Author Topic: Leuckart-Wallach Reductive Amination @ 50-70°C  (Read 1274 times)

0 Members and 1 Guest are viewing this topic.


  • Guest
Leuckart-Wallach Reductive Amination @ 50-70°C
« on: November 22, 2002, 10:36:00 PM »

A Cp*Rh(III) complex catalyzes reductive amination of ketones using HCOONH4 at 50-70°C to give the corresponding primary amines in high yields. The reaction is clean and operationally simple and proceeds at a lower temperature and with higher chemoselectivity than the original Leuckart-Wallach reaction.

J. Org. Chem., 67 (24), 8685-8687 (2002)





  • Guest
J. Org. Chem., 2002, 67, 8685-8687
« Reply #1 on: November 21, 2003, 12:06:00 PM »


Catalytic Leuckart-Wallach-Type Reductive Amination of Ketones

Masato Kitamura, Donghyun Lee, Shinnosuke Hayashi, Shinji Tanaka, and  Masahiro Yoshimura

Research Center for Materials Science and the Department of Chemistry, Nagoya University, Chikusa, Nagoya 464-8602, Japan

Received June 3, 2002


A Cp*Rh(III) complex catalyzes reductive amination of ketones using HCOONH4 at 50-70 °C to give the corresponding primary amines in high yields. The reaction is clean and operationally simple and proceeds at a lower temperature and with higher chemoselectivity than the original Leuckart-Wallach reaction. The new method has been applied to the synthesis of alpha-amino acids directly from alpha-keto acids.

Reductive amination of carbonyl compounds is attractive in organic synthesis because ketones or aldehydes can be transformed, in one reaction vessel, directly to the corresponding secondary or primary alkylamines without isolation of the intermediary imines or hydroxy amines.1 The reaction with formic acid as a reducing agent is called the Leuckart-Wallach (LW) reaction.2 The LW reaction is very simple and clean, but it suffers from several drawbacks such as the requirement of high temperature (mostly above 180 °C), the formation of N-formyl derivative, and the difficulty of the selective synthesis of primary amine from ammonia.3 Such a reaction is in general most useful and efficient when performed catalytically, rather than stoichiometrically, but during the past 100 years, only a few reports on the catalytic version of LW reaction have been made.4 This is apparently because the reported methods using Raney Ni or Co could not overcome the above deficiencies. In this paper, we describe a new and efficient catalytic LW-type reductive amination of ketones.

The 8, 9, and 10 group metal complexes having Cp*, Cp, COD, or P(C6H5)3 ligand were selected because most of these complexes are able to hydrogenate the unsaturated organic molecules.5 The catalytic activity for the LW reaction was screened by use of 3-5 mmol of acetophenone (1a) and ammonium formate by fixing the concentrations of the complex, 1a, and ammonium formate, temperature, reaction time, and solvent to 5 mM, 1 M, 5 M, 70 °C, 2 h, and methanol, respectively. The yields of the possible products, 1-phenylethylamine (2a), di(1-phenylethyl)amine (3a), N-formyl-1-phenylethylamine (4a), and 1-phenylethanol (5a), were determined by 1H NMR analysis (delta 2.62 (s, CH3 of 1a), delta 4.18 (q, J = 6.6 Hz, CH of 2a), delta 3.59 (q, J = 6.6 Hz, CH of meso-3a), delta 3.86 (q, J = 7.4 Hz, CH of dl-3a), delta 4.69 (dq, J = 7.4, 7.4 Hz, CH of the minor rotamer of 4a), delta 5.22 (dq, J = 7.3, 7.3 Hz, CH of the major rotamer of 4a), delta 4.90 (q, J = 6.6 Hz, CH of 5a)).

Figure 1 illustrates the reactivity and selectivity of the complexes investigated. [RhCp*Cl2]2 (6)6 shows the highest efficiency among others. Under the standard conditions, 98% of acetophenone is converted to 2a, 3a, 4a, and 5a in a 96.5:0.5:1.0:2.0 ratio. The desired product 2a can be isolated in pure form in 90% yield by a simple partition between organic and aqueous layers. [Ir(cod)Cl]27 also catalyzes the LW reaction to give a 96:0:1:2 mixture of 2a, 3a, 4a, and 5a, although the reactivity is lowered. Table 1 lists the results of the optimization of the conditions using [RhCp*Cl2]2. The complete consumption of 1a takes 6 h, while the N-formyl compound 4a is formed in 7% yield (entry 2). Without any special care about moisture and air, 98% of acetophenone is converted to 2a, 3a, 4a, and 5a in a 95.5:0.8:1.3:2.4 ratio (entry 3). The amount of 4a is increased to 34% after 31 h (entry 4). At 50 °C, both the reactivity and selectivity is dramatically decreased due to the low solubility of HCOONH4 in methanol (entry 5). A 5 mol amount of HCOONH4 is essential. Lowering the concentration to 2 M, the 2a/5a ratio is decreased to 9 (entry 6). With 1 M HCOONH4, the reactivity is halved and the alcohol product 5a is produced in >50% yield (entry 7). A 10 mol amount of HCOONH4 results in completed reaction with high chemoselectivity after 1 h (entry 8). The total concentration can be reduced to 0.33 M without loss of the amine/alcohol selectivity (entry 9), but an increase to 5 M results in the insolubility of HCOONH4 (entry 10). Methanol is the solvent of choice. The reactivity is decreased in aqueous methanol, 1:1 alcohol-CF3CH2OH, and 2-propanol (entries 11-14). In aprotic solvents, the yields of the reduction products never exceed 15% (entries 15-20). In CH3CN, CH2Cl2, benzene, and cyclohexane, the alcohol 5a was obtained selectively (entries 15, 17, 19, and 20). On the other hand, the LW product 2a was predominantly produced in DMF and THF (entries 16 and 18). Use of an excess either of ammonia or formic acid decreases the reactivity (entries 21 and 22). [RhCp*I2]26 and [Rh2Cp*2Cl3]BARF8 also showed the same reactivity and selectivity as those of [RhCp*Cl2]2 (entries 23 and 24).
Figure 1 The reactivity and selectivity profiles of the 8, 9, and 10 group metal complexes in the catalytic LW reaction using acetophenone and ammonium formate: dark shading, 1-phenylethylamine (2a); lighter shading, di(1-phenylethyl)amine (3a); white, N-formyl-1-phenylethylamine (4a); striped, 1-phenylethanol (5a); light shading, others.

The reproducibility was confirmed on a 10 g scale reaction using 1a. Thus, a 1:5 mixture of 1a and HCOONH4 was reacted in methanol (83 mL) containing 257 mg of [RhCp*Cl2]2 at 70 °C for 7 h, giving 2a in 92% yield as determined by NMR analysis of the crude mixture obtained by a usual workup under basic condition.9 The pure 2a was isolated in 85% yield (see the Experimental Section). The generality is high. All of primary, secondary alkyl methyl ketones (1b and 1c) studied can be converted to the corresponding primary amines in greater than 90% yield (entries 25 and 26). The cyclic ketone (1g) remains a high reactivity, but the 2g/3g ratio is decreased to 6 (entry 30). The reactivities of pinacolone (1d) and diphenyl ketone (1e) are low, and with dicyclohexyl ketone (1f) no reaction at all occurred.

As shown in Table 2, the present catalytic LW-type reaction can be applied to the alpha-keto acids.10 When benzoylformic acid (7) was subjected to the above established conditions ([6] = 5 mM, [7] = 1 M, [HCOONH4] = 5 M, CH3OH, 50 °C), the reductive amination product was precipitated from the reaction mixture. Filtration gives pure phenylglycine in 81% isolated yield. Other keto acids possessing indole and thiophene groups (8 and 9) were also converted to the corresponding amino acids in good isolated yields. These amino acids cannot be synthesized via enzymatic methods.11 tert-Butylglycine was obtained in 70% yield by use of 3,3-dimethyl-2-oxobutanoic acid (10). 3-(2-Furanyl)-2-oxoethanoic acid and pyruvic acid possessing alpha proton did not work under the present conditions.

When the Rh(III) complex 6 is mixed with a 37 mol amount of HCOONH4 in CD3OH, the 1H NMR signals appear at delta -8.7 (t, J = 27.5 Hz) and delta -9.4 (t, J = 26.0 Hz) after 20 min at room temperature. These converge, after 2 h at 70 °C, to the signals at delta -18.4 (dd, J = 26.0 Hz) and delta -18.5 (dd, J = 26.0 Hz). These hydride species can be assigned to hydride-bridged dinuclear Rh complexes,12 which would be just kinetic repositories for the real catalytic species.13 We assume that [RhCp*Cl2]2 is converted, by the action of NH3 and HCOOH, into an ammonia-coordinated metal hydride RhCp*HCl(NH3) that acts as a chain carrier in the catalytic cycle.14 Coordination of NH3 onto Rh enhances the acidity of the hydrogen atom of NH3 and also the nucleophilicity of the hydride of RhH.13,15 The synergetic effect facilitates the formation of a catalyst-imine complex and then stabilizes the transition state by realizing the charge alternation on the C=N···H-N-Rh-H six atoms.13 The hydride transfer from RhH to the C=N carbon gives a catalyst-product complex, which releases a free amine product together with the formation of a metal amide species. The Rh-NH2 reacts quickly with formic acid to generate CO2 and RhCp*HCl(NH3), completing the catalytic cycle.

In summary, a novel catalytic system facilitating the Leuckart-Wallach-type reaction at a lower temperature with high chemoselectivity and generality has been established. Other than the desired primary amine products, the reaction produces only CO2 and H2O. Using 0.005 mol amount of catalyst and HCOONH4, a variety of substrates including simple ketones are converted to the corresponding primary amines. With alpha-keto acids, alpha-amino acids are the products. The reaction is clean and operationally simple. In most cases, only filtration is necessary to arrive at alpha-amino acids in high yields. Related studies on the asymmetric synthesis as well as the mechanism are being carried out. These results will be reported in due course.


  • Guest
« Reply #2 on: February 02, 2004, 07:54:00 PM »
Can someone help me synth this catalyst