HK1131964A - Method for manufacture of escitalopram - Google Patents

Description

Method for preparing escitalopram

Field of the invention

The invention relates to a method for preparing a known antidepressant drug namely escitalopram.

Background of the invention

Escitalopram is a well-known antidepressant drug having the following structure:

it is a selective, centrally active inhibitor of serotonin (5-hydroxytryptamine; 5-HT) resorption, with corresponding antidepressant activity.

U.S. Pat. No. 4,943,590 discloses escitalopram and its pharmaceutical activity. Two methods of preparing escitalopram are disclosed. One is the fractional crystallization of 4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile in 2-propanol in the form of a salt with (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid. The crystalline product produced in this way consists of small crystals which are discharged slowly and tend to retain the mother liquor. Inadequate removal of the mother liquor results in a product with low enantiomeric purity, requiring additional purification. Purification is time consuming and solvent consuming. These problems are more pronounced on an industrial scale.

Brief description of the invention

It has now been found that 4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile can be resolved by fractional crystallization of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile in the form of a salt with (+) - (S, S) -or (-) - (R, R) -O, O '-di-p-methylbenzoyl tartaric acid in a solvent system comprising 1-propanol and wherein per mol of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile, the use of not more than 0.5mol of (+) - (S, S) -or (-) - (R, R) -O, O' -di-p-methylbenzoyltartaric acid gives a crystalline product in which the crystals are larger than those obtained from 2-propanol and have different shapes. This process has proven to be a robust and stable method of producing crystals with good filtration properties. This results in better drainage characteristics and reduces filtration times which have a significant impact on mass production. Typical filtration times for industrial scale batch production are several hours or less. This resolution method can be used to prepare escitalopram.

Detailed Description

In one aspect of the invention there is provided a process for the resolution of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile in the form of a racemic or non-racemic mixture of enantiomers into its isolated enantiomers, said process comprising reacting the compound of formula (I) with the (+) - (S, S) -or (-) - (R, R) -enantiomer of O, O' -di-p-methylbenzoyltartaric acid, a step of fractional crystallization of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile in a solvent system comprising 1-propanol, ethanol or acetonitrile.

In a particular embodiment, no more than 1 mole, more particularly no more than 0.5 mole of the (+) - (S, S) -or (-) - (R, R) -enantiomer of O, O '-di-p-methylbenzoyl tartaric acid is used per mole of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile.

In one embodiment, (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid is used. In another embodiment, (-) -O, O' -di-p-methylbenzoyl- (R, R) -tartaric acid is used.

In a particular embodiment, 1-propanol is the major component in the solvent system. In a more particular embodiment, the 1-propanol component comprises at least 50%, such as at least 75%, at least 90% or at least 95% of the solvent system, and in most particular embodiments 1-propanol is the only solvent.

In an equivalent embodiment, ethanol is the major component of the solvent system. In a more particular embodiment, the ethanol component comprises at least 50% (v/v) of the solvent system, such as at least 75%, at least 90%, or at least 95%, and in most particular embodiments ethanol is the only solvent.

In an equivalent embodiment, acetonitrile is the major component of the solvent system. In a more particular embodiment, the acetonitrile component comprises at least 50% (v/v) of the solvent system, such as at least 75%, at least 90%, or at least 95%, and in most particular embodiments acetonitrile is the only solvent.

In another embodiment, the solvent system comprises one or more organic co-solvents, in particular, an organic co-solvent selected from the group consisting of toluene, diethyl ether, ethyl acetate, dichloromethane and acetonitrile, more particularly from toluene. In more particular embodiments, the amount of co-solvent is from 0 to 20% (v/v) of the solvent system, e.g., from 0 to 15%, from 0 to 10%, from 0.5 to 8%, from 1 to 5%, or from 1.5 to 3%.

In yet another embodiment, the solvent system comprises water. In a more particular embodiment, the amount of water is 0-8% (v/v), such as 0.05-5%, 0.1-3%, or 0.15-2% of the solvent system.

In yet another embodiment, the solvent system comprises an achiral acid which under the present conditions is capable of protonating 4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile without precipitating 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile in salt form. In a particular embodiment, the achiral acid is selected from the group consisting of organic acids such as formic acid, acetic acid, trifluoroacetic acid, methanesulfonic acid, more particularly from acetic acid. In a more particular embodiment, the amount of achiral acid is 0-0.5 equivalents, for example 0-0.4 equivalents, relative to the amount of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile.

In a further embodiment, the solvent system is cooled, together with dissolved 4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile and (+) - (S, S) -or (-) - (R, R) -O, O' -di-p-methylbenzoyl tartaric acid, from a first temperature in the range of 20 ℃ to the reflux temperature of the solvent system, in particular from 25 ℃ to 70 ℃, more in particular from 30 ℃ to 50 ℃, to a second temperature in the range of 0 ℃ to 40 ℃, in particular from 10 ℃ to 30 ℃, more in particular from 15 ℃ to 25 ℃. In a particular embodiment, the difference between the first temperature and the second temperature is between 5 ℃ and 50 ℃, in particular between 10 ℃ and 40 ℃, more in particular between 15 ℃ and 30 ℃.

In a particular embodiment, the mixture of 4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile, (+) - (S, S) -or (-) - (R, R) -O, O' -di-p-methylbenzoyl tartaric acid and the solvent system is kept at the first temperature for a period of time within the range of 0 to 4 hours, more particularly 0.5 to 3 hours, most particularly 1 to 2 hours, before cooling.

In another particular embodiment, a mixture of 4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile, (+) - (S, S) -or (-) - (R, R) -O, O' -di-p-methylbenzoyltartaric acid and a solvent system is seeded with crystallization of the desired salt at a first temperature or during cooling. Typically, the amount of seed crystals is 0.4 to 0.8g of seed crystals per kg of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile, more typically 0.45 to 0.7g/kg, most typically 0.5 to 0.6 g/kg.

In an equivalent particular embodiment, the cooling is performed within 8 hours, particularly within 4 hours, more particularly within 2 hours. In another equivalent particular embodiment, the precipitated salt is separated from the mother liquor within 8 hours, more particularly within 4 hours, after the precipitation has started. In yet another equivalent particular embodiment, the separated salt is washed within 4 hours, more particularly within 2 hours.

In yet another further embodiment, the isolated salt is reslurried or recrystallized one or more times in a solvent system comprising 1-propanol or ethanol by heating to a temperature in the range of 30 ℃ to the reflux temperature of the solvent, more particularly 40 ℃ to 60 ℃, followed by cooling to a temperature in the range of 0 ℃ to 40 ℃, particularly 10 ℃ to 30 ℃, more particularly 15 ℃ to 25 ℃. In a particular embodiment, 1-propanol or ethanol is the major component of the solvent system. In a more particular embodiment, 1-propanol or ethanol comprises at least 50% of the solvent, such as at least 75%, at least 90%, or at least 95%, and in a most particular embodiment, 1-propanol or ethanol is the only solvent. In a particular embodiment, the major components of the solvent system used for reslurry are the same as the major components of the solvent system used for crystallization.

In another aspect of the invention, there is provided a process for the preparation of escitalopram comprising resolving 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile into the above enantiomer thereof.

In one embodiment, one of the isolated enantiomers of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile is stereoselectively converted to escitalopram.

In a particular embodiment, S-4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile is stereoselectively converted to escitalopram.

In a more particular embodiment, S-4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile is reacted with an active acid derivative, such as an acid chloride or anhydride, in the presence of a base, such as triethylamine or pyridine, in particular with methanesulfonyl chloride or toluenesulfonyl chloride.

The terms "resolution" and "resolving" as used throughout the specification and claims refer to processes in which the enantiomeric purity of a racemic or non-racemic mixture of enantiomers is increased, for example, by reducing the proportion of the undesired enantiomer in the mixture by at least 20% to the extent that the enantiomeric proportion in the resulting mixture is in favor of the desired enantiomer, as illustrated by the following two examples:

i) converting the racemic mixture (50:50) into a mixture having an enantiomeric ratio of at least 60:40, or

ii) converting the 80:20 mixture to a mixture having an enantiomeric ratio of at least 84: 16.

As used throughout the specification and claims, the term "racemic mixture" refers to a 50:50 enantiomeric mixture, while the term "non-racemic mixture" refers to any enantiomeric mixture that is not 50: 50.

The term "isolated enantiomer" as used throughout the specification and claims refers to an enantiomer that is at least 95% enantiomerically pure, specifically at least 97% enantiomerically pure, more specifically at least 98% enantiomerically pure, and most specifically at least 99% enantiomerically pure.

The term "fractional crystallization" as used throughout the specification and claims refers to a process in which one enantiomer crystallizes as a salt and a chiral acid more preferentially than the other enantiomer, in which process crystallization may begin with dissolution of the salt or suspension of the salt.

The term "solvent system" as used throughout the specification and claims refers to a combination of an organic solvent and water when present. The term "organic solvent" includes any protic or aprotic solvent, such as alcohols, esters, alkanes, ethers and aromatics, but does not include acids such as carboxylic acids and bases such as amines.

The term "reslurry" as used throughout the specification and claims refers to a process in which a crystalline material is suspended in a solvent at a temperature such that the crystalline material is partially dissolved, and then cooled such that the dissolved material is again partially crystallized.

The term "recrystallization" as used throughout the specification and claims refers to the process of dissolving a crystalline material in a solvent at a temperature, optionally filtering to remove insoluble material, and then cooling to partially crystallize the dissolved material again.

If the chiral purity of the S-or R-4- [4- (dimethylamino) -1- (4 ' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile product obtained from the process disclosed herein is not sufficiently high, especially as may occur when using mother liquors as products, the racemic 4- [4- (dimethylamino) -1- (4 ' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile can be precipitated from a solution of the product enriched in S-or R-4- [4- (dimethylamino) -1- (4 ' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile, leaving further enriched S-or R-4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile in the solution, thereby further improving the chiral purity, as disclosed in WO 2004/056754.

In EP0347066, it is disclosed that S-4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile can be converted stereoselectively into escitalopram.

In WO03/000672, it is disclosed that R-4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile can be converted into escitalopram.

If the chiral purity of the S-citalopram product resulting from the process disclosed herein is not sufficiently high, as may occur especially when using a mother liquor as the product, the chiral purity may be further improved by precipitating racemic citalopram from a solution of the product enriched in S-citalopram, leaving further enriched S-citalopram in the solution, as disclosed in WO 2003/000672.

Experimental stage

The following abbreviations are used throughout the specification and claims:

eq means an equivalent weight, calculated as molar ratio with respect to the amount of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile.

V refers to volume, calculated as milliliters of solvent per gram of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile free base.

Molar yields were calculated as moles of S-4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile in the product per mole of racemic 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile starting material.

Seed crystal

A solution of S-4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile (10g) in 1-propanol (9.5ml) and a solution of (+) O, O' -di-p-methylbenzoyl (S, S) tartaric acid (11.6g) in 1-propanol (88ml) were mixed, alternatively, seed crystals were prepared by mixing a solution of R-4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile (10g) in 1-propanol (9.5ml) and a solution of (-) -O, O' -di-p-methylbenzoyl (R, R) tartaric acid (11.6g) in 1-propanol (88 ml). Alternatively, the seed crystals can be prepared in an approximate manner using ethanol as a solvent instead of 1-propanol. Preferably, the same solvent is used for crystallization to crystallize the seed crystals. The crystals produced according to the following examples can also be used as seed crystals.

S-or R-4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile for use in the preparation of seed crystals can be obtained according to the disclosure of EP0347066 or WO 03/006449.

Experimental example 1

(+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid (0.39eq) was dissolved in 1-propanol (3.44V). The mixture was heated to about 40 ℃ and acetic acid (0.2eq) was added. The solution was transferred to 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile free base (0.95V) containing 0.1V toluene within 1 hour. The dissolution mixture, which now contains 4.4V 1-propanol as a whole, was inoculated with seed crystals containing S-4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile and (+) -O, O' -di-p-methylbenzoyl- (S, S) tartaric acid, and then stirred at 40 ℃ for 2 hours. The mixture was cooled to 20-25 ℃ within 2 hours. The product was filtered and washed twice with 1-propanol. Enantiomeric purity is typically in the range of about 91% to about 98% S.

The product was reslurried in 1-propanol (2.5V) at about 50 ℃ for 2 hours. The mixture was cooled to 20-25 ℃. The product was filtered and washed with 1-propanol. Enantiomeric purity is typically about 99.3% S.

Molar yields are typically 34-36%.

Experimental example 2

(+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid (0.4eq) was dissolved in 1-propanol (3.5V). The mixture was heated to about 40 ℃ and acetic acid (0.2eq) was added, whereupon the solution was transferred to a solution of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile free base in 1-propanol containing 0.1V of toluene. The dissolution mixture, which now contains 4.5V of 1-propanol as a whole, was seeded with seed crystals containing S-4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile and (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid, and then stirred at 40 ℃ for 2 hours. The mixture was cooled to 20-25 ℃ within 2 hours. The product was filtered (filter reactor) and washed with 1-propanol.

Enantiomeric purity is typically about 97% S or higher.

Typical batch production gives the following molar yields: 33.8%, enantiomeric purity 99.0% S.

Experimental example 3

The general procedure of Experimental example 2 was followed, but using 0.5eq (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid and 10V 1-propanol. Toluene or acetic acid was not present in the system.

Typical batch production gives the following molar yields: 29.5 percent; enantiomeric purity 99.2% S.

Experimental example 4

The general procedure of experimental example 2 was applied. To a solution of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile free base in 1-propanol was added 0.05V of water. Toluene or acetic acid was not present in the system.

Typical batch production gives the following molar yields: 29.3 percent; enantiomeric purity 99.3% S.

Experimental example 5

The conventional procedure of Experimental example 2 was applied using only 0.25eq (+) O, O' -di-p-methylbenzoyl- (S, S) tartaric acid. Acetic acid is not present in the system.

Typical batch production gives the following molar yields: 29.4 percent; enantiomeric purity 99.0% S.

Experimental example 6

The general procedure of experimental example 2 was applied. Acetic acid is not present in the system.

Typical batch production gives the following molar yields: 32.6 percent; enantiomeric purity 98.0% S.

Experimental example 7

The general procedure of experimental example 2 was applied. Acetic acid is not present in the system. The experiment was carried out with a small amount of water (0.01V).

Typical batch production gives the following molar yields: 32.5 percent; enantiomeric purity 98.7% S.

Experimental example 8

The general procedure of experimental example 2 was applied. Acetic acid is not present in the system. The experiment was carried out at a higher water content (0.05V).

A typical batch production gives:

molar yield: 34.7 percent; enantiomeric purity 99.0% S.

Experimental example 9

The general procedure of experimental example 2 was applied. To a solution of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile free base in 1-propanol was added an additional small amount of water (0.05V).

Typical batch production gives molar yields: 33.0 percent; enantiomeric purity 99.1% S.

Experimental example 10

100g (0.292 mol) of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile are dissolved in 150ml of pure ethanol at 40 ℃. Maintaining the temperature at about 40 ℃, a solution made of 57.5g (0.148 mole) of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid and 350ml of pure ethanol was added over 1 hour. The mixture was inoculated and then cooled to room temperature overnight. The suspension was cooled to 0 ℃ and then filtered.

Molar yield: 29.5 percent; enantiomeric purity 98.2% S.

Experimental example 11

(+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid (0.25eq) was dissolved in 1-propanol (200 ml). The mixture was heated to about 40 ℃ and then transferred to a solution of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile free base (100g) in 1-propanol (100ml) containing 11g of toluene. The dissolution mixture, which now comprises 3V 1-propanol as a whole, was seeded with seeds comprising S-4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile and (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid at 40 c and then stirred for 2 hours at 40 c. The mixture was cooled to 20 ℃ within 2 hours and kept at 20 ℃ overnight. The product was filtered (filter reactor) and washed with 1-propanol.

Molar yield: 31.8 percent; enantiomeric purity 95.5% S.

Experimental example 12

Experimental example 11 was repeated except that the total volume of 1-propanol was 10V.

Molar yield: 30.7 percent; enantiomeric purity 98.9% S.

Experimental example 13

Experimental example 11 was repeated except that the total volume of 1-propanol was 4.3V and 0.39eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used. This example was repeated several times. The crystallization batch is kept at 20 ℃ for up to 16 hours, typically up to 8 hours.

Molar yield: about 35%; enantiomeric purity: > 98% S.

Experimental example 14

Experimental example 11 was repeated except that the total volume of 1-propanol was 4.5V, 0.50eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was 0.5 hr.

Molar yield: 36 percent; enantiomeric purity: 97.2% S.

Experimental example 15

Experimental example 11 was repeated except that the total volume of 1-propanol was 4.4V, 0.60eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was 0.5 hr.

Molar yield: 38.9 percent; enantiomeric purity: 82.8% S.

Experimental example 16

Experimental example 11 was repeated except that the total volume of 1-propanol was 4.5V, 0.675eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 35.2 percent; enantiomeric purity: 76.2% S.

Experimental example 17

Experimental example 11 was repeated except that the total volume of 1-propanol was 6V, 0.75eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was 0.5 hours.

Molar yield: 24.8 percent; enantiomeric purity: 99.4% S.

Experimental example 18

Experimental example 11 was repeated except that the total volume of 1-propanol was 6V, 0.75eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 31 percent; enantiomeric purity: 99.4% S.

Experimental example 19

Experimental example 11 was repeated except that the total volume of 1-propanol was 4.5V, 0.75eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 30.3 percent; enantiomeric purity: 99.0% S.

Experimental example 20

Experimental example 11 was repeated except that the total volume of 1-propanol was 4.5V, 0.75eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was 4 days.

Molar yield: 32.2 percent; enantiomeric purity: 92.8% S.

Experimental example 21

Experimental example 11 was repeated except that acetonitrile was used as a solvent in the total volume of 10V instead of 1-propanol, 0.25eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 30.0 percent; enantiomeric purity: 96.0% S.

Experimental example 22

Experimental example 11 was repeated except that acetonitrile was used as a solvent in place of 1-propanol in a total volume of 4.5V, 0.50eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 24.7 percent; enantiomeric purity: 99.2% S.

Experimental example 23

Experimental example 11 was repeated except that a mixture of 1-propanol and dichloromethane (50:50) in a total volume of 2V was used instead of 1-propanol as a solvent ((+) -O, O ' -di-p-methylbenzoyl- (S, S) -tartaric acid and 4- [4- (dimethylamino) -1- (4 ' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile free base was dissolved in 4.5V of dichloromethane, 3.5V of dichloromethane was distilled off and 1V of 1-propanol was added), 0.25eq of (+) -O, O ' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 18.5 percent; enantiomeric purity: 96.9% S.

Experimental example 24

Experimental example 11 was repeated except that a mixture of 1-propanol and dichloromethane (95:5) was used as a solvent in a total volume of 4.5V, 0.35eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 35.7 percent; enantiomeric purity: 78.8% S.

Experimental example 25

Experimental example 11 was repeated except that 1-propanol and methylene chloride (85:15) in a total volume of 4.5V were used as solvents instead of 1-propanol, 0.4eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 31 percent; enantiomeric purity: 98.2% S.

Experimental example 26

Experimental example 11 was repeated except that a mixture of 1-propanol and dichloromethane (50:50) was used as a solvent in a total volume of 4.4V, 0.5eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 16.4 percent; enantiomeric purity: 98.9% S.

Experimental example 27

Experimental example 11 was repeated except that a mixture of 1-propanol and dichloromethane (75:25) was used as a solvent in a total volume of 4.5V, 0.5eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 34.2 percent; enantiomeric purity: 98.8% S.

Experimental example 28

Experimental example 11 was repeated except that the crystallization mixture contained no toluene, a mixture of 1-propanol and dichloromethane (85:15) in a total volume of 4.5V was used instead of 1-propanol as the solvent, 0.5eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used, and the retention time before filtration was overnight.

Molar yield: 37.8 percent; enantiomeric purity: 98.8% S.

Experimental example 29

Experimental example 11 was repeated except that a mixture of 1-propanol and dichloromethane (85:15) was used as a solvent in a total volume of 4.5V, 0.5eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 36.6 percent; enantiomeric purity: 97.6% S.

Experimental example 29

Experimental example 11 was repeated except that a mixture of 1-propanol and dichloromethane (90:10) was used as a solvent in a total volume of 4.5V, 0.5eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight. The results of this experiment were performed twice as follows:

molar yield: 38.9 percent; enantiomeric purity: 97.7% S.

Molar yield: 35.8 percent; enantiomeric purity: 98.5% S.

Experimental example 30

Experimental example 11 was repeated except that a total volume of 6.0V of a mixture of 1-propanol and dichloromethane (92.5:7.5) was used instead of 1-propanol as the solvent, 0.5eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight. The results of this experiment were performed twice as follows:

molar yield: 35.1 percent; enantiomeric purity: 98.6% S.

Molar yield: 39.0 percent; enantiomeric purity: 81.3% S.

Experimental example 31

Experimental example 11 was repeated except that a mixture of 1-propanol and methylene chloride (95:5) was used as a solvent in a total volume of 4.5V, 0.5eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was 0.5 hours.

Molar yield: 35.0 percent; enantiomeric purity: 98.4% S.

Experimental example 32

Experimental example 11 was repeated except that a mixture of 1-propanol and dichloromethane (90:10) was used as a solvent in a total volume of 4.6V, 0.6eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 38.5 percent; enantiomeric purity: 99.1% S.

Experimental example 33

Experimental example 11 was repeated except that a mixture of 1-propanol and acetonitrile (15:85) was used as a solvent in a total volume of 4.5V, 0.5eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was 0.5 hours.

Molar yield: 25.9 percent; enantiomeric purity: 99.2% S.

Experimental example 34

Experimental example 11 was repeated except that a mixture of 1-propanol and acetonitrile (85:15) was used as a solvent in a total volume of 4.5V, 0.5eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 18.5 percent; enantiomeric purity: 99.4% S.

Experimental example 35

Experimental example 11 was repeated except that a mixture of 1-propanol and acetonitrile (90:10) was used as a solvent in a total volume of 4.5V, 0.5eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was overnight.

Molar yield: 29.9 percent; enantiomeric purity: 99.3% S.

Experimental example 36

Experimental example 11 was repeated except that a mixture of 1-propanol and ethyl acetate (31:69) was used as a solvent in a total volume of 4.5V instead of 1-propanol and additional 2V of 1-propanol was added thereto, 0.25eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was 0.5 hour.

Molar yield: 28.6 percent; enantiomeric purity: 98.4% S.

Experimental example 37

Experimental example 11 was repeated except that a mixture of 1-propanol and ethanol (50:50) was used as a solvent in a total volume of 4.4V, 0.5eq of (+) -O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid was used and the retention time before filtration was 0.5 hours.

Molar yield: 27.4 percent; enantiomeric purity: 99.4% S.

Experimental example 38

A series of experiments were performed with (+) - (S, S) -DTT to examine the diol resolution. The general procedures are described below, and specific descriptions and results for each reaction are shown in table 1.

Racemic diol (20g, 58.4mmol) was dissolved in approximately one at 40 deg.CHalf of the solvent was used for the experiment. Will (+) - (S, S) -DTT.H₂O (the amount is detailed in the table) is added as a solution to the other half of the solvent. The solution was kept at 40 ℃ and seeded with crystals of (S) -diol 1/2(+) - (S, S) -DTT (approximately 5mg) over 2 minutes. Crystallization typically begins within 5-10 minutes after seeding. After 2 hours at 40 ℃, the temperature of the solution was reduced to 20 ℃ over 2 hours and the solution was kept at this temperature for a further 1 hour. The product was then isolated by filtration, washed with the appropriate solvent (2 × 20ml) and dried under reduced pressure at 60 ℃ overnight.

Table 1: results of Experimental example 38

Experimental number	Solvent or solvent mixture (mixture is expressed as v/v%)	Total volume of solvent used (mL)	DTT equivalent used	Yield (%)	Ratio (S/R)
						38a	1-propanol	60	0.25	16.7	96.0/4.0
38b	1-propanol	86	0.39	19.9	97.0/3.0
						38c	1-propanol	90	0.5	26	77.3/22.7
38d	1-propanol	90	0.68	15.8	98.4/1.6
						38e	1-propanol	120	0.75	11.5	96.6/3.4
38f	Acetonitrile	200	0.25	94	91.8/82
						38g	Acetonitrile	90	0.5	17.2	78.7/21.3
38h	1-propanol/acetonitrile (15/85)	90	0.5	14.2	99.3/0.7
						38i	1-propanol/acetonitrile (85/15)	90	0.5	98	99.0/1.0
38j	1-propanol/acetonitrile (90/10)	90	0.5	13.9	99.4/0.6
						38k	1-propanol/ethyl acetate (31/69)	90	0.25	15.1	94.0/6.0
38l	1-propanol/ethanol (50/50)	90	0.5	15.4	99.3/0.7
						38m	1-propanol/DCM (50/50)	40	0.25	11.7	96.0/4.0
38n	1-propanol/DCM (85/15)	90	0.4	34.6	98.6/1.4
						38o	1-propanol/DCM (75/25)	90	0.5	26.6	98.4/1.6
38p	1-propanol/DCM (85/15)	90	0.5	33.7	98.8/1.2
						38q	1-propanol/DCM (90/10)	90	0.5	358	99.3/0.7
38r	1-propanol/DCM (92.5/7.5)	120	0.5	37.6	99.0/1.0
						38s	1-propanol/DCM (95/5)	90	0.5	36.6	99.4/0.6

38t	1-propanol/DCM (90/10)	90	06	296	99 1/0 9
						38u	ethanol/DCM (50/50)	60	05	0	n/a
38v	ethanol/DCM (75/25)	90	05	07	96 5/3 5
						38w	ethanol/DCM (85/15)	90	05	99	98 8/1 2
38x	Ethanol	100	05	207	99 6/0 4

Although all the above experiments were carried out using (+) O, O ' -di-p-methylbenzoyl- (S, S) -tartaric acid, which precipitates with S-4- [4- (dimethylamino) -1- (4 ' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile leaving a mother liquor rich in R-4- [4- (dimethylamino) -1- (4 ' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile, it will be clear to the skilled person that (-) -O, O ' -di-p-methylbenzoyl- (R, R) -tartaric acid, which precipitates with R-4- [4- (dimethylamino) -1- (4 ' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile are precipitated together, leaving a mother liquor rich in S-4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile.

Although the standard patterns added are: adding (+) -O, O '-di-p-methylbenzoyl- (S, S) -tartaric acid to 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile; the addition procedure can be reversed (4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile added to (+) O, O' -di-p-methylbenzoyl- (S, S) -tartaric acid).

Claims (22)

1. A process for the resolution of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile in the form of a racemic or non-racemic mixture of enantiomers into its isolated enantiomers, said process comprising reacting the compound of formula (I) with the (+) - (S, S) -or (-) - (R, R) -enantiomer of O, O' -di-p-methylbenzoyltartaric acid, a step of fractional crystallization of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile in a solvent system comprising 1-propanol, ethanol or acetonitrile.

2. A process as claimed in claim 1, characterized in that not more than 1mol, in particular not more than 0.5mol, of the (+) - (S, S) or (-) - (R, R) -enantiomer of O, O '-di-p-methylbenzoyltartaric acid is used per mol of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile.

3. The process as claimed in claim 1 or 2, characterized in that (+) O, O' -di-p-methylbenzoyl (S, S) tartaric acid is used.

4. The process as claimed in claim 1 or 2, characterized in that (-) O, O' -di-p-methylbenzoyl (R, R) tartaric acid is used.

5. The process according to any one of claims 1 to 5, characterized in that 1-propanol is the main component of the solvent system.

6. The process according to any one of claims 1 to 5, characterized in that ethanol is the main component of the solvent system.

7. The process of any one of claims 1-5, characterized in that acetonitrile is the main component of the solvent system.

8. The process according to any one of claims 1 to 7, characterized in that the solvent system comprises one or more organic co-solvents, in particular selected from the group consisting of toluene, diethyl ether, ethyl acetate and dichloromethane, more particularly selected from toluene.

9. The process of any one of claims 1 to 8, wherein the solvent system comprises water.

10. The process of any one of claims 1 to 9, characterized in that the solvent system comprises an achiral acid which under the present conditions is capable of protonating 4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile without precipitating 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile in salt form.

11. The process according to claim 10, characterized in that the achiral acid is selected from the group consisting of organic acids such as formic acid, acetic acid, trifluoroacetic acid and methanesulfonic acid, more particularly from acetic acid.

12. A process according to any one of claims 1 to 11 characterised in that the solvent system is cooled, together with dissolved 4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile and (+) - (S, S) -or (-) - (R, R) -O, O' -di-p-methylbenzoyl tartaric acid, from a first temperature in the range of 20 ℃ to the reflux temperature of the solvent system to a second temperature in the range of 0 ℃ to 40 ℃.

13. The process of claim 12, wherein the mixture of 4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile, (+) - (S, S) -or (-) - (R, R) -O, O' -di-p-methylbenzoyl tartaric acid and the solvent system is maintained at the first temperature for a period of time within the range of 0 to 4 hours prior to cooling.

14. The process of claim 12 or 13, wherein the mixture of 4- [4- (dimethylamino) -1- (4 '-fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile, (+) - (S, S) -or (-) - (R, R) -O, O' -di-p-methylbenzoyl tartaric acid and the solvent system is seeded with the crystallization of the desired salt at the first temperature or during cooling.

15. The method of any one of claims 12-14, wherein the cooling is completed within 8 hours.

16. The process as claimed in any of claims 1 to 15, characterized in that the precipitated salt is separated from the mother liquor within 8 hours after the start of the precipitation.

17. The process as claimed in any of claims 1 to 16, characterized in that the separated salt is washed within 4 hours.

18. The process according to any of claims 1-17, characterized in that the isolated salt is reslurried or recrystallized one or more times in a solvent system comprising 1-propanol or ethanol by heating to a temperature in the range of 30 ℃ to the reflux temperature of the solvent, more particularly to 40 ℃ to 60 ℃, followed by cooling to a temperature in the range of 0 ℃ to 40 ℃.

19. A process for the preparation of escitalopram comprising a process according to any one of claims 1 to 18.

20. The method of claim 19, further comprising stereoselectively converting one of the isolated enantiomers of 4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile to escitalopram.

21. The method of claim 20, wherein S-4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile is stereoselectively converted to escitalopram.

22. A process according to claim 21, wherein S-4- [4- (dimethylamino) -1- (4' -fluorophenyl) -1-hydroxybutyl ] -3- (hydroxymethyl) -benzonitrile is reacted with a reactive acid derivative, such as an acid chloride or anhydride, in particular with methanesulfonyl chloride or toluenesulfonyl chloride, in the presence of a base, such as triethylamine or pyridine.