalysts for further investigations
with the potiential for additional synthetic power. Currently, work is in progress to test other imines in
this reaction.
Acknowledgements
We are grateful for the financial support provided by Aventis Research and Technologies GmbH
(Frankfurt, Germany), the European Commision (Inco-Copernicus, ERBIC 15 CT 960722) and the Fonds
der Chemischen Industrie. We thank Dr. Ch. Fischer and Mrs. A. Modler for HPLC-analysis of the
hydrogenation product. We thank Prof. Dr. R. Selke, Mrs. U. Schmidt and Dr. U. Berens for generous
gifts of ligands. It’s a pleasure to thank Dr. habil. D. Heller for helpful discussion and support of this
work.

V. I. Tararov et al. / Tetrahedron: Asymmetry 10 (1999) 4009–4015
4015
References
1
2
3
. (a) Noyori, R. Asymmetric Catalysis in Organic Synthesis Wiley: New York, 1994; p. 82. (b) Ojima, I. Catalytic
Asymmetric Synthesis; VCH: New York, 1993; p. 31.
. Spindler, F.; Blaser, H.-U. Enantioselective reduction of C_N bonds and enamines with hydrogen. In Transition Metals
for Organic Synthesis; Beller, M.; Bolm, C., Eds.; Wiley-VCH: Weinheim, 1998; Vol. 2, p. 69.
. For recent publications and reviews, see: (a) Kobayashi, S.; Ishitani, H. Chem. Rev. 1999, 99, 1069. (b) Cahill, J. P.;
Lightfoot, A. P.; Goddard, R.; Rust, J.; Guiry, P. J. Tetrahedron: Asymmetry 1998, 9, 4307. (c) Schnider, P.; Koch, G.;
Prétôt, R.; Wang, G.; Bohnen, F. M.; Krüger, C.; Pfaltz, A. Chem. Eur. J. 1997, 3, 887.
4
. Chiral biphenyl-bridged titanocene complexes were also suggested affording good to excellent enantioselectivity.
However, the conditions applied were rather severe and a relatively large amount of catalyst had to be utilized: (a)
Willoughby, C. A.; Buchwald, S. L. J. Am. Chem. Soc. 1992, 114, 7562. (b) Ringwald, M.; Stürmer, R.; Brintzinger,
H. H. J. Am. Chem. Soc. 1999, 121, 1524.
5
. (a) Ng Cheon Chan, Y. P.; Osborn, J. A. J. Am. Chem. Soc. 1990, 112, 940. (b) Sablong, R.; Osborn, J. A. Tetrahedron:
Asymmetry 1996, 7, 3059. (c) Tani, K.; Onouchi, J.-i.; Yamagata, T.; Kataoka, Y. Chem. Lett. 1995, 955. (d) Spindler, F.;
Pugin, B.; Blaser, H.-U. Angew. Chem. 1990, 102, 561; Angew. Chem., Int. Ed. Engl. 1990, 29, 558.
. Tani, K.; Onouchi, J.-i.; Yamagata, T.; Kataoka, Y. Chem. Lett. 1995, 955.
6
7
. (a) Morimoto, T.; Achiwa, K. Tetrahedron: Asymmetry 1995, 6, 2661. (b) Zhu, G.; Zhang, X. Tetrahedron: Asymmetry
1998, 9, 2415.
8
9
. Burk, M. J.; Feaster, J. E. J. Am. Chem. Soc. 1992, 114, 6266.
. Sablong, R.; Osborn, J. A. Tetrahedron Lett. 1997, 37, 4937.
1
1
0. Broger, E. A.; Burkart, W.; Hennig, M.; Scalone, M.; Schmid, R. Tetrahedron: Asymmetry 1998, 9, 4043.
1. For the poor catalytic performance of a cationic-(i-Pr PCp) Fe Rh complex in the hydrogenation of a related imine, see:
Cullen, W. R.; Fryzuk, M. D.; James, B. R.; Kutney, J. P.; Kang, G.-J.; Herb, G.; Thorburn, I. S.; Spogliarich, J. Mol. Catal.
990, 62, 243.
2
2
1
1
1
1
1
1
2. Inoguchi, K.; Sakuraba, S.; Achiwa, K. Synlett 1992, 169.
3. Achiwa, K. J. Am. Chem. Soc. 1976, 98, 8265.
4. Kagan, H. B.; Dang, T. P. J. Am. Chem. Soc. 1972, 94, 6429.
5. Ru(R-BINAP)-complexes gave low conversions and poor enantioselectivity.
6. For the poor catalytic performance of a neutral DIOP–Rh complex in the asymmetric hydrogenation of a related imine,
see: Ref. 11.
1
7. (a) Börner, A.; Holz, J.; Kless, A.; Heller, D.; Berens, U. Tetrahedron Lett. 1994, 35, 6071. (b) Holz, J.; Börner, A.; Kless,
A.; Borns, S.; Trinkhaus, S.; Selke, R.; Heller, D. Tetrahedron: Asymmetry 1995, 6, 1973.
8. (a) Borns, S.; Kadyrov, R.; Heller, D.; Baumann, W.; Spannenberg, A.; Kempe, R.; Holz, J.; Börner, A. Eur. J. Inorg.
Chem. 1998, 1291. (b) Borns, S.; Kadyrov, R.; Heller, D.; Baumann, W.; Holz, J.; Börner, A. Tetrahedron: Asymmetry
1
1999, 10, 1425.
19. (a) Trinkhaus, S.; Holz, J.; Selke, R.; Börner, A. Tetrahedron Lett. 1997, 38, 807. (b) Trinkhaus, S.; Kadyrov, R.; Selke,
R.; Holz, J.; Götze, L.; Börner, A. J. Mol. Catal. A: Chemical 1999, 144, 15.
20. (a) Bakos, J.; Orosz, A.; Heil, B.; Laghmari, M.; Lhoste, P.; Sinou, D. J. Chem. Soc., Chem. Commun. 1991, 1684. (b)
Lensink, C.; de Vries, J. G. Tetrahedron: Asymmetry 1992, 3, 235.
21. Selke, R.; Pracejus, H. J. Mol. Catal. 1986, 37, 213.
2
2. (a) Kumar, A.; Oehme, G.; Roque, J. P.; Schwarze, M.; Selke, R. Angew. Chem. 1994, 106, 2272; Angew. Chem., Int. Ed.
Engl. 1994, 33, 2197. (b) Selke, R.; Ohff, M.; Riepe, A. Tetrahedron 1996, 52, 15079.
3. (a) Krause, H. W.; Foken, H.; Pracejus, H. New J. Chem. 1989, 13, 615. (b) Krause, H. W.; Schmidt, U.; Taudien, S.;
Costisella, B.; Michalik, M. J. Mol. Catal. A: Chemical 1995, 104, 147.
2
2
2
2
2
4. Kadyrov, R.; Heller, D.; Selke, R. Tetrahedron: Asymmetry 1998, 9, 329.
5. Tanaka, M.; Ogata, I. J. Chem. Soc., Chem. Commun. 1975, 735.
6. Vastag, S.; Bakos, J.; Torös, S.; Takach, N. E.; King, R. B.; Heil, B.; Marko, L. J. Mol. Catal. 1984, 22, 283.
7. Corresponding cationic Ir(I)-complexes with DIOP or bdpch as a ligand gave inferior results. [Ir(DIOP)COD]BF
conversion, % ee not analyzed; [Ir(DIOP)Cl] (in situ): 63% conversion, 3% ee (S); [Ir(bdpch)COD]BF : 4% conversion,
ee not analyzed; [Ir(bdpch)Cl] (in situ): 57% conversion, 4% ee (S).
4
: 3%
2
4
%
2