Synthesis of tipifarnib
11542244 · 2023-01-03
Assignee
Inventors
Cpc classification
International classification
Abstract
Provided herein are methods of preparing a desired enantiomer 6-[amino(4-chlorophenyl)(1-methyl-1H-imidazol-5-yl) methyl]-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone, otherwise known as tipifarnib.
Claims
1. A method for preparing a desired enantiomer of tipifarnib comprising the steps of: (i) obtaining a starting material comprising tipifarnib that is not enantiopure in the desired enantiomer; (ii) transforming the starting material from step (i) to a racemic mixture of tipifarnib; and (iii) recovering the desired enantiomer of tipifarnib from the racemic mixture of tipifarnib of step (ii); wherein step (ii) comprises the steps of: (ii)(a) reacting the starting material with sodium nitrite in a reaction solvent to give a product mixture; (ii)(b) recovering a racemic alcohol of Formula (II) from the product mixture of step (ii)(a); and ##STR00026## (ii)(c) transforming the racemic alcohol of step (ii)(b) to the racemic mixture of tipifarnib.
2. The method of claim 1, wherein the reaction solvent of step (ii)(a) is an organic solvent, water, or a mixture thereof.
3. The method of claim 2, wherein the organic solvent is miscible with water.
4. The method of claim 1, wherein step (ii)(a) takes place in the presence of an additive.
5. The method of claim 1, wherein step (ii)(b) comprises: adjusting the product mixture's pH with a base; extracting the product mixture with an extraction solvent; and crystallizing the racemic alcohol.
6. A method for preparing a desired enantiomer of tipifarnib comprising the steps of: (i) obtaining a starting material comprising tipifarnib that is not enantiopure in the desired enantiomer; (ii) transforming the starting material from step (i) to a racemic mixture of tipifarnib; and (iii) recovering the desired enantiomer of tipifarnib from the racemic mixture of tipifarnib of step (ii), wherein step (ii) comprises the steps of: (ii)(a) heating the starting material in a reaction solvent to give a product mixture; (ii)(b) recovering a racemic alcohol of Formula (II) from the product mixture of step (ii)(a); and ##STR00027## (ii)(c) transforming the racemic alcohol of step (ii)(b) to the racemic mixture of tipifarnib.
7. The method of claim 6, wherein the reaction solvent of step (ii)(a) is an organic solvent, water, or a mixture thereof.
8. The method of claim 7, wherein the organic solvent is miscible with water.
9. The method of claim 6, wherein step (ii)(a) takes place in the presence of an additive.
10. The method of claim 6, wherein step (ii)(b) comprises: adjusting the product mixture's pH with a base; extracting the product mixture with an extraction solvent; and crystallizing the racemic alcohol.
11. The method of claim 1, wherein step (iii) comprises: (iii)(a) crystallizing the desired enantiomer of tipifarnib from the racemic mixture of tipifarnib in the presence of a chiral resolving agent; and (iii)(b) separating crystals of the desired enantiomer of tipifarnib from a mother liquor.
12. The method of claim 11, wherein the method further comprises: (iv) recycling the mother liquor of step (iii)(b) to be used as the starting material in step (i).
13. The method of claim 12, wherein the steps (i) to (iv) are run in multiple cycles.
14. The method of claim 1, wherein the desired enantiomer of tipifarnib is (R)-(+)-tipifarnib.
15. The method of claim 1, wherein the starting material of step (i) comprises an enantiomeric excess of an undesired enantiomer of tipifarnib.
16. The method of claim 6, wherein the desired enantiomer of tipifarnib is (R)-(+)-tipifarnib.
17. The method of claim 6, wherein the starting material of step (i) comprises an enantiomeric excess of an undesired enantiomer of tipifarnib.
18. The method of claim 7, wherein reaction solvent of step (ii)(a) is a mixture of an organic solvent and water.
19. The method of claim 18, wherein the organic solvent is acetonitrile, methylethylketone, acetone, DMF, or a mixture thereof.
20. The method of claim 18, wherein the organic solvent is acetone.
21. The method of claim 7, wherein step (ii)(a) takes place at a temperature ranging from about 60° C. to about 80° C.
22. The method of claim 9, wherein the additive is an acid.
23. The method of claim 22, wherein the acid is hydrochloric acid.
24. The method of claim 22, wherein the acid is sulfuric acid.
25. The method of claim 10, wherein the product mixture's pH is adjusted to about 8.
26. The method of claim 25, wherein the base is sodium hydroxide.
27. The method of claim 25, wherein the extraction solvent is isopropyl acetate.
28. The method of claim 25, wherein the racemic alcohol is crystallized as a hydrochloride salt.
29. The method of claim 11, wherein the chiral resolving agent of step (iii)(a) is (-)dibenzoyl-L-tartaric acid.
Description
V. EXAMPLES
(1) Certain embodiments are illustrated by the following non-limiting examples. The discussion below is offered to illustrate certain aspects of the present invention and is not intended to limit the scope of the present invention. Changes can be made to the embodiments in light of the detailed description below. Although specific embodiments have been described herein for purposes of illustration, various modifications of the modes described herein for carrying out the disclosure that are obvious to persons of skill in the art are intended to be within the scope of the claims. All publications, patents, and patent applications cited in this specification are incorporated herein by reference as if each such publication, patent, or patent application were specifically and individually indicated to be incorporated herein by reference.
(2) Hereinafter “eq.” or “X” means equivalent or equivalents; “h” means hour or hours; “IPC” means in process control; “N.D.” means not detected; “RT” means retention time; “temp.” means temperature, and “V” means volume or volumes.
Example A1
Synthesis of (±)-4-(3-chlorophenyl)-6-[(4-chlorophenyl)hydroxy(1-methyl-1H-imidazol-5-yl)methyl]-1-methyl-2(1H)-quinolinone monohydrochloride (IIa)
(3) As illustrated in Scheme 2, the transformation from 6-(4-chlorobenzoyl)-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone (I) to (±)-4-(3-chlorophenyl)-6-[(4-chlorophenyl)hydroxy(l-methyl-1H-imidazol-5-yl)methyl]-1-methyl-2(1H)-quinolinone monohydrochloride (IIa) consists of two consecutive steps:
(4) The first step was the condensation of 6-(4-chlorobenzoyl)-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone (I) with 1-methyl-imidazole in the presence of n-butyllithium in hexane 23% and chlorotriethylsilane using tetrahydrofuran as solvent. After completion of the reaction, the reaction mixture was diluted with water and neutralized with acetic acid. After separation of the layers, water was added to the organic layer and it was neutralized with sodium hydroxide. The layers were separated and the organic layer was evaporated.
(5) In the second step, the residue was diluted with 2-propanone. Formation of the hydrochloric acid salt (IIa) was performed by addition of hydrochloric acid. The product was crystallized. The precipitate was isolated, washed with 2-propanone and dried.
(6) Specifically, hydrochloride (IIa) was prepared as follows:
(7) 1. Charge a reactor with minimum 1.7 L tetrahydrofuran and 1.75 moles 1-methylimidazole, stir and cool.
(8) 2. Add 0.11 kg n-butyllithium in hexane 23% (1.75 moles) and stir.
(9) 3. Add 0.27 kg chlorotriethylsilane (1.8 moles) and stir.
(10) 4. Add 0.10 kg n-butyllithium in hexane 23% (1.55 moles) at −75° C. to −68° C. and stir.
(11) 5. Charge another reactor with 1 mole ketone (I) and minimum 2 L tetrahydrofuran.
(12) 6. Stir and heat until ketone (I) is completely dissolved.
(13) 7. Cool and add the solution to the reaction mixture of step 4 while keeping the temperature at −75° C. to −68° C. and stir.
(14) 8. Add water and stir.
(15) 9. Add glacial acetic acid and stir.
(16) 10. Allow the layers to settle. Separate the layers. Discard the aqueous layer.
(17) 11. Add water and sodium hydroxide 29% to the organic layer and stir.
(18) 12. Allow the layers to settle. Separate the layers. Discard the aqueous layer.
(19) 13. Evaporate the organic layer.
(20) 14. Add 2-propanone and evaporate.
(21) 15. Repeat step 14 twice.
(22) 16. Add minimum 4 L 2-propanone to the evaporated residue, stir and cool.
(23) 17. Add hydrochloric acid and stir.
(24) 18. Centrifuge the precipitate and wash the product with 2-propanone.
(25) 19. Dry the product in a suitable drying unit.
(26) The above described process was scaled to accommodate 549 to 822 moles of ketone (I). The process yielded 45-69% of alcohol (IIa).
Example A2
Synthesis of (±)-6-[amino(4-chlorophenyl)(1-methyl-1H-imidazol-5-yl)methyl]-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone (IV)
(27) As illustrated in Scheme 2, the transformation from (±)-4-(3-chlorophenyl)-6-[(4-chlorophenyl)hydroxy(l-methyl-1H-imidazol-5-yl)methyl]-1-methyl-2(1H)-quinolinone monohydrochloride (IIa) to (±)-6-[amino(4-chlorophenyl)(1-methyl-1H-imidazol-5-yl)methyl]-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone (IV) consists of two consecutive steps:
(28) The first step was the chlorination of alcohol (IIa) with thionylchloride in 1,3-dimethyl-2-imidazolidinone as solvent.
(29) The second step was the amination of the in-situ intermediate chloride (III) (not shown in Scheme 2) to amine (IV) using a solution of ammonia in methanol. The product was crystallized by addition of water. The precipitate was isolated, washed with water and dried.
(30) Specifically, amine (IV) was prepared as follows:
(31) 1. Charge a reactor with 2 L 1,3-dimethyl-2-imidazolidinone and 1 mole hydrochloride (IIa) and stir.
(32) 2. Add minimum 0.26 kg thionylchloride (2.2 moles) while keeping the temperature below 45° C.
(33) 3. Stir at 20-45° C. for at least 3 hours.
(34) 4. Charge another reactor with minimum 1.71 L ammonia in methanol (12 moles), stir and cool.
(35) 5. Add the reaction mixture from step 3 while keeping the temperature below 45° C.
(36) 6. Stir at 12-45° C. for at least 15 hours.
(37) 7. Add maximum 5 L water and stir.
(38) 8. Centrifuge the product and wash each centrifuge load with water.
(39) 9. Dry the product in a suitable drying unit.
(40) The above described process was scaled to accommodate 480 to 720 moles of alcohol (IIa). The process yielded 65-85% of amine (IV).
Example A3
Synthesis of (R)-6-[amino(4-chlorophenyl)(1-methyl-1H-imidazol-5-yl)methyl]-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone [R-(R*,R*)]-2,3-bis(benzoyloxy)butanedioate (2:3) (VIa)
(41) As illustrated in Scheme 2, the transformation from (±)-6-[amino(4-chlorophenyl)(1-methyl-1H-imidazol-5-yl)methyl]-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone (IV) to (R)-6-[amino(4-chlorophenyl)(1-methyl-1H-imidazol-5-yl)methyl]-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone [R-(R*,R*)]-2,3-bis(benzoyloxy)butanedioate (2:3) (VIa) consisted of the formation of the [R-(R*,R*)]-2,3-bis(benzoyloxy)-butanedioic acid salt of amine (IV) with [R-(R*,R*)]-2,3-bis(benzoyloxy)-butanedioic acid monohydrate using 2-propanone as solvent. The product was isolated, washed with 2-propanone and dried.
(42) Specifically, salt (VIa) was prepared as follows:
(43) 1. Charge a reactor with minimum 3.6 L 2-propanone, 1 mole amine (IV) and minimum 2.8 moles [R-(R*,R*)]-2,3-bis(benzoyloxy)butanedioic acid monohydrate and stir at a maximum temperature of 25° C.
(44) 2. Centrifuge the product and wash each centrifuge load with 2-propanone.
(45) 3. Collected the filtrate as mother liquor (VIb).
(46) 4. Dry the product in a suitable drying unit.
(47) The above described process was scaled to accommodate 460 to 1255 moles of amine (IV). The process yielded 27-37% of salt (VIa).
Example A4
Synthesis of “crude” (R)-6-[amino(4-chlorophenyl)(1-methyl-1H-imidazol-5-yl)methyl]-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone (V)
(48) As illustrated in Scheme 2, the transformation from (R)-6-[amino(4-chlorophenyl)(1-methyl-1H-imidazol-5-yl)methyl]-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone [R-(R*,R*)]-2,3-bis(benzoyloxy)butanedioate (2:3) (VIa) to “crude” (R)-6-[amino(4-chlorophenyl)(1-methyl-1H-imidazol-5-yl)methyl]-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone (V) consisted of the liberation of the base of salt (VIa) with ammonium hydroxide using a mixture of water and ethanol as solvent. The product was crystallized by cooling. The precipitate was isolated, washed with water and dried.
(49) Specifically, “crude” (R)-(+)-tipifarnib (V) was prepared as follows:
(50) 1. Charge a reactor with ethanol anhydrous denatured and 1 mole salt (VIa) and stir.
(51) 2. Add ammonium hydroxide.
(52) 3. Add water, heat to reflux and reflux for maximum 150 minutes.
(53) 4. Cool and stir.
(54) 5. Centrifuge the product and wash each centrifuge load with water.
(55) 6. Dry the product in a suitable drying unit.
(56) The above described process was scaled to accommodate 147 to 706 moles of salt (VIa). The process yielded 70-95% of “crude” (V).
Example A5
Synthesis of “not milled” (R)-6-[amino(4-chlorophenyl)(1-methyl-1H-imidazol-5-yl)methyl]-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone (V)
(57) As illustrated in Scheme 2, “crude” (R)-6-[amino(4-chlorophenyl)(1-methyl-1H-imidazol-5-yl)methyl]-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone (V) was dissolved in ethanol and treated with activated carbon. After filtration over infusorial earth, the reaction mixture was partly evaporated. The product was crystallized by cooling. The precipitate was isolated and washed with ethanol. The wet product was again dissolved in ethanol and treated with activated carbon. After filtration over infusorial earth, the reaction mixture was partly evaporated. The product was crystallized by cooling. The precipitate was isolated, washed with ethanol and dried.
(58) Specifically, “not milled” (R)-(+)-tipifarnib (V) was prepared as follows:
(59) 1. Charge a reactor with ethanol anhydrous denatured, 1 mole crude (V), minimum 12.5 g activated carbon type norit A supra, infusorial earth and stir.
(60) 2. Heat to reflux and reflux while stirring.
(61) 3. Filter the reaction mixture to another reactor, wash the filter with ethanol anhydrous denatured, stir and evaporate maximum 5.72 L solvent over a period of maximum 24 hours.
(62) 4. Cool and stir.
(63) 5. Centrifuge the product and wash each centrifuge load with ethanol anhydrous denatured.
(64) 6. Charge a reactor with ethanol anhydrous denatured, the wet product from step 5, minimum 12.5 g activated carbon type norit A supra, infusorial earth and stir.
(65) 7. Heat to reflux and reflux while stirring.
(66) 8. Filter the reaction mixture to another reactor, wash the filter with ethanol anhydrous denatured, stir and evaporate maximum 5.43 L solvent over a period of maximum 24 hours.
(67) 9. Repeat steps 4 and 5.
(68) 10. Dry the product in a suitable drying unit until the loss on drying ≤0.20% w/w.
(69) The above described process was scaled to accommodate 392 to 588 moles of “crude” (V). The process yielded of 69-84% of “not milled” (V).
Example A6
Synthesis of (R)-(+)-tipifarnib (V)
(70) As illustrated in Scheme 2, “not milled” (R)-6-[amino(4-chlorophenyl)(1-methyl-1H-imidazol-5-yl)methyl]-4-(3-chlorophenyl)-1-methyl-2(1H)-quinolinone (V) was milled, and optionally sieved and homogenized.
(71) As milling is a semi-continuous flow-through process, there was not a typical batch size. The process yielded 90% or higher of (V).
Example B1
Feasibility Study on Recovery of Mother Liquor
(72) In order to recover the undesired enantiomer in the form of a dibenzoyltartrate salt (VIb) in the mother liquor and realize the transformation from salt (VIb) of the undesired enantiomer to alcohol (IIa) (see Scheme 2), several experiments were conducted to study the feasibility of the proposed recovery and transformation process. The details of the experiments are summarized in Tables 1, 2 and 3. The experiments were first conducted using salt (VIa) of the desired enantiomer.
(73) TABLE-US-00001 TABLE 1 Racemization of VIa without NaNO.sub.2 Batch Materials IPC (IV/II) %* No. VIa H.sub.2O MeOH Acid 19 h at 40° C. 4 h at 75° C. T1-1 1.5 g 4.5 mL 15 mL 64.22/18.43 3.24/50.55 1.0 X 3 V 10 V Solid VIa H.sub.2O acetone Acid 2 h at 50° C. 18 h at 50° C. 24 h at 50° C. T1-2 1.0 g 20 mL 96.32/2.99 86.15/13.01 71.66/27.07 1.0 X 20 V Solid T1-3 20 mL 0.1 mL 97.57/1.96 86.39/12.81 69.28/29.53 20 V 37% HCl T1-4 1 mL 20 mL 90.55/5.62 68.51/19.64 44.28/36.51 1 V 20 V *Determined by peak area integration of HPLC graph.
(74) In Experiment T1-1, 10 volumes of methanol and 3 volumes of water were used as the solvent, and the reaction was conducted at 40° C. and 75° C. In Experiment T1-2, 20 volumes of water was used as the solvent. In Experiment T1-3, 0.1 mL of concentrated hydrochloric acid was used as an additive. In Experiment T1-4, 1 volume of water and 20 volumes of acetone were used as the solvent. The last three experiments T1-2/3/4 were conducted at 50° C.
(75) The data from Experiment T1-1 indicate that conversion from salt (VIa) to alcohol (II) is feasible. Furthermore, analysis of alcohol (II) with chiral chromatography demonstrates complete racemization of the chiral center. However, the reaction conditions lead to the formation of significant amounts of impurities.
(76) The data from Experiments T1-2/3/4 indicate that racemization without sodium nitrite has a low conversion rate even with the help of acid at 50° C. For example in Experiment T1-3, only about 30% of alcohol (II) was observed after stirring at 50° C. for 24 hours.
(77) Further experiments employed sodium nitrite as an oxidant in an attempt to improve the reaction conversion and purity profile. Details of the experiments are summarized in Table 2.
(78) TABLE-US-00002 TABLE 2 Racemization of VIa with NaNO.sub.2 Batch Materials IPC (IV/II) %* No. VIa NaNO.sub.2 Solvent Acid 39 h at 20° C. T2-1 1.5 g 0.3 g 15 mL MeOH 80.32/16.10 1.0 X 4.5 mL H.sub.2O Solid VIa NaNO.sub.2 Solvent Acid 2 h at 20° C. 18 h at 20° C. 24 h at 20° C. T2-2 1.0 g 0.1 g 20 mL 42.38/48.69 49.24/41.76 29.57/57.89 1.0 X 1.5 eq. t-AmOH Solid 1 mL H.sub.2O T2-3 20 mL 16.65/72.77 24.17/64.75 20.23/69.23 Acetone 1 mL H.sub.2O T2-4 20 mL MeCN 36.53/60.72 38.40/59.03 36.10/61.15 1 mL H.sub.2O T2-5 20 mL MeCN H.sub.2SO.sub.4 1.67/92.55 0.11/97.41 1 mL H.sub.2O 2.0 eq. *Determined by peak area integration of HPLC graph.
(79) Five experiments were conducted using salt (VIa), and 1.5 equivalents of sodium nitrite were used as a reaction reagent. In Experiment T2-2, 20 volumes of tert-amyl alcohol and 1 volume of water were used as the solvent. In Experiment T2-3, 20 volumes of acetone and 1 volume of water were used as the solvent. In Experiment T2-4, 20 volumes of acetonitrile and 1 volume of water were used as the solvent. In Experiment T2-5, 20 volumes of acetonitrile and 1 volume of water were used as the solvent, and 2.0 equivalents of sulfuric acid were used as an additional additive. Experiments T2-2/3/4/5 were conducted at 20° C.
(80) As the data in Table 2 reveal, sodium nitrite can convert the desired enantiomer salt (VIa) to alcohol (II) at a mild temperature of 20° C. In particular, almost all the starting material in Experiment T2-5 was converted to alcohol (II) after stirring at 20° C. for 2 hours. Furthermore, analysis of alcohol (II) with chiral chromatography demonstrates complete racemization of the chiral center.
(81) Further experiments employed the mother liquor containing salt (VIb) to investigate the recovery process directly. The details of these experiments are summarized in Table 3.
(82) TABLE-US-00003 TABLE 3 Racemization of mother liquor with NaNO.sub.2 IPC (IV/II) %* Materials Solution Batch 50% Initial after solvent 2 h 18 h No. VIb NaNO.sub.2 H.sub.2SO.sub.4 Solvent purity switch at 20° C. at 20° C. T3-1 1.0 X 0.16 g 20 mL 89.93/1.16 15.60/74.10 0.54/87.86 Mother 1.5 eq. Acetone liquor 1 mL H.sub.2O T3-2 2.0 20 mL 85.55/4.34 0.51/89.61 0.48/90.02 eq. MeCN 1 mL H.sub.2O *Determined by peak area integration of HPLC graph.
(83) In Experiment T3-1, 20 mL of acetone and 1 mL of water were used as the solvent. In Experiment T3-2, 20 mL of acetonitrile and 1 mL of water were used as the solvent, and 2.0 equivalents of sulfuric acid were used as an additive. Both of these experiments were conducted at 20° C.
(84) As summarized in Table 3, almost all the starting material was converted to alcohol (II) after stirring at 20° C. for 18 h in both experiments. Chiral HPLC confirmed that the product alcohol (II) was racemic in both experiments. Using acetone or acetonitrile as the solvent gave similar in-process control results.
Example B2
Investigating the Process Parameters for Working Up the Racemization Reaction
(85) In order to find a suitable method to work up the reaction, several experiments were conducted using the solution from T2-5. Details of these experiments are summarized in Table 4, wherein “DBTA” refers to dibenzoyltartaric acid.
(86) TABLE-US-00004 TABLE 4 Process parameters for working up the racemization reaction Purity (DBTA/II) %* Solution before Solution after work-up work-up and phase separation Batch Materials for work-up (including Organic Aqueous No. II Solvent base pH DBTA) layer layer T4-1 1 mL 2 mL iPrOAc 10% 5 35.20/63.22 16.70/79.60 85.24/13.92 T4-2 T2-5 1 mL H.sub.2O NaOH 7 0.31/95.48 89.70/7.63 T4-3 10 0.14/95.66 80.92/15.89 T4-4 NH.sub.3•H.sub.2O 0.15/95.57 90.59/8.02 T4-5 2 mL 2-MeTHF 10% 3.77/93.23 89.05/3.23 1 mL H.sub.2O NaOH *Determined by peak area integration of HPLC graph.
(87) In Experiments T4-1/2/3, the reaction mixture was diluted with 2 volumes of isopropyl acetate and 1 volume of water, and then treated with 10% sodium hydroxide to adjust the mixture's pH value. In Experiment T4-4, the reaction mixture was diluted with 2 volumes of isopropyl acetate and 1 volume of water, and then treated with ammonium hydroxide to adjust the mixture's pH to about 10. In Experiment T4-5, the reaction mixture was diluted with 2 volumes of 2-methyltetrahydrofuran and 1 volume of water, and then treated with 10% sodium hydroxide to adjust the mixture's pH to about 10.
(88) A few observations can be made from the experimental data in Table 4. First, pH 5 is not a suitable pH point for work-up since as much as 16.70% of DBTA remains in the organic layer. Good separation is obtained at pH values from about 7 to about 10, where the amount of residual DBTA in the organic layer is below 1% when isopropyl acetate is used. Second, any residual alcohol (II) in aqueous layer may be recovered by back extraction of the aqueous layer. Last, using 2-methyltetrahydrofuran resulted in 3.77% of residual DBTA in the organic layer.
(89) It is determined from these experiments that isopropyl acetate and 10% sodium hydroxide are the optimal materials for the work-up procedure, and pH adjustment to about 10 can ensure effective removal of DBTA.
Example B3
Optimization of the Recovery Process
(90) Six experiments were conducted at different temperatures and with different solvents in order to further optimize the reaction conditions. The details of these experiments are summarized in Tables 5 and 6, wherein “DBTA” refers to dibenzoyltartaric acid.
(91) TABLE-US-00005 TABLE 5 Investigation of the reaction conditions Observation during the reaction Materials Before the After the Batch 50% Reaction addition reaction No. VIb Solvent Water NaNO.sub.2 H.sub.2SO.sub.4 Temp. of NaNO.sub.2 completes T5-1 1.0 X 40 mL 2 mL 0.2 g 20° C. Clear Much solid VIb Acetone (1 V) 1.5 eq. solution (DBTA) (Mother (20 V) after the precipitated out, liquor) addition of and the mixture water still can be stirred. T5-2 28 mL 2 mL 20° C. Much solid Acetone (2 V) (DBTA) (14 V) precipitated out, and the mixture was difficult to stir. T5-3 14 mL 2 mL 20° C. Much solid Acetone (1 V) (DBTA) (7 V) precipitated out, and the mixture still can be stirred. (Similar to the 1.sup.st reaction) T5-4 14 mL 2 mL 40° C. Much solid Acetone (1 V) (DBTA) (7 V) precipitated out, and the mixture still can be stirred. (Similar to the 1.sup.st reaction) T5-5 14 mL 4 mL 20° C. Clear solution Acetone (2 V) (7 V) T5-6 10 mL 4 mL 2.0 eq. 20° C. Clear Clear solution MeCN (2 V) solution (5 V) after stirring for 0.5 h
(92) All six experiments were conducted using the mother liquor (VIb) obtained from large scale manufacturing. The mother liquor was concentrated, and then water was charged into the reaction mixture. The behavior of the reactions was observed and noted in Table 5. Next, sodium nitrite and sulfuric acid (if needed) were added into the reaction mixture, and the reactions were monitored by HPLC.
(93) In Experiment T5-1, the volume of the reaction mixture was close to the full capacity of the reactor. The reaction mixture was a clear solution after the addition of water. However, a large amount of solid precipitated out during the reaction. The reaction mixture could still be stirred after the completion of the reaction, but some solid clung to the inner wall of the reactor.
(94) In order to increase the reactor capacity, the next two experiment employed a smaller volumes of solvent. In Experiment T5-2, 14 volumes of acetone were used. In Experiment T5-3, 7 volumes of acetone were used. A large amount of solid precipitated out in both experiments. Particularly in Experiment T5-2, which used 14 volumes of solvent, the reaction mixture exhibited a much worse stirring behavior. The experiments indicate that water is helpful in dissolving DBTA.
(95) Experiment T5-4 was an attempt to improve the stirring behavior by increasing the reaction temperature to 40° C. However, it did not achieve the desired result.
(96) In Experiment T5-5, 2 volumes of water were used. The reaction mixture was still a clear solution even after the completion of the reaction, which could make the process robust.
(97) In Experiment T5-6, the reaction solvent was switched to acetonitrile with two volumes of water. However, the reaction mixture required stirring for about half an hour to reach a clear solution. The reaction mixture was also a clear solution when the reaction was complete.
(98) The in-process control data of the six experiments are shown in Table 6.
(99) TABLE-US-00006 TABLE 6 Investigation of the reaction conditions IPC (IV/II) %* Materials Solution after Batch Reaction concentration or No. Solvent Water Temp. solvent switch 2 h 18 h T5-1 40 mL Acetone 2 mL 20° C. 72.07/11.39 17.69/65.29 2.49/79.35 (20 V) (1 V) T5-2 28 mL Acetone 2 mL 20° C. 73.21/10.94 31.63/52.19 2.59/78.78 (14 V) (1 V) T5-3 14 mL Acetone 2 mL 20° C. 72.60/11.12 10.83/70.68 2.29/79.64 (7 V) (1 V) T5-4 14 mL Acetone 2 mL 40° C. 72.14/11.23 4.06/77.49 2.00/75.26 (7 V) (1 V) T5-5 14 mL Acetone 4 mL 20° C. 71.15/11.77 6.73/76.18 1.65/81.70 (7 V) (2 V) T5-6 10 mL MeCN 4 mL 20° C. 71.50/11.64 1.40/82.40 1.22/82.32 (5 V) (2 V) *Determined by peak area integration of HPLC graph.
(100) Several observations can be made from the experimental results in Table 6. First, the reactions at 20° C. affords similar results after stirring for 18 hours, but raising the reaction temperature to 40° C. leads to a lower product purity. Second, using a smaller volume of solvent (and thus a high concentration of reactants) resulted in a higher reaction rate. For example, only about 7% of salt (VIb) remained after stirring for 2 hours in Experiment T5-5. Last, using either acetone or acetonitrile as the solvent gives similar results.
Example B4
Further Investigation of the Recovery Process
(101) Three experiments were conducted on 20 gram scale to further investigate the reaction conditions. The details of the experiments are summarized in Tables 7, 8, and 9.
(102) TABLE-US-00007 TABLE 7 Investigation of the reaction conditions Observation during the reaction Materials Before the After the Batch 50% Reaction addition of reaction No. VIb Solvent Water NaNO.sub.2 H.sub.2SO.sub.4 Temp. NaNO.sub.2 complete T7-1 1.0 X 249 mL 20 mL 2.44 g 20° C. Clear solution Much solid VIb Acetone (1 V) 1.8 eq. after the (DBTA) (Mother (12.5 V, addition of precipitated liquor) without water. out, and the concentration) mixture was difficult to stir. T7-2 100 mL MeCN 40 mL 2.0 eq. Clear solution No solid (5 V) (2 V) after stirring (DBTA) for 0.5 h. precipitated out. The mixture was clear. T7-3 120 mL 40 mL Clear solution No solid Acetone (2 V) after the (DBTA) (6 V) addition of precipitated water. out. The mixture was clear.
(103) All three experiments were conducted using the mother liquor (VIb) obtained from large scale manufacturing. The mother liquor was concentrated, and then water was charged into the reaction mixture. The behavior of the reactions was observed and noted in Table 7. Next, sodium nitrite and sulfuric acid (if needed) were added into the reaction mixture, and the reactions were monitored by HPLC.
(104) In Experiment T7-1, the reaction was conducted using the mother liquor directly without concentration. The reaction mixture was a clear solution after the addition of water. However, a large amount of solid precipitated out during the reaction which rendered the reaction mixture difficult to stir.
(105) In Experiment T7-2, a solvent switch was first carried out to obtain a solution in acetonitrile. The reaction mixture was initially a sticky solution but became a clear solution after adding water and stirring for half an hour. No solids of DBTA precipitated out during the course of reaction, and the reaction mixture stayed clear.
(106) In Experiment T7-3, the reaction was concentrated to 6 volumes. The reaction mixture became a clear solution after the addition of water. No solids of DBTA precipitated out during the course of the reaction, and the reaction mixture was also a clear solution when the reaction was complete.
(107) The in-process control results of these three experiments are summarized in Table 8.
(108) TABLE-US-00008 TABLE 8 Investigation of the reaction conditions IPC (IV/II) %* Solution after Batch Materials Initial concentration or No. Solvent Water purity solvent switch 2 h 18 h T7-1 249 mL Acetone 20 mL 78.19/7.13 7.65/77.51 0.48/82.72 (12.5 V, without (1 V) concentration) T7-2 100 mL MeCN 40 mL 77.46/7.50 0.75/85.10 0.34/85.21 (5 V) (2 V) T7-3 120 mL Acetone 40 mL 77.00/8.22 1.15/83.71 0.86/83.04 (6 V) (2 V) *Determined by peak area integration of HPLC graph.
(109) The purity of alcohol (II) obtained from the three experiments was almost the same. Chiral HPLC confirmed that all three experiments produced alcohol (II) as a racemic mixture. However, the reaction without concentration (Experiment T7-1) required a longer period of time to convert all the starting material to the product.
(110) Using lower volumes of solvent (and thus higher concentrations) resulted in higher reaction rate in Experiments T7-2/3. There was only about 1% of the starting material (VIb) remaining after stirring for 2 hours.
(111) All three experiments were further continued with work-up and crystallization procedures. In Experiments T7-1/2, the reaction mixture was diluted with 100 mL isopropyl acetate and 100 mL water, then treated with 10% aqueous sodium hydroxide to adjust the mixture's pH to about 10. In Experiment T7-3, the reaction mixture was diluted with 100 mL isopropyl acetate, then treated with about 80 mL of 10% aqueous sodium hydroxide to adjust the mixture's pH to about 10. In order to achieve high recovery, the aqueous layer was back-extracted with 100 mL isopropyl acetate. The combined organic layer was used to crystallize salt (IIa), by first switching solvent to 10 volumes of acetone and then acidified with 1.5 equivalents of 37% HCl.
(112) The final results for the dry cakes of salt (IIa) are shown in Table 9. RT stands for HPLC retention time. The major impurity at RT 7.5 min is identified and fully characterized. The chemical structure is assigned as VII in Table 9.
(113) TABLE-US-00009 TABLE 9 Results from crystallization studies Expt. # T7-1 T7-2 T7-3 RT 5.4 min Mesityl oxide
(114) After crystallization with acetone, the purity of alcohol (IIa) is almost 98% in the final product.
Example B5
Investigation of Impurity (VII)
(115) Eight experiments were conducted to study the formation of impurity (VII). The details of these experiments are summarized in Table 10.
(116) TABLE-US-00010 TABLE 10 Investigation of impurity (VII) Observation during the reaction IPC for 18 h Before the After the Batch Materials Impurity addition of reaction No. VIb Acetone Water VII NaNO.sub.2 complete T10-1 1.0 X 12.5 V 0 V 0.12% Clear solution Much solid (DBTA) VIb (Without precipitated out, and the (Mother concentration) mixture was not able to stir. T10-2 liquor) 0.5 V 0.15% Much solid (DBTA) precipitated out, and the mixture was difficult to stir. T10-3 1 V 0.30% Much solid (DBTA) precipitated out, and the mixture still can be stirred. T10-4 2 V 0.52% No solid (DBTA) precipitated out. The mixture was clear. T10-5 6 V 0.5 V 0.17% Much solid (DBTA) precipitated out, and the mixture still can be stirred. T10-6 1 V 0.40% Much solid (DBTA) precipitated out, and the mixture was difficult to stir. T10-7 2 V 0.61% No solid (DBTA) precipitated out. The mixture was clear. T10-8 6 V 1.04% Much solid (B) Much solid (DBTA) precipitated out precipitated out, and the mixture still can be stirred.
(117) The experimental data in Table 10 indicate that more water in the reaction mixture leads to a higher content of impurity (VII). This is especially the case in Experiment T10-8, where 6 volumes of water resulted in 1.04% of the impurity. However, it should be noted that while a low volume of water can afford better results in term of the purity profile, but at the same time the low volume of water may causes mixing problems. This is the case in Experiment T10-1, where the reaction mixture was almost not stirrable since it had no water in the reaction mixture.
(118) Experiment T10-5 (6 volumes of acetone solution with 0.5 volume of water) resulted in only 0.17% of impurity (VII) in a 3 gram scale reaction. In order to further verify the process, the same procedure in Experiment T10-5 was repeated on larger scales in three further experiments. The details of the three experiments are summarized in Tables 11 and 12.
(119) TABLE-US-00011 TABLE 11 Further investigation of the recovery process Observation during the reaction Materials Before the After the Batch Reaction addition reaction No. VIb.sup.† Acetone Water NaNO.sub.2 Temp. of NaNO.sub.2 complete T11-1 20 g 6 V 0.5 V 1.8 eq. 20° C. Clear Much solid (DBTA) (Concentrate solution precipitated out, and the the solution mixture still can be stirred. T11-2 40 g to 6 V) Much solid (DBTA) precipitated out. The upper reaction mixture was too viscous to stir, and it was T11-3 100 g almost jelly. .sup.†Calculated mass of salt (VIb) contained in mother liquor.
(120) TABLE-US-00012 TABLE 12 Further investigation of the recovery process Expt. # T11-1 T11-2 T11-3 RT 5.4 min Mesityl oxide
(121) In the experiment on 20 g scale (Experiment T11-1), although much solid (DBTA) precipitated out at the completion of the reaction, the reaction mixture could still be stirred. However, when this process was magnified to the scales of 40 g or 100 g, the process experienced mixing problems. The reaction mixture was too viscous to stir at the completion of the reaction. This is especially so in the upper portion of the reaction mixture, where it was almost jelly-like.
(122) Furthermore, the content of impurity (VII) was also higher than usual (about 0.2%) in the scaled up reactions.
(123) In an attempt to further optimize the process, yet another three experiments were conducted with different amounts of acetone and water on a 60 gram scale. The details of these experiments are summarized in Table 13.
(124) TABLE-US-00013 TABLE 13 Further investigation of the recovery process Batch Materials Observation during the reaction (VIb/VII/II)* No. VIb.sup.† Acetone Water IPC for 2 h IPC for 18 h T13-1 60 g 9 V 0.75 V Clear 3.66/0.20/86.47 Much solid (DBTA) 0.14/0.27/90.17 solution precipitated out. The T13-2 6 V 1 V Clear 0.13/0.55/89.04 upper reaction mixture 0.13/0.58/89.26 solution was too viscous to stir, and it was almost jelly. T13-3 6 V 1.5 V Clear 0.03/0.82/89.22 Clear 0.10/0.77/90.28 solution solution *Determined by peak area integration of HPLC graph. .sup.†Calculated mass of salt (VIb) contained in mother liquor.
(125) The results in Table 13 indicate that the reactions with low volumes of water (0.75 and 1 volume) has mixing problems, especially after stirring for long hours. The reaction with 1.5 volumes of water always gives a clear solution even after the completion of reaction, but it resulted in a higher level of impurity.
(126) Two experiments were conducted with procedure in T13-1 at different temperatures (30 and 40° C.). The details of these experiments are summarized in Table 14.
(127) TABLE-US-00014 TABLE 14 Further study of the recovery process Batch Materials Observation during the reaction (VIb/VII/II)* No. VIb.sup.† Acetone Water Temp. IPC for 1 h IPC for 18 h Work-up T14-1 60 g 9 V 30° C. Clear 2.47/0.18/87.78 Much solid (DBTA) 0.11/0.51/87.91 solution precipitated out. The T14-2 0.75 V 40° C. Clear 0.13/1.13/89.47 upper reaction mixture 0.09/0.38/88.38 solution was too viscous to stir, and it was almost jelly. *Determined by peak area integration of HPLC graph. .sup.†Calculated mass of salt (VIb) contained in mother liquor.
(128) As the experimental results in Table 14 demonstrate, the reactions at high temperatures (30° C. and 40° C.) still experience mixing problems, especially after stirring for long hours. Moreover, the content of impurity (VII) were also higher than at 20° C. (20° C., RT 7.5 min about 0.2%).
(129) It is determined from the experimental results here that using 6 volumes of acetone and 1.5 volumes of water minimizes the mixing problems and provides the optimal results in the final recovery process.
Example B6
Demo Batch in Kilo Lab
(130) A demo batch of the mother liquor recovery process was conducted on 1.445 kg scale in a kilo lab with a 30-liter reactor. The details of the demo batch are summarized in Tables 15 and 16. In this experiment, multiple batches of the mother liquor from large scale manufacturing were combined, and their total volume was about 20 liters.
(131) The mother liquor was concentrated to 6 volumes, and 1.5 volumes of water were added. Next, 1.8 equivalents of sodium nitrite were charged into the reactor slowly while maintaining the reaction mixture at a temperature below 40° C. At the completion of the reaction, 5 volumes of isopropyl acetate were added, and 10% aqueous solution of sodium hydroxide was charged to adjust the mixture to pH 10. Then the aqueous layer after phase cut was re-extracted with 5 volumes of isopropyl acetate. Finally, 1.5 equivalents of concentrated hydrochloric acid were charged into the reaction mixture to crystallize the product after solvent switch.
(132) TABLE-US-00015 TABLE 15 Demo batch in kilo lab Materials Work-up Batch Reaction 10% Crystallization No. VIb.sup.† Acetone Water NaNO.sub.2 Temp. .sup.iPrOAc NaOH Conc. HCl T15 1.445 kg The mother 1.5 V 1.8 eq. 20° C. 10 V 9 L 1.5 eq. liquor was 2.16 L 175 g 14.44 L pH = 10 208.4 g concentrated to 6 V .sup.†Total mass of salt (VIa) and (VIb) contained in mother liquor.
(133) TABLE-US-00016 TABLE 16 Results from demo batch in kilo lab Expt. # T15 Mother liquor Solution before before Mother Dry Chart reaction IPC crystallization liquor cake RT 5.4 min Mesityl oxide
(134) In this experiment, the reaction mixture was a clear solution even after stirring overnight. The entire process showed good behavior at the various stages of the demo batch, including reaction, work-up and crystallization. In the final isolated product, the purity of salt (IIa) was 97.10%, and the content of impurity (VII) was 0.73%. Furthermore, the product salt (IIa) isolated after this recycling process is racemic.
(135) The yield loss from crystallization was 5%, and the yield for the overall recovery process was 97%, based on the amount of (S)-(−) tipifarnib in the mother liquor (VIb).
Example C1
Recycling Process without Sodium Nitrite
(136) Four experiments were carried out with various solvents at elevated temperatures in order to further study the recovery process without using sodium nitrite. These experiments are detailed in Tables 17 and 18. All the experiments used the dry solid of salt (VIa). The behavior of the reactions were observed and recorded after adding solvent, after adding water, and after stirring the reaction mixture for 48 hours.
(137) TABLE-US-00017 TABLE 17 Investigation of reaction solvent: reaction set up. Observation Materials After After After Batch H.sub.2O Solvent adding adding stirring No. VIa 1.0 V 6.0 V solvent water for 48 h T17-1 1.0 g 1 mL 6 mL Undissolved Some solid Some solid 1.0 X MeCN precipitated out precipitated Solid (RT) out T17-2 6 ml Undissolved Some solid Some solid MEK precipitated out precipitated (RT) out T17-3 6 ml Undissolved Some solid Some solid Acetone precipitated out precipitated (RT) out T17-4 6 ml Dissolved Clear solution Clear solution DMF
(138) TABLE-US-00018 TABLE 18 Investigation of reaction solvent: experimental results. Batch No. Results Enantiomeric Temperature RT = 1.26 Benzoic Composition (Solvent) Stirring Time min Acid IV DBTA II of II S.M. VIa N.D. N.D. 64.89 34.07 1.04 T17-1 3 h 3.46 4.97 14.77 17.45 55.62 80° C. 6 h 4.12 7.70 7.70 10.19 64.96 (MeCN/H.sub.2O) 24 h N.D. 15.45 4.91 0.74 75.21 48 h 1.78 16.26 4.01 0.41 70.64 48.69/51.31 T17-2 3 h 3.88 4.58 37.04 20.73 32.49 80° C. 6 h 5.58 7.80 22.43 12.89 47.06 (MEK/H.sub.2O) 24 h 4.95 16.80 2.04 0.98 69.67 48.09/51.91 T17-3 3 h 2.40 2.47 44.43 25.65 22.41 reflux at 60° C. 6 h 4.17 4.34 36.63 22.25 31.33 (acetone/H.sub.2O) 24 h 5.90 10.10 11.50 7.99 59.67 48 h 4.52 13.86 4.98 2.39 67.48 47.87/52.13 T17-4 3 h 3.16 4.50 39.54 19.24 27.59 80° C. 6 h 3.87 7.00 29.91 12.18 37.92 (DMF/H.sub.2O) 24 h 3.13 13.39 11.87 1.32 58.44 48 h 1.01 17.12 5.01 0.11 65.11 48.38/51.62
(139) In experiment T17-1 where MeCN was used as the reaction solvent, only 7.70% of unconverted amine (IV) remained after stirring for 6 hours at 80° C. However, the reaction of reaction became slower subsequently, with 4.01% of unconverted amine (IV) still remaining after stirring for 24 hours. In experiment T17-2 where methylethylketone (MEK) was used as the reaction solvent, 22.43% of unconverted amine (IV) remained after stirring for 6 hours at 80° C. The amount of unconverted amine (IV) decreased further to about 2% after stirring for 24 hours. In experiment T17-4 where DMF was used the reaction solvent, the experiment did not afford better results than the other experiments that used different solvents.
(140) The enantiomeric composition of alcohol (II) in the product solution was tested, and it was racemic within the error of measurement.
Example C2
Investigation of the Volume of Water
(141) Six further experiments were carried out to investigate the impact of different volumes of water. These experiments are detailed in Tables 19 and 20. All the experiments used the dry solid of salt (VIa). The behavior of the reactions were observed and recorded after adding solvent, after adding different volumes of water, and after stirring the reaction mixture for 48 hours.
(142) TABLE-US-00019 TABLE 19 Investigation of the volume of water: reaction set up. Observation After After After stirring Batch Materials adding adding for 48 h at No. VIa H.sub.2O Solvent solvent water 80° C. T19-1 1.0 g 1 mL 6 mL Undissolved Some solid Some solid 1.0 X MeCN precipitated out precipitated out Solid (RT) T19-2 2 mL Undissolved Some solid Some solid precipitated out precipitated out (RT) T19-3 3 mL Undissolved Some solid Clear solution precipitated out (RT) T19-4 1 mL 6 mL Undissolved Some solid Some solid MEK precipitated out precipitated out (RT) T19-5 2 mL Undissolved Some solid Clear solution precipitated out with two phase (RT) T19-6 3 mL Undissolved Some solid Clear solution precipitated out with two phase (RT)
(143) TABLE-US-00020 TABLE 20 Investigation of the volume of water: experimental results. Results (%) RT = Benzoic Batch No. Stirring Time 1.26 min Acid IV DBTA II IV/(IV + II) S.M. VIa N.D. N.D. 64.89 34.07 1.04 98.4 T19-1 3 h 3.46 4.97 14.77 17.45 55.62 21.0 (MeCN/H.sub.2O = 6 h 4.12 7.70 7.70 10.19 64.96 10.6 6 V/1 V) 24 h N.D. 15.45 4.91 0.74 75.21 6.1 48 h 1.78 16.26 4.01 0.41 70.64 5.4 T19-2 3 h 3.34 4.06 10.36 21.34 58.46 15.1 (MeCN/H.sub.2O = 6 h 4.27 5.86 5.85 16.34 64.57 8.3 6 V/2 V) 24 h 4.90 12.33 3.79 3.85 70.69 5.1 48 h 4.11 15.08 3.56 1.07 71.61 4.7 T19-3 3 h 3.22 3.23 7.03 24.51 59.86 10.5 (MeCN/H.sub.2O = 6 h 4.23 4.65 3.64 20.48 64.50 5.3 6 V/3 V) 24 h 6.76 9.65 2.42 8.89 68.90 3.4 48 h 6.90 12.070 2.42 3.85 70.61 3.3 T19-4 3 h 3.88 4.58 37.04 20.73 32.49 53.3 (MEK/H.sub.2O = 6 h 5.58 7.80 22.43 12.89 47.06 32.3 6 V/1 V) 24 h 4.95 16.80 2.04 0.98 69.67 2.8 T19-5 3 h 3.30 4.44 34.72 21.48 34.51 50.2 (MEK/H.sub.2O = 6 h 4.42 6.78 24.64 16.17 46.13 34.8 6 V/2 V) 24 h 6.62 15.50 2.62 2.68 70.86 3.6 48 h 5.19 18.68 1.16 0.74 72.54 1.6 T19-6 3 h 2.64 4.14 35.16 22.30 34.48 50.5 (MEK/H.sub.2O = 6 h 3.80 6.83 22.62 16.57 48.62 31.8 6 V/3 V) 24 h 8.48 13.08 4.61 5.29 67.60 6.4 48 h 6.67 17.79 0.95 1.35 72.28 1.3
(144) It can be concluded from the experimental results in Tables 19 and 20 that increasing the amount of water from 1.0 to 3.0 volumes moderately increased the reaction rate, because the increased amount of water could solubilize any solid precipitated during the reaction. However, there was nonetheless approximately 3% to 6% of unconverted amine (IV) after stirring for 24 hours.
(145) It can also be concluded from the experimental results in Tables 19 and 20 that MeCN appeared to be a superior solvent than MEK, especially when looking at the results at 6 h.
Example C3
Investigation of Acid Additives
(146) Four further experiments were carried out to investigate the accelerating effect of an acid additive, 5 wt % sulfuric acid, on the rate of reaction. The experiments were conducted using the mother liquor (VIb) with the same solvent/water ratio (6.0 V to 3.0 V) at 80° C. The experiments are summarized in Tables 21 and 22.
(147) TABLE-US-00021 TABLE 21 Investigation of acid additive: reaction set up. Materials Observation VIb.sup.† After After After After Batch 1.0 X Solvent adding adding heating stirring for No. (mother liquor) H.sub.2O 6.0 V Additive solvent water to 80° C. 24 h at 80° C. T21-1 45 g 3.0 V MeCN None Undissolved Undissolved Clear Clear T21-2 5 g 5% H.sub.2SO.sub.4 solution solution (RT) T21-3 45 g MEK None Undissolved Undissolved Clear Clear T21-4 5 g 5% H.sub.2SO.sub.4 solution solution with two phase .sup.†Calculated mass of salt (VIb) contained in mother liquor.
(148) TABLE-US-00022 TABLE 22 Investigation of acid additive: experimental results. Results (%) RT = Benzoic Batch No. Stirring Time 1.26 min Acid IV DBTA II IV/(IV + II) S.M. VIb 0.16 0.24 61.04 35.86 0.41 99.3 (Mother liquor) T21-1 2 h 3.98 3.05 1.95 53.05 36.07 5.1 (MeCN) 4 h 7.05 5.87 0.13 45.13 38.88 0.3 6 h 10.26 9.40 N.D. 36.80 39.48 N.D. 8 h 10.65 9.88 N.D. 35.84 39.47 N.D. T21-2 2 h 3.05 2.21 1.47 55.72 36.13 3.9 (5% H.sub.2SO.sub.4 in 4 h 5.42 4.04 0.11 50.65 37.98 0.3 MeCN) 6 h 8.48 6.64 N.D. 43.41 39.29 N.D. 8 h 8.57 6.78 N.D. 43.68 38.68 N.D. T21-3 2 h 3.31 2.26 13.94 54.89 24.32 36.4 (MEK) 4 h 5.93 4.31 5.68 49.18 33.17 14.6 6 h 9.05 7.16 1.67 41.58 38.16 4.2 8 h 9.45 7.55 1.24 41.57 38.12 3.2 16 h 12.88 12.83 0.05 30.68 40.49 0.1 T21-4 2 h 2.96 1.75 12.19 55.37 26.56 31.5 (5% H.sub.2SO.sub.4 in 4 h 13.33 4.18 6.21 42.10 32.94 15.9 MEK) 6 h 8.54 5.48 1.04 46.00 37.44 2.7 8 h 7.67 5.66 0.84 46.58 37.99 2.2
(149) When using the mother liquor (VIb) directly in the racemization reaction, the rate of reaction was much faster when using MeCN as the solvent than MEK. Almost all of the starting material (IV) converted to alcohol (II) within four hours in MeCN, while it took approximately 16 hours under analogous conditions in MEK.
(150) The acid additive (5 wt % sulfuric acid) appeared to have limited accelerating effect on the rate of reaction. Experimental results in the presence and absence of the acid additive are largely similar.
Example C4
Manufacture Procedure
(151) The transformation from the mother liquor (VIb) to alcohol hydrochloride (IIa) may be effected in the following procedure:
(152) 1. Charge a reactor with mother liquor (VIb) containing 1.64 kg (1.6 mol) of salts (VIa) and (VIb).
(153) 2. Concentrate to 5 L (3.0 V), then charge 7.3 kg (4.5 X) of methylethylketone and 4.5 kg (2.7 X) of process water.
(154) 3. Stir the resulting biphasic mixture at 76° C. for 14 hours.
(155) 4. Monitor the reaction by UPLC until the ratio IV/(IV+II) is equal to or less than 2%.
(156) 5. Add 3.8 kg (2.3 X) process water.
(157) 6. Add 2.4 kg of 30 w/w % NaOH solution such that the reaction mixture has pH ≥8.
(158) 7. Separate layers.
(159) 8. Extract the aqueous layer with IPAc (6.0 kg, 3.7 X).
(160) 9. Add 4.5 kg (2.7 X) process water to the combined organic layers.
(161) 10. Age the resulting mixture at 20° C. for 12 hours.
(162) 11. Concentrate the mixture at 50° C. under reduced pressure until its volume is approximately 9 L (5.5 V).
(163) 12. Add 3.3 kg (2.0 X) IPAc.
(164) 13. Age the resulting slurry at 20° C. for 12 hours.
(165) 14. Filter the slurry and rinse the wet cake with 1.3 kg (0.8 X) of acetone.
(166) 15. Charge the wet cake with acetone (4.7 X, 6 V).
(167) 16. Add 35% HCl solution (224 g, 2.1 mol, 0.14 X) at 20° C. over 5 hours and age for 14 hours.
(168) 17. Filter, rinse with acetone (3.3 kg, 2.0 X), and dry at 80° C. for 72 hours under vacuum.
(169) The above described process has a typical yield of approximately 80-90%.