Process for preparing esters of lactic acid, and 2-hydroxy-3-butenoic acid or α-hydroxy methionine analogues from sugars

Abstract

A continuous flow process for the preparation of one or more esters of lactic acid and 2-hydroxy-3-butenoic acid or -hydroxy methionine analogs from a sugar in the presence of a solid Lewis acid catalyst and a solvent comprising an organic solvent and water. The invention provides a means for stabilizing a Lewis acid catalyst for use in a continuous reaction process wherein the water is present in an amount of up to or equal to 10 vol. % of the organic solvent.

Claims

1. A continuous flow process for the preparation of one or more esters of lactic acid and 2-hydroxy-3-butenoic acid from a sugar selected from one or more of the group consisting of glucose, fructose, mannose, sucrose, xylose, erythrose, erythrulose, threose and glycolaldehyde, in the presence of a solid Lewis acid catalyst and a solvent comprising an organic solvent and water, wherein the water is present in an amount of up to or equal to 10 vol. % of the organic solvent, wherein the solid Lewis acid catalyst has a framework structure, which is selected from the group consisting of BEA, MFI, FAU, MOR, FER, MWW, MCM-41 and SBA-15.

2. A process according to claim 1, wherein the yield of the one or more lactic acid esters decreases by up to 0.25% per hour on stream on average, wherein the yield is amount of ester relative to the amount of sugar.

3. A process according to claim 1, wherein the yield of the one or more lactic acid esters decreases by up to 0.10% per hour on average, wherein the yield is amount of ester relative to the amount of sugar.

4. A process according to claim 1, wherein the yield of the one or more lactic acid esters decreases by up to 5% after 50 hours on stream, wherein the yield is amount of ester relative to the amount of sugar.

5. A process according to claim 1, wherein the yield of the one or more lactic acid esters is greater than 40% after 50 hours on stream, wherein the yield is amount of ester relative to the amount of sugar.

6. A process according to claim 1, wherein the solid Lewis acid catalyst is calcined after 450 hours of continuous flow process.

7. A process according to claim 1, wherein the solid Lewis acid catalyst comprises an active metal selected from one or more of the group consisting of Sn, Ti, Pb, Zr, Ge and Hf.

8. A process according to claim 1, wherein the solid Lewis acid catalyst is selected from the group consisting of Sn-BEA, Sn-MFI, Sn-FAU, Sn-MOR, Sn-MWW, Sn-MCM-41 and Sn-SBA-15.

9. A process according to claim 1, wherein the solid Lewis acid catalyst comprises Sn and the loss of Sn from the solid Lewis acid catalyst is less than or equal to 0.11% of the initial amount of Sn per hour on stream.

10. A process according to claim 1, wherein the solid Lewis acid catalyst comprises Sn and the loss of Sn from the solid Lewis acid catalyst is less than 8% of the initial amount of Sn after 50 hours on stream.

11. A process according to claim 1, wherein the yield of 2-hydroxy-3-butenoic acid ester is greater than 12% after 50 hours on stream, wherein the yield is amount of ester relative to the amount of sugar.

12. A process according to claim 1, wherein an alkaline earth metal or alkali metal ion is present in the process.

13. A process according to claim 1, wherein the solvent is selected from one or more of the group consisting of methanol, ethanol, 1-propanol, 1-butanol and isopropanol.

14. A process according to claim 1, wherein the temperature of the process is from 140 C. to 200 C.

15. A process according to claim 1, wherein at least a fraction of the water is introduced to the solvent by mixing the organic solvent with an aqueous sugar solution.

16. A process according to claim 1, wherein the aqueous sugar solution is a sugar syrup with a sugar dry matter content of 30% or higher.

Description

(1) FIGS. 1-5 cover the conversion of C6-sugars (fructose) in continuous flow mode. FIGS. 1-3 refer to diminishing of Sn leaching and diminishing of catalyst deactivation. FIGS. 4 and 5 demonstrate the improvement in MVG yield when water is present. More specifically, the figures have the following meanings:

(2) FIG. 1: Improved stability of the Sn-BEA Lewis acid catalysts with 1%, 5% and 10% water added to the process solvent compared to no water added to the process solvent. The catalyst stability is illustrated by a significantly consistent yield of methyl lactate product.

(3) FIG. 2: Improved stability of the Sn-BEA Lewis acid catalysts with 1% water added to the process solvent (squares) compared to no water added to the process solvent (triangles). The catalyst stability is illustrated by a significantly consistent yield of methyl lactate product observed over a prolonged period of process time on stream (ca. 500 hours).

(4) FIG. 3: Improved stability of the Sn-BEA Lewis acid catalysts with 1% water added to the process solvent (squares) compared to no water added to the process solvent (triangles). The improved stability is illustrated by a significant decrease of Sn leaching (loss of Sn) from the catalyst.

(5) FIG. 4: Improved yield of esters of 2-hydroxy-3-butenoic acid with the addition of water to the continuous flow process: (a) Yield of 2-hydroxy-3-butenoic acid methyl ester (MVG); (b) Combined yield of esters of lactic acid (methyl lactate) and 2-hydroxy-3-butenoic acid methyl ester (MVG).

(6) FIG. 5: Yield of bio-monomers obtained from fructose when using Sn-Beta zeolite in flow. Total yield [methyl lactate (ML), glycolaldehyde dimethyl acetal (GLAD) and methyl vinyl glycolate (MVG)] is 70% from fructose and it is stable over 400 h. Methyl vinyl glycolate (MVG) is equivalent to 2-hydroxy-3-butenoic acid methyl ester.

(7) FIGS. 6-8 cover the conversion of C2-sugar (glycolaldehyde) in continuous flow mode. FIGS. 6 and 7 refer to the improvement in MVG yield when water is present. FIG. 8 refers to the improvement in S-MVG yield when water and methanethiol are present. More specifically, the figures have the following meanings:

(8) FIG. 6: Improved yield of methyl vinyl glycolate (MVG) from glycolaldehyde with the addition of water in continuous flow mode. Feed composition: 20 g/l glycolaldehyde in methanol as solvent. Methyl vinyl glycolate (MVG) is equivalent to 2-hydroxy-3-butenoic acid methyl ester.

(9) FIG. 7: Yield of bio-monomers obtained from glycolaldehyde when using Sn-Beta zeolite in flow. Feed composition: 20 g/l glycolaldehyde in methanol as solvent, 8.5 wt % water. Total yield [glycolaldehyde dimethyl acetal (GLAD), methyl vinyl glycolate (MVG) and methyl-4-methoxy-2-hydroxybutenoate (MMHB)] is 65% from glycolaldehyde at the beginning of the reaction. Methyl vinyl glycolate (MVG) is equivalent to 2-hydroxy-3-butenoic acid methyl ester.

(10) FIG. 8: Yield of methyl ester of -hydoxy methionine analogue with Sn-BEA as catalyst in the presence of water and methanol in continuous flow reaction from glycolaldehyde. Feed composition: 9 g/l glycolaldehyde in methanol as solvent, 8.5 wt % water, 1.2 g/l methanethiol.

(11) FIG. 9 exemplifies the use of syrup as feed in the reaction. It demonstrates the improvement in MVG yield when water is present and the use of sugar in the form of a sugar syrup. More specifically, the figure has the following meaning:

(12) FIG. 9: Yield of bio-monomers obtained from sucrose syrup when using Sn-Beta zeolite in flow. Feed composition: 55 g/l sucrose. The total yield [methyl lactate (ML), methyl vinyl glycolate (MVG)] is 80% from sucrose syrup and it is stable. Methyl vinyl glycolate (MVG) is equivalent to 2-hydroxy-3-butenoic acid methyl ester.

(13) The process of the invention is illustrated further by the following examples.

EXAMPLE 1

(14) Preparation of Catalyst

(15) Sn-BEA (Si/Sn=125) is prepared according to a modification of the procedure described in U.S. Pat. No. 4,933,161. Commercial zeolite Beta (Zeolyst, Si/Al 12.5, ammonium form) is calcined (550 C. for 6 h) to obtain the H form (de-aluminated form) and treated with 10 grams of concentrated nitric acid (Sigma-Aldrich, 65%) per gram of zeolite beta powder for 12 h at 80 C. The resulting solid is filtered, washed with ample water and calcined (550 C. for 6 h) to obtain the dealuminated Beta. This solid is impregnated by incipient wetness methodology with a Sn/Si ratio of 125. For this purpose, tin (II) chloride (0.128 g, Sigma-Aldrich, 98%) is dissolved in water (5.75 ml) and added to the de-aluminated Beta (5 g). After the impregnation process, the samples are dried 12 h at 110 C. and calcined again (550 C. for 6 h).

(16) Catalytic Reaction in Continuous Flow Mode:

(17) Fructose (Sigma-Aldrich, 99%) was dissolved in methanol (Sigma-Aldrich, 99.9%) at room temperature to reach a concentration of 12.5 g/l. Additionally, deionized water (0, 10, 50 or 100 ml/l) and potassium carbonate (Sigma-Aldrich, 99%, 2.5 mg/l) were added to the feed solution. Catalyst Sn-Beta (Si:Sn 125) prepared according to the above preparation was fractionized (0.25 g, 300-600 m.) and loaded into a stainless steel 0.25 inch reactor. Glass wool was used to hold the catalyst in place. The reactor was introduced into an oven and the temperature of the reactor increased to 160 C. When the temperature was over 140 C., the pump was started with a flow of 0.15 ml/min of a 1.25 wt. % fructose solution in methanol.

(18) Glycolaldehyde (glycolaldehyde dimer, Sigma) was dissolved in methanol (Sigma-Aldrich, 99.9%) at room temperature to reach a concentration of 9 g/l. Additionally, deionized water (0, 10, 30 ml/l) and if necessary methanethiol (Sigma, 1.7 bar) were added to the feed solution. Catalyst Sn-BEA (Si:Sn 125) prepared according to the above preparation was fractionized (0.25 g, 300-600 m) and loaded into a stainless steel 0.25 inch reactor. Glass wool was used to hold the catalyst in place. The reactor was introduced into an oven and the temperature of the reactor increased to 160 C. When the temperature was over 140 C., the pump was started with a flow of 0.05 ml/min (see FIG. 8).

(19) Sucrose syrup (65 wt %, KNO.sub.3 1 g/l) and methanol (Sigma-Aldrich, 99.9%) were pumped separately and mixed at 160 C. to reach a sucrose concentration of 55 g/l. Catalyst Sn-BEA (Si:Sn 125) prepared according to the above preparation was extruded (40 g, 1/32 cylinders) and loaded into a stainless steel reactor. Glass wool was used to hold the catalyst in place. The reactor was introduced into an oven, and the temperature of the reactor increased to 160 C. (see FIG. 9).

(20) Samples were collected after different times on stream and analysed by HPLC (Agilent 1200, Biorad Aminex HPX-87H column at 65 C., 0.05 M H.sub.2SO.sub.4, 0.6 ml/min) to quantify unconverted hexoses and dihydroxyacetone (DHA), glyceraldehyde (GLA); and GC (Agilent 7890 with a Phenomenex Solgelwax column) was used to quantity: methyl lactate (ML), methyl vinyl glycolate (MVG, methyl 2-hydroxy-3-butenoate), glycolaldehyde dimethylacetal (GLAD) and sulfur-methylvinylglyclate (S-MVG, methyl 2-hydroxy-4-(methylthio)-butanoate).

EXAMPLE 2

(21) Determination of the Total Amount of Soluble Tin in the Liquid Medium:

(22) The determination of the total amount of soluble tin (Sn) was carried out using inductively coupled plasma mass spectrometry (ICP-MS). The methanol sample was diluted by weight with an 80/20 xylene/2-propanol mixture. The total Sn content is quantified by ICP-MS (Agilent 7500ce ICP-MS instrument) at the Sn isotope masses 118 and 120 by comparison with a calibration curve made from a 900 ppm Conostan organo-metallic Sn standard diluted with xylene. Indium is used as an internal standard to correct for drift and matrix effects. Removal of molecular interferences in the ICP-MS analysis is done with Helium kinetic energy discrimination. EnviroMAT Used oil certified reference standard which gives an informational value for Sn (305 mg/kg) is analyzed with each sample batch to verify the precision of the method.

EXAMPLE 3

(23) This example illustrates the conversion of C2-sugars (glycolaldehyde) to MVG with increased yield due to the effect of water in batch experiments.

(24) Catalytic Reactions in Batch:

(25) A stainless steel pressure vessel (40 cm.sup.3, Swagelok) was charged with 15.0 g of methanol (Sigma-Aldrich, >99.8%), the required amount of water, 0.200 g glycolaldehyde (glycolaldehyde dimer, Sigma) and 0.150 g of catalyst. The reactor was closed and heated to 160 C. under stirring (900 rpm). The reaction was continued for 16 hours, and after this period, the reaction was quenched by submerging the vessel in cold water. Samples from the reaction vessel were filtered and analyzed by HPLC (Agilent 1200, Biorad Aminex HPX-87H column at 65 C., 0.05 M H.sub.2SO.sub.4, 0.5 ml/min) to quantify unconverted glycolaldehyde (GA); and GC (Agilent 7890 with a Phenomenex Solgelwax column was used to quantify the following: Methyl lactate (ML), methyl vinylglycolate (MVG, methyl-2-hydroxy-3-butenoate), glycolaldehyde dimethylacetal (GLAD) and methyl-4-methoxy-2-hydroxybutanoate (MMHB).

(26) Table 2 shows the effect of the amount of water in batch experiments from glycolaldehyde using Sn-Beta in methanol. An improved yield of methyl vinylglycolate (MVG) and methyl-4-methoxy-2-hydroxybutanoate (MMHB) is obtained with the addition of water to the batch reaction. Methyl vinylglycolate (MVG) is equivalent to 2-hydroxy-3-butenoic acid methyl ester.

(27) TABLE-US-00002 TABLE 2 Effect of the amount of water in batch experiments from glycolaldehyde using Sn-BEA in methanol Batch Wt % GLAD MVG MMHB Total C4 exp. No water yield yield yield yield 1 0 35% 32% ~11% 43% 2 3 wt % 0 52% ~14% 67% 3 8.5 wt % 0 55% ~15% 70% 4 21 wt % 0 46% ~12% 58%

(28) In Table 3, the effect of the presence (amount) of alkali in batch experiments from glycolaldehyde using Sn-Beta in methanol is shown. An improved yield of methyl vinylglycolate (MVG) and methyl-4-methoxy-2-hydroxybutanoate (MMHB) is obtained with the addition of water to the batch reaction in the absence of alkali. This experiment shows that water is the component responsible for the increase in yields, while the presence of alkali is less important. It is, however, preferred to operate in the absence of alkali. The results in batch experiment No. 7 are comparable to conditions mentioned in Green Chemistry 2012, 14, p. 702. Results from said paper: ML 16%, MVG 27%, MMHB 6%.

(29) TABLE-US-00003 TABLE 3 Effect of the presence (amount) of alkali in batch experiments from glycolaldehyde using Sn-BEA in methanol mM K.sub.2CO.sub.3 Total Batch Water in GLAD ML MVG MMHB C4 exp. No wt % MeOH yield yield yield yield yield 3 8.5 0 0 0 55% ~15% 70% 6 8.5 0.13 0 0 48% ~16% 66% 1 0 0 35% 0 32% ~11% 43% 7 0 0.13 0 13% 37% ~15% 52%

(30) Table 4 shows the effect of the type of catalyst in batch experiments from glycolaldehyde using different stannosilicates in methanol. Optimum yield of methyl vinylglycolate (MVG) and methyl-4-methoxy-2-hydroxybutanoate (MMHB) is obtained with Sn-BEA as catalyst.

(31) TABLE-US-00004 TABLE 4 Effect of the type of catalyst in batch experiments from glycolaldehyde using Sn-silicates in methanol Total Batch GLAD ML MVG MMHB C4 exp. No Catalyst yield yield yield yield yield 3 Sn-Beta 0 0 55% ~15% 70% 8 Sn-SBA-15 8% 3% 2% 2% 4% 9 Sn-MCM-41 1% 6% 6% 8% 14%

EXAMPLE 4

(32) This example relates to catalytic reactions in batch to produce sulfur-methyl vinylglycolate (S-MVG, methyl 2-hydroxy-4-(methylthio)-butanoate) and sulfur-ethyl vinylglycolate (S-EVG, ethyl 2-hydroxy-4-(methylthio)-butanoate) from C2-sugars (glycolaldehyde).

(33) 1.6 g of an aqueous solution containing glycolaldehyde (34.2 g/l) produced from commercial glycolaldehyde (Glycolaldehyde dimer, Sigma) or obtained from fragmentation of a 40 wt % glucose syrup (WO 2014/131743) was mixed either with pure methanol (13.8 g; Sigma-Aldrich 99.9%) or with pure ethanol (13.8 g; CCS Healthcare 99.9%). Then 0.16 g of catalyst and the desired amount of methanethiol (Sigma, 1.7 bar) were added, and the mixture was reacted in a pressure reactor at 160 C. (temperature of the oil bath) with 900 rpm stirring under autogenous pressure. An initial sample of the reaction mixture was used for calculation of the conversion and the yields. Samples were collected after 16 h of reaction and analysed by HPLC (Agilent 1200, Biorad Aminex HPX-87H column at 65 C., 0.05 M H.sub.2SO.sub.4, 0.6 ml/min) to quantify unconverted C2 sugars and formed C4 sugars; and GC (Agilent 7890 with Phenomenex Solgelwax column) was used to quantify the following: Methyl vinylglycolate (MVG, methyl 2-hydroxy-3-butenoate), ethyl vinylglycolate (EVG, ethyl 2-hydroxy-3-butenoate), sulfur-methyl vinylglycolate (S-MVG, methyl 2-hydroxy-4-(methylthio)-butanoate) and sulfur-ethyl vinylglycolate (S-EVG, ethyl 2-hydroxy-4-(methylthio)-butanoate).

(34) Table 5 shows batch reactions of the conversion of glycolaldehyde to esters of methionine -hydoxy analogue with Sn-BEA as catalyst in the presence of water and solvent. Amount of methanethiol: 3.6 mmol. Optimum yield was obtained in methanol to S-EVG.

(35) TABLE-US-00005 TABLE 5 Batch reactions of the conversion of glycolaldehyde to esters of methionine -hydroxy analogue [Methyl vinylglycolate (MVG, methyl 2-hydroxy-3-butenoate), ethyl vinylglycolate (EVG, ethyl 2-hydroxy-3-butenoate), sulfur- methyl vinylglycolate (S-MVG, methyl 2-hydroxy-4- (methylthio)-butanoate) and sulfur-ethyl vinylglycolate (S- EVG, ethyl 2-hydroxy-4-(methylthio)-butanoate)] Yield of Batch -hydroxy MVG or exp. methionine EVG No Solvent analogue yield Conversion Selectivity 10 MeOH 39.4% S- 13.9% 98.6 39.9% S-MVG MVG (MVG) 11 EtOH 47.3% S- 19.5% 100 47.3% S-EVG EVG (EVG)

(36) Table 6 shows batch reactions of the conversion of glycolaldehyde to esters of methionine -hydoxy analogue with Sn-Beta as catalyst in the presence of water and methanol. The results illustrate the effect of the GA/thiol molar ratio. Optimum yield to S-MVG is obtained with a molar ratio of 0.8.

(37) TABLE-US-00006 TABLE 6 Batch reactions of the conversion of glycolaldehyde to esters of -hydroxy methionine analogue (S-MVG, methyl 2-hydroxy-4-(methylthio)-butanoate) Batch experiment GA/thiol No (molar ratio) S-MVG yield MVG yield 10 0.8 37.5 7.3 12 0.6 33.6 0 13 1 30.6 7.3 14 5 18.1 33.1 16 0.2 8.3 0 17 No thiol 0 55.0