Asymmetric hydrogenation of prochiral N-arylketimines to give optically active secondary amines at a temperature of -20° to 80° C., a hydrogen pressure of 105 to 6.106 Pa with the addition of catalytic amounts of an iridium compound of the formula III or IIIa [XIrYZ] (III) or [XIrY⊕ A⊖ (IIIa) in which X is two olefin ligands or one diene ligand, Y is a chiral diphosphine, the secondary phosphine groups of which are linked by 2-4 C atoms and which, together with the Ir atom, forms a 5-, or 6- or 7-ring, or Y is a chiral diphosphinite, the phosphinite groups of which are linked via 2 C atoms and which together with the Ir atom forms a 7-ring, Z is Cl, Br or I and A63 is the anion of an oxygen acid or complex acid, and if appropriate with the addition of an ammonium or alkali metal chloride, bromide or iodide.
1. A process for the preparation of an optically active seconday N-arylamine of the formula I [XIrYZ] (III) or [XIrY]⊕A⊖ (IIIa) in which X is two olefin ligands or one diene ligand, Y is a chiral diphosphine, the secondary phosphine groups of which are linked by 2-4 C atoms and which, together with the Ir atom, forms a 5-, 6- or 7-ring, or Y is a chiral diphosphinite, the phosphinite groups of which are linked via 2 C atoms and which together with the Ir atom forms a 7-ring, Z is Cl, Br or I and A⊖ is the anion of an oxygen acid or complex acid. 2. The process according to claim 1, wherein the reaction temperature is -20° to 50° C. 3. The process according to claim 1, wherein the hydrogen pressure is 2.105 Pa to 3.106 Pa. 4. The process according to claim 1 wherein X in the formulae III and IIIa is two ethylene or one open-chain or cyclic diene, the diene groups of which are linked via 1 or 2 C atoms. 5. The process according to claim 4, wherein the diene is hexadiene, norbornadiene or cyclooctadiene. 6. The process according to claim 1, wherein Y in the formulae III and IIIa is a chiral diphosphine, the phosphine groups of which are linked by 4 C atoms and which together with the Ir atom forms a 7-ring. 7. The process according to claim 1, wherein the phosphine groups contain C1 -C12 -alkyl, cycloalkyl which has 5 to 8 ring C atoms and can be substituted by 1 to 3 C1 -C6 -alkyl groups, phenyl, C7 -C12 -phenylalkyl or alkylphenylalkyl with 1 to 6 C atoms in the alkyl groups and 1 to 5 C atoms in the alkylene group. 8. The process according to claim 1, wherein A⊖ is CLO4, CF3 SO3⊖, BF4⊖, B(phenyl)4⊖, PF6⊖, SbCl6⊖, AsF6⊖ or SbF6⊖. 9. The process according to claim 1, wherein the iridium compound is added in an amount of 0.01 to 5 mol %, based on the compound of the formula II. 10. The process according to claim 1, wherein an ammonium or alkali metal chloride, bromide or iodide is additionally added. 11. The process according to claim 1, wherein, in formula III, X is cyclooctadiene, Z is Cl and Y is (R)- or (S)- 12. The process according to claim 1, wherein, in formula II, R1 is 2,6-dimethylphen-1-yl or 2-methyl-6-ethylphen-1yl, R2 is methyl and R3 is methoxymethyl. 13. The process according to claim 1, wherein the diphosphonite is 1-O-phenyl-4,6-O-(R)-benzylidene-2,3-O-bis(diphenylphosphino)-β-D-gly copyranoside.
The present invention relates to a process for the preparation of optically active secondary amines by asymmetric hydrogenation of prochiral N-arylketimines with chiral iridium diphosphine or diphosphinite complexes. European Patent A-No. 0,104,375 describes chiral diphosphine ligands, complexes of which with metals of group VIII of the periodic table can be used as catalysts for asymmetric hydrogenation of α-(acylamine)-acrylic acids. It has been found that iridium compounds with chiral diphosphine or diphosphinite ligands are suitable homogeneous asymmetric catalysts for the hydrogenation of prochiral N-arylketimines. This reaction leads to optically active secondary N-arylamines with high chemical conversions and good optical yields. Optically active means an excess of one enantiomer with the R- or S-configuration. The present invention relates to a process for the preparation of optically active secondary N-arylamines of the formula I [XIrYZ] (IIIa) or [XIrY]⊕A⊖ (IIIa) in which X is two olefin ligands or one diene ligand, Y is a chiral diphosphine, the secondary phosphine groups of which are linked by 2-4 C atoms and which, together with the Ir atom, forms a 5-, 6- or 7-ring, or Y is a chiral diphosphinite, the phosphinite groups of which are linked via 2 C atoms and which together with the Ir atom forms a 7-ring, Z is Cl, Br or I and A⊖ is the anion of an oxygen acid or complex acid. R1 can be substituted in any desired positions by identical or different radicals, for example with 1 to 5, preferably 1 to 3, substituents. The substitution in the two ortho-positions relative to the N atom may have a favourable influence on the desired yields, and in this case R2 is preferably not aryl. The two ortho-positions are preferably substituted, in particular by C1 -C12 -alkyl. Suitable substituents for aryl and aralkyl R1, R2 and R3 are: C1 -C12 -, preferably C1 -C6 - and in particular C1 -C4 -alkyl, -alkoxy or -alkylthio, for example methyl, ethyl, propyl, n-, i- and t-butyl, the isomers of pentyl, hexyl, octyl, nonyl, decyl, undecyl and dodecyl and corresponding alkoxy and alkylthio radicals; C1 -C6 -, preferably C1 -C4 -haloalkyl with preferably F and Cl as halogen, for example trifluoro- or trichloromethyl, difluorochloromethyl, fluorodichloromethyl, 1,1-difluoroeth-1-yl, 1,1-dichloroeth-1-yl, 1,1,1-trichloro- or -trifluoroeth-2-yl, pentachloroethyl, pentafluoroethyl, 1,1,1-trifluoro-2,2-dichloroethyl, n-perfluoropropyl, i-perfluoropropyl, n-perfluorobutyl, fluoro- or chloromethyl, difluoro- or dichloromethyl, 1-fluoro- or -chloro-eth-2-yl or -eth-1-yl, 1-, 2- or 3- fluoro- or -chloro-prop-1-yl or -prop-2-yl or -prop-3-yl, 1-fluoro-or-chloro-but-1-yl, -but-2-yl,-but-3-yl or-but-4-yl, 2,3-dichloro-prop-1-yl, 1-chloro-2-fluoro-prop-3-yl and 2,3-dichloro-but-1-yl; halogen, preferably F and Cl; C6 -C12 -aryl, -aryloxy or -arylthio, in which aryl is preferably naphthyl or, in particular, phenyl, C7 -C16 -aralkyl, -aralkoxy and -aralkylthio, in which the aryl radical is preferably naphthyl or, in particular, phenyl and the alkylene radical is linear or branched and contains 1 to 10, preferably 1 to 6 and in particular 1-3 C atoms, for example benzyl, naphthylmethyl, 1- or 2-phenyleth-1-yl or -eth-2-yl or 1-, 2- or 3-phenyl-prop-1-yl, -prop-2-yl or -prop-3-yl, benzyl being particularly preferred; the abovementioned radicals containing aryl groups can in turn be mono- or polysubstituted, for example by C1-C4 -alkyl, -alkoxy or -alkylthio, halogen, --OH, --CN, --CONR4 R5 or --COOR4, R4 and R5 being as defined above; examples are methyl, ethyl, n- and i-propyl, butyl, corresponding alkoxy and alkylthio radicals, F, Cl, Br, dimethyl-, methylethyl- and diethylcarbamoyl and methoxy-, ethoxy-, phenoxy- and benzyloxycarbonyl, secondary amino with 2 to 24, preferably 2 to 12 and in particular 2 to 6 C atoms, the secondary amino preferably containing 2 alkyl groups, for example dimethyl-, methylethyl-, diethyl-, methylpropyl-, methyl-n-butyl-, di-n-propyl-, di-n-butyl and di-n-hexylamino; --CONR4 R5, in which R4 and R5 independently of one another are C1 -C12 -, preferably C1 -C6 - and in particular C1 -C4 -alkyl, or R4 and R5 together are tetra- or pentamethylene or 3-oxapentylene, it being possible for the alkyl to be linear or branched, for example dimethyl-, methylethyl-, diethyl-, methyl-n-propyl-, ethyl-n-propyl-, di-n-propyl-, methyl-n-butyl-, ethyl-n-butyl-, n-propyl-n- butyl- and di-n-butylcarbamoyl; -COOR4, in which R4 is C1 -C12 -, preferably C1 -C6 -alkyl, which can be linear or branched, for example methyl, ethyl, n- and i-propyl, n-, i-and t-butyl, and the isomers of pentyl, hexyl, heptyl, octyl, nonyl, decyl, undecyl and dodecyl. Aryl R1 is preferably unsubstituted or substituted naphthyl or, in particular, phenyl. Heteroaryl R1 is preferable a 5- or 6- membered ring with 1 or 2 identical or different heteroatoms, in particular O, S or N, which preferably contains 4 or 5 C atoms and can be fused with benzene. Examples of heteroaromatics from which R1 can be derived are furan, pyrrole, thiophene, pyridine, pyrimidine, indole and quinoline. The substituents of R2 and R3 have the same preferred meanings as the substituents of R1. Alkyl R2 and R3 is preferably unsubstituted or substituted C1 -C6 -, in particular C1 -C4 -alkyl, which can be linear or branched. Examples are methyl, ethyl, i- and n-propyl, i-, n- and t-butyl, and the isomers of pentyl, hexyl, heptyl, octyl, nonyl, decyl, undecyl and dodecyl. Unsubstituted or substituted cycloalkyl R2 and R3 preferably contains 3 to 6, in particular 5 or 6, ring C atoms. Examples are cyclopropyl, cyclobutyl, cyclopentyl, cyclohexyl, cycloheptyl and cyclooctyl. Aryl R2 and R3 is preferably unsubstituted or substituted naphthyl or, in particular, phenyl. Aralkyl R2 and R3 is preferably unsubstituted or substituted phenylalkyl with 1-10, preferably 1 to 6 and in particular 1 to 4 C atoms in the alkylene, it being possible for the alkylene to be linear or branched. Examples are, in particular, benzyl, and 1-phenyleth-1-yl, 2-phenyleth-1-yl, 1-phenylprop-1-yl, 1-phenylprop-2-yl, 1-phenylprop-3-yl, 2-phenylprop-1yl, 2-phenylprop-2-yl and 1-phenylbut-4-yl. R4 and R5 in -- CONR4 R5 and --COOR4 R2 and R3 are preferably C1 -C6, in particular C1 -C4 -alkyl, or R4 and R5 together are tetramethylene, pentamethylene or 3-oxapentylene. Examples of alkyl have been mentioned above. Alkylene R2 and R3 together or R1 bounded to R3 is preferably interrupted by 1 --O--, --S-- or --NR4 --, preferably --O--. R2 and R3 or R1 bounded to R3, together with the C atom or the --N=C-group to which they are bonded preferably form a 5- or 6-membered ring. The substituents have the abovementioned preferred meanings. Fused alkylene R2 and R3 or R1 bonded to R3 is preferably alkylene fused with benzene or pyridine. Examples of alkylene are: ethylene, 1,2- or 1,3-propylene, 1,2-, 1,3- or 1,4-butylene, 1,5pentylene and 1,6-hexylene. Examples of alkylene which is interrupted or substituted by =O are 2-oxa-1,3-propylene, 2-oxa-1,4-butylene, 2-oxa- or 3-oxa-1,5-pentylene, 3-thia-1,5-pentylene, 2-thia-1,3-propylene, 2-methylimino-1,3propylene, 2-ethylimino-1,4-butylene, 2- or 3-methylimino-1,5-pentylene, 1-oxo-2-oxa-1,3-propylene, 1-oxo-2-oxa-1,4-butylene, 2-oxo-3-oxa-1,4-butylene and 1-oxa-2-oxo-1,5-pentylene. Examples of fused alkylene are: Examples of fused and interrupted alkylene which is unsubstituted or substituted by =O are In a preferred group, in formula II R1 is 2,6-dimethylphen-1-yl or 2-methyl-6-ethylphen-1-yl, R2 is methyl and R3 is methoxymethyl. N-Arylimines of the formula II are known or can be prepared from ketones and arylamines by known processes. In one embodiment of the process, the N-arylimines of the formula II can also be prepared in situ from the corresponding ketones and arylamines. The process is preferably carried out at a temperature of -20° to 50° C., in particular -20° to 20° C. and especially ---20° to 10° C., and preferably under a hydrogen pressure of 2.105 to 3.106 Pa, in particular 8.105 to 3.106 Pa. In the formulae III and IIIa, an olefin ligand X can be, for example, butene, propene or, in particular, ethylene, and the diene ligand is preferably an open-chain or cyclic diene, the diene groups of which are linked by one or two C atoms. The diene is preferably hexadiene, cyclooctadiene or norbornadiene. In the chiral diphosphine, the phosphine groups are preferably linked via an aliphatic group with 2-4 C atoms which can be substituted by C1 -C4 -alkyl, C5 - or C6 -cycloalkyl, phenyl or benzyl. The aliphatic group can be alkylene or a cycloaliphatic group with 5 or 6 ring C atoms, or an aliphatic-heterocyclic group with 1 or 2 --O--or =N--C1 -C12 -alkyl or -acyl or -aminocarbonyl, -phenyl or -benzyl and 3-5 C. atoms in the ring. The rings can be substituted by C1 -C4 -alkyl, C5 -C7 -cycloalkyl, phenyl or benzyl. Y in the formulae III or IIIa is preferably a chiral diphosphine, the phosphine groups of which are linked by 4 C atoms and which, together with the Ir atom, forms a 7-ring. The phosphine groups and phosphinite groups preferably contain C1 -C12 -alkyl, cycloalkyl which has 5 to 8 ring C atoms and can be substituted by 1 to 3 C1 -C6 -alkyl groups, phenyl, C7 -C12 -phenylalkyl or alkylphenylalkyl with 1 to 6 C atoms in the alkyl groups and 1 to 5 C atoms in the alkylene group. t-Butyl, phenyl, benzyl and cyclohexyl are particularly preferred. Suitable chiral diphosphines are described in H. B. Kagan, Chiral Ligands for Asymmetric Catalysis, Asymmetric Synthesis, Volume 5, pages 13 -23, Academic Press, Inc., N.Y. (1985). Examples are (Ph is phenyl): An example of diphosphinites is 1-0-phenyl-4,6-0-(R)-benzylidene-2,3-0-bis(diphenylphosphino)-β-D-glu copyranoside of the formula In formula III, Z is preferably Cl or Br. A⊖in formula IIIa is preferably ClO4⊖, CF3 SO3⊖, BF4⊖, B(phenyl)4⊖, PF6⊖, SbCl6⊖, AsF6⊖ or SbF6⊖. A preferred group of iridium compounds are those of III in which X is cyclooctadiene, Z is Cl and Y is (R)-- or (S)-- The iridium compounds of the formulae III and IIIa are known or can be prepared by known processes, see, for example, R. Uson et al., Inorg. Chim. Acta 73, page 275 et seq. (1983); S. Brunie et al., Journal of Organometallic Chemistry, 114 (1976), pages 225-235 and M. Green et al., J. Chem. Soc. (A), pages 2334 et seq. (1971). The iridium compounds can be used as isolated compounds. It is advantageous to prepare the compounds in situ and to use them directly. The iridium compounds are preferably used in amounts of 0.01 to 5, in particular 0.05 to 2 mol %, based on the compounds of the formula II. A preferred process procedure comprises additionally using an ammonium or alkali metal chloride, bromide or iodide. The addition of chlorides, bromides or iodides is particularly advantageous if compounds of the formula IIIa are used as catalysts. The chlorides, bromides and iodides are preferably used in amounts of 0.01 to 100, in particular 0.05 to 50 mol %, based on the compounds of the formula II. Preferred salts are the iodides. Ammonium is preferably tetraalkylammonium with 1 to 6 C atoms in the alkyl groups, and the alkali metal is preferably sodium, lithium or potassium. The reaction can be carried out in the absence or presence of solvents. Suitable solvents, which can be used by themselves or as a mixture of solvents, are, for example: aliphatic and aromatic hydrocarbons, for example pentane, hexane, cyclohexane, methylcyclohexane, benzene, toluene and xylene; alcohols, such as, for example, methanol, ethanol, propanol and butanol; ethers, such as, for example, diethyl ether, diethylene glycol dimethyl ether, tetrahydrofuran and dioxane; halogenohydrocarbons, for example methylene chloride, chloroform, 1,1,2,2-tetrachloroethane and chlorobenzene; esters and lactones, for example ethyl acetate, butyrolactone and valerolactone; and acid amides and lactams, for example dimethylformamide, dimethylacetamide and N-methylpyrrolidone. The compounds of the formula I are biologically active substances or intermediates for the preparation of such substances with an N-aryl-secondary amine group, in particular in the pharmaceuticals and agrochemicals sector. Thus, for example, o,o-dialkylarylketamine derivatives, in particular those with alkyl and/or alkoxyalkyl groups, act as fungicides, in particular as herbicides. The derivatives can be amine salts, acid amides, for example of chloroacetic acid, tertiary amines and ammonium salts (see, for example, European Patent A-No. 0,077,755 and European Patent A-0,115,470). The following examples illustrate the invention in more detail. 10 ml each of methanol and benzene are introduced into a 50 ml two-necked flask under an N2 inert gas atmosphere. Thereafter, 4.10- mol of [Ir(cyclooctadiene)Cl]2, 8.8×10-5 mol of diphosphine and 1.2×10-3 mol of tetrabutylammonium iodide (only in the case of Example 1 and 2) are added in succession. After each addition, the mixture is stirred until a homogeneous solution is present (solution B). Solutions A and B are introduced in succession into a 0.3 l steel autoclave with a capillary with exclusion of air. 2×106 Pa of hydrogen are forced in through a gas inlet valve. At the same time, 1.4×107 Pa of hydrogen are forced into a 100 ml reservoir. The temperature is 20-22° C. The reaction is carried out under a constant pressure of hydrogen of 2×106 Pa until no further uptake of hydrogen takes place. The reaction mixture is then flushed into a 250 ml flask with nitrogen. The solvent is removed at 80° C. on a rotary evaporator. A crude product is obtained and is distilled under a high vacuum (1-10 Pa). The optical yield is then determined by polarometry (A. F. Lee et al., J. Chem. Arc. 1954, 145) or by means of1 H-NMR, using shift reagents. In Example 2, 80 mmol of ketimine are used. In the remaining examples, the catalyst concentration is 1 mol % (40 mmol), 20 ml of solvent being used. The reactions are carried out in a 120 ml glass or 65 ml steel autoclave. The optical yields (ee in %), reaction times, conversion, solvents and reaction temperatures are shown in Table 1. The procedure is as in Examples 3-10 and the reaction conditions are changed. N(2-Methyl-6-ethylphen-1-yl) -methyl-methoxymethyl-ketimine is used as the ketimine. (-)-DIOP (Example 11) and (+)-DIOP (Examples 12-14) are used as the diphosphine. The hydrogen pressure is 2×106 Pa. The results are shown in Table 2. The procedure is as in Examples 3-10, using various ketimines. Methylene chloride is used as the solvent in Examples 23, and otherwise benzene/methanol is used. The results are shown in Table 3. The procedure is as in Examples 3-10. The reaction conditions and results are shown in Table 4. The procedure is as in Examples 3-10, but using 1--O--phenyl-4,6--O--(R)-benzylidene-2,3--O--bis(diphenyl-phosphino)-.beta .-D-glucopyranoside, 2 equivalents of TBAI per iridium and N-(2-methyl-6-ethylphen-1-yl)-methyl-methoxmethylketimine. The reaction time is 21.5 hours. At a conversion of 67%, the enantiomer excess (ee) is 44% (S). The procedure is according to the reaction conditions of Example 3 and the ketimine is produced in situ from 40 mmol of 2,6-dimethylaniline and 40 mmol of methoxyacetone. Tetrahydrofuran is used as the solvent. The reaction time is 144 hours. At a conversion of 37%, an enantiomer excess of 64% ee (S) is achieved.EXAMPLES 1-10: p 160 mmol of N(2,6-dimethylphen-1-yl)-methoxymethylmethylketimine are introduced into a 250 ml two-necked flask under an N2 inert gas atmosphere. The flask is evacuated to 5×103 Pa and flushed with nitrogen. Thereafter, 30 ml each of methanol and benzene are added and the mixture is stirred at room temperature for 2 minutes (solution A).
EXAMPLE 11-14
EXAMPLES 15-23
EXAMPLES 24-40
EXAMPLE 41
EXAMPLE 42: