ASYMMETRIC CATALYTIC REDUCTION OF OXCARBAZEPINE

Abstract

A process for preparing (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide or (R)-()-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide, by reduction of oxcarbazepine in the presence of a catalyst and a hydride source is disclosed. The catalyst is prepared from a combination of [RuX.sub.2(L)].sub.2 wherein X is chlorine, bromine or iodine, and L is an aryl or aryl-aliphatic ligand, with a ligand of formula (A) or formula (B):

##STR00001##

wherein R.sup.1 is chosen from C.sub.1-6 alkoxy and C.sub.1-6 alkyl, n is a number from 0 to 5, and when n is a number from 2 to 5, R.sup.1 can be the same or different, and R.sup.2 is alkyl, substituted alkyl, aryl, substituted aryl, alkaryl or substituted alkaryl. The hydride source is either NR.sup.3R.sup.4R.sup.5 and formic acid, [R.sup.3R.sup.4R.sup.5NH][OOCH] and optionally formic acid, or [M][OOCH].sub.x and formic acid, wherein R.sup.3, R.sup.4 and R.sup.5 are C.sub.1-6 alkyl, M is an alkali metal or alkaline earth metal and x is 1 or 2. A pH from 6.5 to 8 is maintained during the process.

Claims

1. A process for preparing (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide or (R)-()-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide, by reduction of oxcarbazepine in the presence of a catalyst and a hydride source, wherein the catalyst is prepared from a combination of [RuX.sub.2(L)].sub.2 wherein X is chlorine, bromine or iodine, and L is an aryl or aryl-aliphatic ligand, with a ligand of formula (A) or formula (B): ##STR00010## wherein R.sup.1 is chosen from C.sub.1-6 alkoxy and C.sub.1-6 alkyl, n is a number from 0 to 5, and when n is from 2 to 5, R.sup.1 can be the same or different, and R.sup.2 is alkyl, substituted alkyl, aryl, substituted aryl, alkaryl or substituted alkaryl; wherein the hydride source is chosen from NR.sup.3R.sup.4R.sup.5 and formic acid, or [R.sup.3R.sup.4R.sup.5NH][OOCH] and optionally formic acid, or [M][OOCH].sub.x and formic acid, wherein R.sup.3, R.sup.4 and R.sup.5 are C.sub.1-6 alkyl, M is an alkali metal or alkaline earth metal and x is 1 or 2, and wherein during the process a pH from 6.5 to 8 is maintained.

2. A process according to claim 1, wherein X is chlorine.

3. A process according to claim 1 or claim 2, wherein L is p-cymene.

4. A process according to any preceding claim, wherein n is 1 and R.sup.1 is a methoxy group or a methyl group.

5. A process according to claim 4, wherein n is 1 and R.sup.1 is a methoxy group or a methyl group in the para position.

6. A process according to any preceding claim wherein the ligand of formula (A) or foliuula (B) is (S,S)-N-(4-toluenesulfonyl)-di(methoxyphenyl)ethylenediamine, ((S,S)-TsDAEN), or (R,R)-N-(4-toluenesulfonyl)-di(methoxyphenyl)ethylenediamine, ((R,R)-TsDAEN).

7. A process according to any preceding claim, wherein the molar ratio of oxcarbazepine to the ruthenium catalyst is at least 500:1

8. A process according to claim 7, wherein the molar ratio of oxcarbazepine to the ruthenium catalyst is at least 1500:1.

9. A process according to claim 7, wherein the molar ratio of oxcarbazepine to the ruthenium catalyst is at least 2500:1.

10. A process according to any preceding claim, wherein the hydride source is NR.sup.3R.sup.4R.sup.5 and formic acid.

11. A process according to claim 10, wherein less than two equivalents of NR.sup.3R.sup.4R.sup.5 and less than 1 equivalent of foiiuic acid are added to the reaction mixture at the start of the process.

12. A process according to any one of claims 1 to 9, wherein the hydride source is [R.sup.3R.sup.4R.sup.5NH][OOCH] and optionally formic acid.

13. A process according to claim 10, wherein less than two equivalents of [R.sup.3R.sup.4R.sup.5NH][OOCH] and less than 0.5 equivalents of formic acid are added to the reaction mixture at the start of the process.

14. A process according to any one of claims 1 to 9, wherein the hydride source is [M][OOCH].sub.x and formic acid.

15. A process according to claim 14, wherein less than two equivalents of [M][OOCH].sub.x and less than 1 equivalent of formic acid are added to the reaction mixture at the start of the process.

16. A process according to any preceding claim wherein R.sup.3, R.sup.4 and R.sup.5 are ethyl, propyl or butyl.

17. A process according to any preceding claim, wherein the pH is maintained from 6.5 to 8 by adding formic acid in a controlled manner during the course of the reaction.

18. A process according to claim 17, wherein the pH is maintained from 7.0 to 7.8 by adding formic acid in a controlled manner during the course of the reaction.

19. A process according to any preceding claim, wherein the reduction takes place in a solvent, wherein the solvent comprises at least one polar aprotic solvent.

20. A process according to claim 19, wherein the solvent comprises dimethylformamide or acetonitrile.

21. A process according to claim 10 claim 11, claim 14 or claim 15, wherein the reduction takes place in a solvent consisting of 0-25% DMF, 0-25% water and 75-95% EtOAc or 0-25% acetonitrile, 0-25% water and 75-95% EtOAc.

22. A process according to claim 12 or claim 13, wherein the reduction takes place in a solvent consisting of 5-25% DMF and 75-95% EtOAc, 5-25% acetonitrile and 75-95% EtOAc, 5-25% DMF and 75-95% water, or 5-25% acetonitrile and 75-95% water.

23. A process according to any preceding claim wherein the process is carried out under reflux conditions.

24. A process according to any preceding claim, wherein the reduction takes place in the presence of a phase transfer catalyst and the phase transfer catalyst is a quaternary alkyl ammonium halide.

25. A process according to any preceding claim, wherein the (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide or (R)-()-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide is isolated either by addition of methyl tert-butyl ether (MTBE) and filtration, or by precipitation from either methanol/water or methanol/MTBE at 0-5 C.

26. A process according to any one of claims 1 to 24, wherein the (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide or (R)-()-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide is precipitated by removing the reaction solvent while adding water to maintain the reaction volume at a substantially constant level.

27. A process according to claim 26, wherein the precipitated (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide or (R)-()-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide is isolated by filtration followed by reslurrying in ethyl acetate and filtration.

28. A process according to any preceding claim wherein the (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide or (R)-()-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide has an optical purity in the range 92-100%.

29. A process for preparing an compound of formula (C) or (D) ##STR00011## wherein R.sup.6 is hydrogen, alkyl, halogenalkyl, aralkyl, cycloalkyl, cycloalkyalkyl, alkoxy, aryl or pyridyl; comprising a first step, which is a process for the production of (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide or (R)-()-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide according to any preceding claim, and a second step, wherein the (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide or (R)-()-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide is acylated.

30. A process for preparing (S)-()-10-acetoxy-10,11-dihydro-5H-dibenz/b,f/azepine-5-carboxamide comprising a first step, which is a process for the production of (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide according to any one of claims 1 to 28, and a second step wherein the (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide is acylated.

31. A process for preparing (R)-(+)-10-acetoxy-10,11-dihydro-5H-dibenz/b,f/azepine-5-carboxamide comprising a first step, which is a process for the production of (R)-()-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide according to any one of claims 1 to 28, and a second step wherein the (R)-()-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide is acylated.

Description

EXAMPLE 1

Asymmetric Reduction of Oxcarbazepine Using NEt.SUB.3 .and pH-Controlled Addition of HCOOH

[0070] In a 2 L 5-necked round bottom flask, oxcarbazepine was charged (635 mmol, 169 g), the flask was fitted with two water reflux condensers connected to a Schlenk line (two reflux condensers were used to secure two escape routes for CO.sub.2 and H.sub.2 gases evolving from the reaction), a burette for titration and a Hamilton gel-filled electrode fitted through a hollow GL25 screw cap equipped with a PTFE/silicone ring. To the starting material under N.sub.2 flow, EtOAc (480 mL, undegassed, HPLC grade), H.sub.2O (48 mL, undegassed, HPLC grade) and Et.sub.3N (1.1 eq., 699 mmol, 97.5 mL, undegassed, Fluka, 99.9% pure) were added with the aid of a graduated cylinder. The catalyst (formed separately in situ in a 50 mL Schlenk tube under N.sub.2 flow by stirring [RuCl.sub.2(p-cymene)].sub.2 (0.1588 mmol, 97.2 mg) and (S,S)-TsDAEN (2.2 eq. with respect to the metal dimer precursor, 0.3493 mmol, 159 mg) in DMF (13 mL, degassed, anhydrous) at room temperature for 10-15 min. was injected. The Schlenk tube was rinsed with small portions of the remaining DMF (57 mL) and injected to the reaction mixture. The solvent combination at this point was 10% DMF-10% H.sub.2O-80% EtOAc (v/v/v) and the substrate concentration before titration was 1.1M. The round bottom flask was placed in an oil bath preheated at 105 C. and the reaction mixture magnetically stirred at reflux (T.sub.oil bath=105 C., internal T=72-77 C.). Once the reaction mixture started refluxing, the reaction pH was approximately 8.8. At this point the titration/slow addition with 12.5 M HCOOH solution in 20% DMF/EtOAc was started. The pH was slowly brought to 7.4 and then maintained constant at this value over 12 hrs by slowly adding the HCOOH solution. The HPLC conversion after 15 hrs was 99%. Further stirring up to 20 hrs resulted in the formation of a white precipitate and has not consumed the remaining 1% oxcarbazepine. A total of approximately 3.7 eq. HCOOH with respect to the starting material was consumed during the reaction. The overall substrate/product concentration at the end of the reaction was 0.86 M. After 20 hrs, the heating was stopped and the reaction mixture was stirred and allowed to cool down slowly. When the temperature of the oil bath reached approximately 80 C., 500 mL MTBE was added to the reaction mixture and allowed to cool to room temperature under stirring. The reaction mixture was stirred at 0-5 C. for approximately 30 min, filtered and the precipitate was washed repeatedly with cold portions of MTBE until the filtrate was colourless. The resulting white precipitate was dried in the air, then under high vacuum, affording a white powder: 95% isolated yield (152 g). HPLC: 99.6% product, 97.8% e.e., 0.4% oxcarbazepine.

[0071] Due to the fact that under these reaction conditions (10% DMF-10% H.sub.2O-80% EtOAc, 1.1eq. Et.sub.3N and HCOOH) the product crystallises out at the refluxing temperature, no solvent was evaporated during the work-up. The very low solubility of the product in MTBE, allows not only further precipitation of the starting material, but also it aids the purification/removal of residual ruthenium, DMF and reagents by washing the filtrate with copious amounts, without a loss in the isolated yield. The ruthenium level in the product was between 5-50 ppm.

EXAMPLE 2

Asymmetric Reduction of Oxcarbazepine Using [Et.SUB.3.NH][OOCH] and pH-Controlled Addition of HCOOH

[0072] In a 500 mL 4-necked round bottom flask, oxcarbazepine was charged (159 mmol, 40 g), the flask fitted with a water reflux condenser connected to a Schlenk line, a burette for titration and a Hamilton gel filled electrode fitted through a hollow GL25 screw cap with a PTFE/silicone ring. The flask was flushed with N.sub.2 for approximately 30 min. To the starting material under N.sub.2 flow, EtOAc (78 mL, degassed, anhydrous), [Et.sub.3NH][OOCH] conunercially available from Fluka (1.07 eq., 170 mmol, 25 mL, tuidegassed, Fluka) were added via a syringe. The catalyst (formed separately in situ in a 20 mL Schlenk tube by stirring. [RuCl.sub.2(p-cymene)].sub.2 (0.0265 mmol, 16.2 mg) and (S,S)-TsDAEN (2.2 eq. with respect to the metal dimer precursor, 0.0582 mmol, 25 mg) in DMF (5 mL, degassed, anhydrous) at room temperature for 10-15 min. was injected. The Schlenk tube was rinsed with small portions of the remaining DMF (53 mL) and injected to the reaction mixture. The solvent combination at this point was 20% DMF-80% EtOAc (v/v) and the oxcarbazepine concentration before titration was 1.3M. The round bottom flask was placed in an oil bath preheated at 105 C. and the reaction mixture was magnetically stirred at reflux (T.sub.oil bath=105 C.). Once the reaction mixture started refluxing, the reaction pH was approximately 6.8. The reaction mixture slowly turned purple and the pH started increasing as HCOOH from the triethylammonium formate was consumed. When the pH reached 7.4-7.45 the titration/slow addition with 12.5 M HCOOH solution in 20% DMF/EtOAc was started. The pH was maintained at pH=7.4 over 12 hrs by slowly adding the HCOOH solution. After 17 hrs the reaction mixture was clear purple, with some catalyst decomposition observed on the walls of the flask. The HPLC conversion after 17 hrs was 98%. At this point the pH of the solution was 7.8 and addition of more HCOOH solution was continued at 7.7. Further stirring up to 23 hrs led to 99% conversion. Approximately 4.7 eq. HCOOH with respect to the starting material was consumed during the reaction. After 23 hrs, the heating was stopped and the reaction mixture was stirred and allowed to cool down. The reaction mixture was concentrated, 100 mL MTBE added and the solvent removed again. 15 mL MeOH was added and the white paste refluxed for about 5 min and then 250 mL MTBE added slowly to this refluxing mixture. The resulting mixture was stirred at reflux for 30 min, cooled to RT, then to 0-5 C. and stirred for 30 min. The mixture was filtered cold and washed with cold portions of MTBE until the filtrate was colourless (850 mL). The resulting white precipitate was dried in the air, then under high vacuum, affording a white powder: 94% isolated yield (37.9 g); HPLC: 99.5% product, 97.8% e.e., 0.5% oxcarbazepine. The ruthenium level in the product was between 5-50 ppm.

Comparison 1a: Asymmetric Reduction with and Without pH Control (Hydride Source is NEt.SUB.3 .and HCOOH)

[0073] Reactions were carried out using a method similar to that of example 1. The substrate/catalyst ratio was 2000 and the solvent was 20% H.sub.2O/EtOAc. The ligand was (S,S)-Ts-DAEN. In example 3, 1 eq. NEt.sub.3 and 1 eq. HCOOH were added to the reaction mixture at the start of the reaction. Further HCOOH was added throughout the course of the reaction to maintain a pH of 7.4. In comparative example 1, 4.4 eq. Et.sub.3N and 4 eq. HCOOH were pre-mixed in H.sub.2O and added to the reaction mixture in EtOAc at the beginning of the reaction. Table 1 shows the results of example 3 and comparative example 1:

TABLE-US-00001 TABLE 1 Time Alcohol e.e Hydride source (h) (%) (%) Example 3 1eq. of NEt.sub.3 and 1eq. of 2 22 98.8 HCOOH added at start of 4 33 98.7 reaction. Further 1.7eq. of 6 46 98.8 HCOOH added during the 8 57 98.5 course of the reaction, 26 82 98.3 maintaining the pH at 7.4. 33 82 98.3 Comparative Mixture of 4.4eq. of NEt.sub.3 and 21 31 96.8 Example 1 4eq. HCOOH in H.sub.2O/EtOAc added at the start of the reaction.

[0074] The yield of the pH controlled reaction (example 3) was much better than the yield of the reaction in which pH was not controlled (comparative example 1), despite the fact that higher quantities of hydride source reagents were used.

Comparison 1b: Asymmetric Reduction with and Without pH Control (Hydride Source is [Et.SUB.3.NH][OOCH] or [Et.SUB.3.NH][OOCH] and HCOOH)

[0075] Reactions were carried out using a method similar to that of example 2. The substrate/catalyst ratio was 1000 and the solvent was 10% DMF/EtOAc. The ligand was Ts-DPEN rather than Ts-DAEN. 20 g of oxcarbazepine were used instead of 40 g. [Et.sub.3NH][OOCH] was added to the reaction mixture at the start of the reaction and no further HCOOH was added. Table 2 shows how the results of two comparative examples using different quantities of [Et.sub.3NH][OOCH]:

TABLE-US-00002 TABLE 2 [Et.sub.3NH][OOCH] Time Conver- Enantiomeric (eq) (h) pH sion (%) excess (%) Comparative 5 0 6.95 0 Example 2 22 8.6 100 97.7 Comparative 2 0 6.7 Example 3 0.15 6.8 0.5 7.55 1.5 8.25 2 8.86 3 8.85 5 8.65 6 8.6 19 8.22 47 98.5
5 equivalents of the expensive reagent [Et.sub.3NH][OOCH] provided 100% conversion after 22 hours, whereas less than 50% conversion was achieved with only 2 equivalents of the reagent. The pH of the reaction mixture increased during the course of the reaction.

[0076] The reaction was repeated on a 10 g scale using a substrate/catalyst ratio of 1500:1 and a 20% DMF/EtOAc solvent. In comparative examples 4 and 5, 5 equivalents of [Et.sub.3NH][OOCH] were used and no further HCOOH was added. In example 4, only one equivalent of [Et.sub.3NH][OOCH] was used but further HCOOH was added during the course of the reaction to maintain the pH at 7.4. Table 3 shows the results of comparative examples 4 and 5 and example 4:

TABLE-US-00003 TABLE 3 Time Conver- Enantiomeric Ligand Hydride Source (h) sion (%) Excess (%) Comparative (S,S)- 5 eq. 25 98 97 Example 4 TsDPEN [Et.sub.3NH][OOCH] Comparative (S,S)- 5 eq. 20 97 96.9 Example 5 TsDAEN [Et.sub.3NH][OOCH] Example 4 (S,S)- 1 eq. 7 99 97.8 TsDAEN [Et.sub.3NH][OOCH] plus HCOOH to maintain pH 7.4
A comparison of comparative examples 4 and 5 with example 4 shows that by controlling the pH by addition of HCOOH, a much smaller quantity of the expensive reagent [Et.sub.3NH][OOCH] can be used, and the reaction reaches almost complete conversion in 7 hours rather than 20-25 hours.

Comparison 2: Asymmetric Reduction Using a Variety of Ligands

[0077] The activity of catalysts comprising the (S,S)-TsDAEN and (S,S)-TsDPEN ligands was compared at a 20-40 g scale. The catalysts were generated in situ by stirring [RuCl.sub.2(p-cymene)].sub.2 and either (S,S)-TsDAEN or(S,S)-TsDPEN for 5-10 minutes in DMF before addition to oxcarbazepine and 1.07 equivalents of [Et.sub.3NH][OOCH]. 12.5M HCOOH solution in 20% DMF/EtOAc was injected slowly at pH=7.4. The ratio of oxcarbazepine to catalyst was 3000:1. Table 4 shows the results for Examples 5 and 6 (using (S,S)-TsDAEN) and example 7 (using (S,S)-TsDPEN):

TABLE-US-00004 TABLE 4 Time Conver- Enantiomeric Scale (g) ligand (h) sion (%) excess (%) Example 5 40 (S,S)- 23 99 98.1 TsDAEN Example 6 25 (S,S)- 4 55 98.2 TsDAEN 6 69 98.3 8 77 98 26 97 97.9 Example 7 20 (S,S)- 5 47 98.4 TsDPEN 6 54 98.2 7 64 98.3 8 77 98 23 83 98 26 85 97.8 30 86 97.8
The (S,S)-TsDAEN examples show significantly better conversion than the (S,S)-TsDPEN example and show similar enantioselectivity.

Comparison 3: Asymmetric Reduction of Oxcarbazepine Using a Phase Transfer Catalyst

[0078] Table 5 shows the results of three asymmetric reduction reactions wherein a phase transfer catalyst was used in addition to the ruthenium catalyst. In each reaction, the catalyst was generated in situ by adding [RuCl.sub.2(p-cymene)].sub.2 and a ligand to EtOAc and stirring. (NB (R,R)-TsDTEN has the same structure as (R,R)-TsDPEN except that the phenyl groups are substituted by tolyl groups). The phase transfer catalyst was 0.1 equivalents of Bu.sub.4NBr. The hydride source was 2 equivalents of [Et.sub.3NH][OOCH] and additional formic acid was slowly added to the reaction mixture during the course of the reaction. The ratio of oxcarbazepine to catalyst was 2000:1 and the external reaction temperature was 110 C. The conversion in example 9, wherein the ligand was (R,R)-TsDAEN was significantly better than the conversion in examples 8 and 10, wherein the ligands were (S,S)-TsDPEN and (R,R)-DTEN.

TABLE-US-00005 TABLE 5 Time Conver- Enantiomeric Ligand (h) sion (%) excess (%) Example 8 (S,S)-TsDPEN 6 58 .sup.98 (S) Example 9 (R,R)-TsDAEN 6 70 98.2(R) Example 10 (R,R)-TsDTEN 6 60 97.8 (R)

EXAMPLE 3

Acetylation of (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide

[0079] (S)-(+)-10,11-dihydro-10-hydroxy-5H-dibenz/b,f/azepine-5-carboxamide (500 g), obtained via asymmetric transfer hydrogenation as described above, and 4-(N,N-dimethylamino)pyridine (4 g) were suspended in dichloromethane (5.07 L). Pyridine (210 mL) was added to the suspension. The reaction mixture was heated to reflux whereupon acetic anhydride (240 mL) was added dropwise. The resulting yellowish-brown solution was stirred for 2 hours and then cooled to 30 C.

[0080] The reaction mixture was then quenched by the addition of sulphuric acid. After stirring for 10 min, the layers were separated. The organic layer was washed twice with saturated aqueous sodium bicarbonate solution and then water. Approximately half of the dichloromethane was then removed by evaporation and isopropanol (5 L) was added to the mixture which was then left to stand overnight. Further solvent was evaporated (approximately 1.5 L) and the resulting slurry was cooled to approximately 3 C. After 3 hours the solid was filtered off, washed with cold isopropanol and then dried under vacuum overnight. The dried solid was suspended in isopropanol (6.5 L) and the resulting white sluny was heated to reflux. Once a solution was obtained heating was stopped and the reaction mixture was stirred for 1 hr at 1-5 C. Solids were isolated by filtration, washed with cold isopropanol and dried under vacuum to yield 524.2 g of white solid, 90% yield, 99.96% chemical purity, (R)-isomer below the limit of detection.

[0081] Residual ruthenium content was found to be less than 2 ppm. According to the regulatory guidelines, the oral concentration limit is 5 ppm.

EXAMPLE 4

Asymmetric Reduction of Oxcarbazepine Using Higher Oxcarbazepine:Catalyst Ratio

[0082] This reduction described in example 1 was carried out on oxcarbazepine (357 mmol, 90 g) using [RuCl2(p-cymene)]2 (0.066 mmol, 40.4 mg) and (S,S)-TsDAEN (0.145 mmol, 61.9 mg) using four times the quantity of water. The reaction was complete in 27 hours.

[0083] Ethyl acetate was distilled from the batch while maintaining the original batch volume by the addition of water (dropwise). Temperature was maintained above 60 C. during the distillation. Approximately into the distillation the product started to precipitate.

[0084] The mixture was cooled to 5 C., held at that temperature for one hour and then filtered. The filter cake was washed with water. The wet cake was then reslurried in ethyl acetate (350 mL) and heated to reflux for 0.5 hours. It was then cooled to 5 C. and held at that temperature for 1 hour. The mixture was then filtered and the recovered solids with ethyl acetate (120 mL). Drying under high vacuum afforded an off-white powder: 88% isolated yield (79.8 g): HPLC: 99.8% product, 98.4% e.e., 0.09% oxcarbazepine

[0085] It will be appreciated that the invention described above may be modified.