Preparation of phosphine ligands

Details Preparation of phosphine ligands Preparation of phosphine ligands

2001-03-29

2002-11-26

Kumar, Shailendra (Department: 1621)

Organic compounds -- part of the class 532-570 series

Organic compounds

Amino nitrogen containing

C564S139000, C564S155000

Reexamination Certificate

active

06486347

ABSTRACT:

FIELD OF THE INVENTION
This invention relates to processes suitable for the large-scale preparation of enantiomerically-enriched phosphines, especially those useful as ligands in asymmetric allylic substitution catalysts.
BACKGROUND OF THE INVENTION
Chiral phosphine ligands such as (1) and (2), and the opposite enantiomers thereon have been shown to be effective in palladium(O)-catalysed asymmetric allylic substitution reactions. For a review, see B. M. Trost and D. L. Van Vranken,
Chem Rev. (
1996) 96:395. For specific examples of such ligands, see WO-A-93 12260, B. M. Trost and X. Ariza,
Angew. Chem. Int. Ed. Engl
. (1997) 36:2635, and B. M. Trost and F. D. Toste,
J. Am. Chem. Soc.
(1998) 120:815.
Such catalysts are eminently suitable for industrial applications, especially for the provision of chiral pharmaceutical intermediates in high enantiomeric purity. For this purpose, and in other industrial applications such as flavour and fragrance fine chemicals, the development of manufacturing processes requires in turn large amounts of a ligand such as (1) or (2), e.g. in kilogram quantity or greater. Thus, there is a requirement for efficient and scaleable methods for synthesis of such ligands.
A process for the manufacture of these ligands has been described by B. M. Trost. D. L. Van Vranken and C. Bingel,
J. Am. Chem. Soc.
(1992) 114:9327. This process is depicted in Scheme 1.
There are several limitations to using the methods described for the manufacture of this class of ligands on a commercially useful scale.
Firstly, the conversion of 2-chlorobenzoic acid to 2-(diphenylphosphino)benzoic acid gives modest yields, using literature methods. A 74% yield for this process is reported in DE-A-2264088, but in our hands this process gave yields in the range of 48 to 54% on a 1 to 2 kg scale. Typically, yields in the range of 40 to 50% are obtained; see
Inorg Synth.
(1982), 21:175, where a 49% yield is quoted.
An alternative method for preparing this class of phosphines has been described, starting from a fluoro-derivative; see Williams et al,
Synlett.
(1993) 509. An example is as follows:
This paper specifically states that 2-fluorobenzoic acid and the corresponding carboxylate did not provide the desired product under these conditions.
Hingst et al, Eur. J Inorg. Chem. (1998) 73, reports that the nucleophilic phosphanylation of 2-fluorobenzoic acid with Ph
2
PH in superbasic medium gave poor yields, and that a “fair yield” is obtained if a solution of Ph
2
PK in THF is used to phosphanylate the K salt of 2-fluorobenzoic acid.
Secondly, the diamine has to be isolated from its salt. The tartrate salt is commonly used to provide enantiomerically-pure diamine; see E. N. Jacobsen et al,
J. Org. Chem.
(1994) 59:1939, and references therein, for 1,2-diaminocyclohexane, and E. J. Corey and S. Pikul,
Org Synth.
(1992) 71:22, for 1,2-diphenyl-1,2-ethylenediamine. This salt has to be cracked, and then the air-sensitive and water-soluble diamine has to be extracted from aqueous solution. U.S. Pat. No. 5,399,771 describes an alternative method, where the tartrate salt is slurried in methanolic KOH, potassium tartrate is removed by filtration, and the diamine is isolated after evaporation of the solvent. This method still requires isolation of the sensitive diamine.
Thirdly, coupling of the diamine with 2-diphenylphosphino)benzoic acid using 1,3-dicyclohexylcarbodiimide (DCC) gives the desired product in yields ranging from 60 to 90%, but requires silica chromatography for purification. The ligand is isolated as a waxy or glassy solid, with a broad melting point range 80-120° C.; see Trost et al (1992) supra.
The known synthesis of the desired ligands is suitable for the preparation of 10 to 100 g samples, but is not economic for kilogram quantities of ligands. The poor yield of 2-(diphenylphosphino)benzoic acid is one factor that directly impacts on the commercial viability of the process. Isolation of the diamine, chromatography of the ligand, and the non-crystalline nature of the ligand isolated, all make the current process inefficient.
U.S. Pat. No. 5,801,263 discloses the following reaction
SUMMARY OF THE INVENTION
The present invention is based on three discoveries that give rise to an efficient, scaleable and economical synthesis of compounds effective as ligands, especially those represented by formula (3)
and the opposite enantiomers thereof, wherein Ar is an aromatic ring bearing the PR
3
R
4
and CO groups in a 1,2-relationship. Ar is also optionally substituted by one or more non-interfering groups. The respective R groups are each any non-interfering group, or R
1
and R
2
may be joined to form a ring.
One aspect of the invention concerns the use of 2-fluorobenzoic acid in the reaction with NaPPh
2
, e.g. generated from triphenylphosphine, sodium and liquid ammonia at −60° C., to provide 2-(diphenylphosphino)benzoic acid. Analysis of the crude product from the reaction of NaPPh
2
with 2-chlorobenzoic acid showed that the major by-product was 3-(diphenylphosphino)benzoic acid, probably arising from benzyne formation and phosphine addition. Thus, several recrystallisations are required to access pure product and lower yields are obtained. Replacing 2-chlorobenzoic acid by the 2-fluoro derivative led to the surprising discovery that yields of 2-(diphenylphosphino)benzoic acid ranging from 75 to 85% could be obtained, with no significant contamination from 3-(diphenylphosphino)benzoic acid. Thus, a key component of the ligand system can be manufactured more efficiently.
A second aspect of this invention concerns the coupling of 2-(diphenylphosphino)benzoic acid, or a derivative thereof represented by formula (4), with a diamine, such as 1,2-diaminocyclohexane, represented by formula (5). An important feature of this invention is that, when the diamine is released from a salt form, such as the tartrate salt represented by formula (6), prior to the coupling reaction, there is no need to isolate the diamine from aqueous solution. Accordingly, the amine is generated in situ, by which is meant that it is not isolated.
Surprisingly, a mixed anhydride of 2-(diphenylphosphino)benzoic acid with diphenyl chlorophosphate (4: R
3
═R
4
═Ph and Y═PO(OPh)
2
) in dichloromethane reacts with an aqueous solution of the diamine (5)/potassium tartrate salt mixture. This mixture is obtained by dissolving the tartrate salt (6) in water and adding potassium carbonate or another, equivalent base. This provides the ligand, e.g. (1), in quantitative crude yield. An unexpected feature is that very little hydrolysis of the mixed anhydride to the acid is observed in this reaction, and excellent conversion to the ligand is obtained.
Another aspect of the present invention is that, surprisingly, the ligand is obtained in a crystalline form, whereas the literature protocol affords ligand as a glassy solid. Thus, the process of the present invention has advantages with respect to material handling and transfer. For example, isolation and purification of the ligand, after the coupling reaction, are simple. The reaction carried out as described above, is very clean and few by-products are seen. No extensive chromatography is required. In particular, after a simple filtration of the crude ligand through silica gel, the ligand can be recrystallised from hot acetonitrile, or acetonitrile/acetone mixtures, to provide the preferred crystalline ligand (1) with a sharp melting point (134-136° C.). The yield of isolated pure, crystalline ligand is between 65 to 75%.
These discoveries allow the ligand to be manufactured on a large scale, with no need for extensive chromatography, and in reproducible quality.
DESCRIPTION OF THE INVENTION
The nature of R
1
, R
2
, R
3
, R
4
or any substituent on the Ar rings. e.g. of up to 10, 20 or 30 C atoms, is not critical to the invention. It will be evident to the skilled man, as to which substituents will or will not affect the reaction. Similarly, while Y is preferably —P(O)(OZ)
2
, and Z is preferably a hydrocarbon substituent, most preferab

Affiliated with

Also associated with

Rating

Preparation of phosphine ligands does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Preparation of phosphine ligands, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Preparation of phosphine ligands will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2932035