Improved process for producing 1,2-alkanediol from the corresponding alkene and hydrogen peroxide

Abstract

The present invention relates to a process for preparing an 1,2-alkanediol from the corresponding alkene and hydrogen peroxide which does not require isolation and purification of the intermediate alkene oxide.

Description

[Technical Field]

[0001] Field of Invention

[0002] This invention relates to a method for preparing 1,2-alkanediol from alkene and hydrogen peroxide, which does not require separation and purification of intermediate alkene oxide. [Previous Technology]

[0003] Background of the Invention

[0004] On an industrial scale, the conventional method for preparing 1,2-alkyldiol, such as propylene glycol, involves first oxidizing an olefin to the corresponding olefin oxide, then separating and purifying the olefin oxide, and finally converting it into the corresponding 1,2-alkyldiol through hydrolysis.

[0005] US 10,214,471 B2 relates to a method for the direct production of 1,2-propanediol from propylene and hydrogen peroxide using a combination of a phase-transfer catalyst and a heteropolytungstate in a two-phase reaction mixture. This method eliminates the need for separation and purification of the intermediate propylene oxide, resulting in reduced equipment costs. In a first step of the continuous method, the propylene reacts with hydrogen peroxide in a liquid two-phase system comprising an aqueous phase having a pH of up to 6 and an organic phase, in the presence of a catalyst mixture comprising a phase-transfer catalyst and a heteropolytungstate. In a second step, the organic phase containing the propylene oxide is separated from the aqueous phase and recycled back to the reaction mixture of the first step, and the 1,2-propanediol is separated from the aqueous phase obtained in the second step. Various phase transfer catalysts, including tertiary amines and tetraalkylammonium salts such as trimethylamine and dodecyltrimethylammonium salt, are described in US 10,214,471 B2. For example, Example 9 of US 10,214,471 describes a continuous method for preparing 1,2-propanediol from propylene and hydrogen peroxide using trioctylamine, with a yield of 52% in the aqueous phase.

[0006] J. Kaur et al. describe the epoxidation of propylene with hydrogen peroxide in a microemulsion or a two-phase solvent mixture catalyzed by peroxo polytungstophosphates, which is prepared by the interaction of H₂WO₄, H₂O₂ and H₃PO₄ in the presence of methyltrioctylammonium chloride as a surfactant (Catal. Commun. 2004, 5, 709-713 DOI: 10.1016 / j.catcom.2004.09.004).

[0007] C. Venturello et al. describe a method for the epoxidation of alkenes such as 1-octene and cyclohexene using hydrogen peroxide in a two-phase solvent mixture in the presence of a polytungsten phosphate catalyst and a tetraalkylammonium phase transfer catalyst such as methyltrioctylammonium chloride (J. Org. Chem. 1983, 48, 3831-3833).

[0008] In the above two-phase method, the formation of a stable emulsion is sometimes observed, making the separation of the aqueous phase and the organic phase both time-consuming and energy-intensive. [Summary of the Invention]

[0009] Summary of the Invention

[0010] Therefore, one object of the present invention is to provide an improved biphasic method for producing 1,2-alkanediols such as 1,2-propanediol from corresponding alkenes, which avoids the disadvantages of prior art methods.

[0011] It has now been surprisingly discovered that the use of a specific tetraalkylammonium cation in this biphasic reaction mixture can prevent the formation of a stable emulsion without impairing the efficiency, yield, or selectivity of the epoxidation process. The tetraalkylammonium cation has the general formula N + R1R2R3R4, wherein R1, R2, R3, and R4 are alkyl groups, which may be independently the same or different. When all R1, R2, R3, and R4 are the same, the tetraalkylammonium cation has a total of at least 32 carbon atoms in the alkyl groups; or, when at least one of R1, R2, R3, and R4 is different from the others, the tetraalkylammonium cation has a total of at least 37 carbon atoms in the alkyl groups.

[0012] Therefore, the present invention relates to a method for preparing 1,2-alkyldiol from a corresponding alkene and hydrogen peroxide, the method comprising the steps of: a) reacting the alkene with hydrogen peroxide in the presence of at least one tungsten polyoxometalate and at least one tetraalkylammonium cation, wherein the reaction is carried out in a biphasic reaction mixture comprising an aqueous phase having a pH of at most 6 and an organic phase, wherein the tetraalkylammonium cation has the general formula N + R1R2R3R4, wherein R1, R2, R3 and R4 are alkyl groups, which may be independently the same or different, wherein when R1, R2, R3 and R4 are all the same, the tetraalkylammonium cation has a total of at least 32 carbon atoms in the alkyl groups, or when R1, R2, R3 and R4 are all the same, the tetraalkylammonium cation has at least 32 carbon atoms in total ... When at least one of 4 is different from the others, the tetraalkylammonium cation has a total of at least 37 carbon atoms in the alkyl groups; b) separating the biphase mixture from step a) into an aqueous phase P1 and an organic phase P2; c) optionally recycling the olefin oxides that may be present in the separated organic phase P2 back to reaction step a); and d) separating 1,2-alkyldiol from the aqueous phase P1 separated in step b).

Implementation Method

[0023] As used herein, the term "comprising" and its variations are used synonymously with the terms "including," "containing," and their variations, and are understood to be open-ended and non-limiting terms that do not exclude the presence of additional undescribed or unrepresented elements, compounds, ingredients, or method steps. As used herein, the term "constituting of" is understood to exclude the presence of unspecified elements, compounds, ingredients, or method steps. When the non-limiting terms "comprising," "including," or "containing" are used in this specification, "constituting of" is included therein. For example, a method described as "comprising" or "including" certain steps may consist of expressly described steps or may further include one or more unrepresented steps. The same applies to, for example, compositions and their corresponding expressly described or unrepresented ingredients.

[0024] As used herein, the term "alkene" is understood to be an unsaturated (with a carbon-carbon double bond) acyclic branched or unbranched hydrocarbon having the general formula CnH2n, or a cyclic hydrocarbon having the general formula CnH2n-2. Suitable alkenes include, but are not limited to, propylene, butene, hexene, octene, and cyclohexene. The alkene is preferably propylene.

[0025] As used herein, the term "alkyl" is understood to be a non-cyclic, saturated, branched or unbranched group having the general formula CnH2n+1 and consisting of hydrogen and carbon, preferably unbranched. The alkyl group may be substituted or unsubstituted, but preferably unsubstituted.

[0026] Unless otherwise stated, ambient temperature refers to room temperature (23°C).

[0027] In step a) of the method of the present invention, the alkene is reacted with hydrogen peroxide in the presence of at least one tungsten polyoxometalate and at least one tetraalkylammonium cation in a catalytic amount. The reaction is carried out in a liquid mixture comprising two liquid phases, namely an aqueous phase having a pH of up to 6 and an organic phase.

[0028] The alkene may be used in pure form or as a mixture with the corresponding alkane, wherein the proportion of the alkane may be up to 20 mol% based on the combined amount of the alkene and the alkane, but preferably less than 5 mol%.

[0029] Hydrogen peroxide can preferably be used in the form of an aqueous solution, preferably having a hydrogen peroxide content of 10 to 80% by weight, and more preferably 30 to 70% by weight, based on the total weight of the aqueous hydrogen peroxide solution. The hydrogen peroxide can be prepared by any conceivable method, including but not limited to anthraquinone methods. For example, the crude hydrogen peroxide product obtained in the extraction stage of the anthraquinone method for producing hydrogen peroxide can be used.

[0030] During the reaction process, the aqueous phase will contain water, unreacted hydrogen peroxide, and the 1,2-alkyldiol formed during the reaction. The organic phase will contain the alkene and the alkene oxide formed as an intermediate. If a mixture of an alkene and a corresponding alkane, such as propylene and propane, is used, the organic phase may additionally contain the alkane. Furthermore, the organic phase may contain at least one solvent that is immiscible with water.

[0031] The reaction mixture of the present invention comprises a tungsten polyoxometalate, preferably a heteropolytungstate. The heteroatom of the heteropolytungstate is preferably phosphorus or arsenic, and more preferably phosphorus; that is, the heteropolytungstate is particularly preferably a polytungstate phosphate. The polytungstate phosphate preferably has a phosphorus to tungsten molar ratio in the range of 1:2 to 1:12. A pre-formed tungsten polyoxometalate may be added to the reaction in step a), or the tungsten polyoxometalate may be generated in situ in the liquid reaction mixture of step a). For example, a polytungstate phosphate may preferably be generated in situ from phosphoric acid and a water-soluble basic tungstate, preferably sodium tungstate, wherein the phosphoric acid and basic tungstate are preferably used in a phosphorus to tungsten molar ratio in the range of 1:2 to 10:1, and particularly in the range of 3:1 to 8:1. Peroxytungstate and peroxytungstate phosphate, such as PO4[WO(O2)2]43- and HPO4[WO(O2)2]22- and their partially protonated forms, are formed in the aqueous phase from a polytungstate phosphate and hydrogen peroxide.

[0032] The reaction mixture used in the method according to the invention further comprises at least one tetraalkylammonium cation. The tetraalkylammonium cation has the general formula N + R1R2R3R4, wherein R1, R2, R3 and R4 are alkyl groups, which may be the same or different independently, wherein when all R1, R2, R3 and R4 are the same, the tetraalkylammonium cation has a total of at least 32 carbon atoms in the alkyl group, or when at least one of R1, R2, R3 and R4 is different from the others, the tetraalkylammonium cation has a total of at least 37 carbon atoms in the alkyl group. If a mixture of different tetraalkylammonium cations is used, the average total number of carbon atoms is preferably at least 32 for those tetraalkylammonium cations in which R1, R2, R3 and R4 are all the same, and the average total number of carbon atoms is preferably at least 37 for those tetraalkylammonium cations in which at least one of R1, R2, R3 and R4 is different from the others.

[0033] When at least one of R1, R2, R3, and R4 is different from the others, R1 is preferably CH3, while R2, R3, and R4 are the same, and preferably the same C12-C18 alkyl group. When all of R1, R2, R3, and R4 are the same, each of R1, R2, R3, and R4 is preferably the same C8-C18 alkyl group. The tetraalkylammonium cation may include a tetraalkylammonium cation, which includes at least one of tetraoctylammonium, tetra-dodecylammonium, tri-dodecylmethylammonium, or mixtures thereof.

[0034] The tetraalkylammonium cation can form an organic phase soluble salt with the tungsten polyoxometalate.

[0035] The tetraalkylammonium cation is preferably added to the reaction in the form of a quaternary tetraalkylammonium salt having an anion, different from the tungsten polyoxometalate, such as, but not limited to, tetraalkylammonium halides (e.g., monochloride, monobromide, or monoiodide), sulfates, or methyl sulfates. Using a salt having a methyl sulfate as an anion, compared to a tetraalkylammonium halide, can reduce the corrosivity of the reaction mixture. The tungsten polyoxometalate and the tetraalkylammonium salt can be fed into the reaction in step a) as a mixture or separately, preferably separately. However, a pre-formed quaternary tetraalkylammonium tungsten polyoxometalate may also be added, wherein the tungsten polyoxometalate represents the counterion of the quaternary tetraalkylammonium cation.

[0036] The tetraalkylammonium cation can preferably be used in an amount such that the molar ratio of the tetraalkylammonium cation to tungsten in the liquid mixture is in the range of 1:1 to 3:1.

[0037] The reaction of the alkene with hydrogen peroxide is carried out at a pH of up to 6 of the aqueous phase. The pH of the aqueous phase is preferably maintained in the range of 1.0 to 3.5, particularly in the range of 2.0 to 3.0, and is measured at room temperature. This pH range can be maintained by adding an acid, preferably sulfuric acid or phosphoric acid, or by adding a base, preferably an aqueous sodium hydroxide solution. The term pH herein refers to the apparent pH measured using a glass electrode calibrated with an aqueous buffer solution. By adjusting the pH to this preferred range, high selectivity for 1,2-alkyldiols can be achieved, and the enrichment of the corresponding alkene oxide in the aqueous phase can be prevented, which simplifies the subsequent separation of 1,2-alkyldiols from the aqueous phase.

[0038] The reaction of the alkene with hydrogen peroxide is preferably carried out using a mole excess of the alkene, wherein the alkene is preferably used in a mole ratio of alkene to hydrogen peroxide from 1.1:1 to 10:1.

[0039] The reaction is preferably carried out at a temperature in the range of 30 to 100°C, and more preferably at 70 to 90°C. The reaction is preferably carried out at a pressure higher than the saturated vapor pressure of the alkene at the reaction temperature, such that the majority of the alkene is present in the organic phase of the liquid mixture.

[0040] The reaction of the alkene with hydrogen peroxide can be carried out with or without the addition of a solvent. The reaction is preferably carried out in the presence of at least one solvent having a boiling point greater than 100°C, preferably greater than 120°C, and a water solubility of less than 0.5 g / kg at 20°C, preferably less than 250 mg / kg. The solvent used can be a halogenated hydrocarbon, such as 1,2-dichloroethane, and / or an aromatic hydrocarbon, such as an alkylated aromatic hydrocarbon. Preferably, the solvent or any other compound present in the reaction mixture of step a) does not contain ester functionality. The solvent preferably contains or is an aromatic hydrocarbon, such as an alkylated aromatic hydrocarbon having 7 to 12 carbon atoms or a mixture of such alkylated aromatic hydrocarbons. Suitable alkylated aromatic hydrocarbon systems include toluene, 1,2-dimethylbenzene (o-xylene), 1,3-dimethylbenzene (m-xylene), 1,4-dimethylbenzene (p-xylene), ethylbenzene, 1,2,3-trimethylbenzene, 1,2,4-triyltoluene, 1,3,5-trimethylbenzene (mesotrimethylbenzene), 1-ethyl-2-methylbenzene, 1-ethyl-3-methylbenzene and 1-ethyl-4-methylbenzene, n-propylbenzene, and 1,2,3,4-tetrahydronaphthalene, and mixtures thereof. Based on the total weight of the solvent mixture, a mixture containing more than 50% by weight, preferably more than 80% by weight, of alkylated aromatic hydrocarbons having 7 to 12 carbon atoms is preferably used as a solvent. By using a solvent containing an alkylated aromatic hydrocarbon having 7 to 12 carbon atoms, the heteropolytungstate in the form of a tetraalkylammonium complex can be extracted in large quantities into the organic phase of the reaction mixture, thereby improving the recycling of the heteropolytungstate and the organic phase, and enabling the simple recovery of the heteropolytungstate from the aqueous phase of the reaction of the alkene with hydrogen peroxide.

[0041] Preferably, the proportion of the solvent can be selected such that, based on the total weight of the organic phase, the proportion of the solvent in the organic phase during the reaction is in the range of 10 to 90% by weight.

[0042] The reaction of the alkene with hydrogen peroxide can be carried out in batches or continuously, with a continuous method being preferred. In a continuous method, the concentration of hydrogen peroxide in the aqueous phase is preferably in the range of 0.1% to 5% by weight, and particularly preferably in the range of 0.5% to 3% by weight, based on the total weight of the aqueous phase. This concentration of hydrogen peroxide can be adjusted by selecting the reaction temperature, the molar ratio of the alkene to hydrogen peroxide, and the residence time of the liquid mixture in the reactor in which the reaction occurs.

[0043] During the reaction, the liquid mixture is preferably mixed to create a large phase interface between the aqueous phase and the organic phase. For this purpose, the reaction is preferably carried out continuously in a loop reactor having fixed internal components, and the liquid mixture is passed through the loop reactor at a flow rate that generates turbulence in the internal components. A loop reactor suitable for continuously carrying out the method of the present invention is described in detail in U.S. Patent No. 10,214,471 B2, particularly in Figures 1 to 4 and columns 9, lines 42 to 10, lines 51.

[0044] In step b) of the method of the present invention, the two-phase reaction mixture from step a) is separated into an aqueous phase P1 and an organic phase P2. This separation is preferably carried out in a setter vessel, and the two-phase mixture may pass through a coalescer element comprising a structured packing or a random packing having a surface wetted by the dispersed phase in the two-phase mixture to facilitate the separation.

[0045] The liquid phase is preferably separated in step b) in the presence of a gas phase. The reaction in step a) can lead to the decomposition of hydrogen peroxide and the formation of oxygen, and the gas phase in step b) may then contain oxygen. To avoid the formation of a flammable gas phase, the oxygen content of this gas phase in step b) is preferably maintained at less than 7% by volume, preferably by supplying an inert gas and venting a gas stream. The inert gas used can be nitrogen, argon, carbon dioxide, or methane, with nitrogen being preferred.

[0046] If present, in any step c) of the method of the present invention, the alkene oxide present in the organic phase P2 is recycled to the reaction in step a) so that the alkene is converted to 1,2-alkyldiol as completely as possible. If the method of the present invention is operated continuously, step c) is preferably present. Preferably, the tungsten polyoxometalate present in the organic phase P2 may be additionally recycled to the reaction in step a). Similarly, the alkene present in the organic phase P2 may preferably be recycled to the reaction in step a). If the alkene is used as a mixture with the corresponding alkane, the same amount of the alkane is separated from the organic phase P2 in the recycling to step a), which is fed into step a) together with the mixture of the alkene and the alkane. In this way, when the reaction in step a) is carried out continuously, the enrichment of the alkane in the organic phase in step a) can be avoided.

[0047] In a preferred embodiment of the method of the present invention, the organic phase P2 separated in step b) may be recycled in whole or in part to the reaction in step a).

[0048] Preferably, the organic phase P2 can be completely or partially separated in step c) by nanofiltration into a residue rich in tungsten polyoxometalates and a permeate poor in tungsten polyoxometalates, and the residue is recycled to the reaction in step a). Preferably, the entire organic phase P2 can be separated into a residue and a permeate by nanofiltration. For the nanofiltration in step c), a nanofiltration membrane can be used, which retains the tungsten polyoxometalates and the tetraalkylammonium cation in the residue and allows the alkene to pass through with the permeate. The nanofiltration can then preferably be operated such that the concentration of tungsten polyoxometalates and the tetraalkylammonium cation in the residue does not increase above the saturation concentration. As used herein, the term "nanofiltration" refers to a pressure-driven separation at a membrane, wherein the membrane retains particles having a diameter of less than 2 nm and dissolved molecules. Membranes based on polymers such as polyimide, polyether ether, polyamide, and polydimethylsiloxane can be used for nanofiltration. Suitable nanofiltration membranes are commercially available, for example, from Evonik Membrane Extraction Technology MET (PuraMem® S600), from GMT Membrantechnik (ONF-2), from SolSep (010306, 030306, 030705, and 030306F), and from AMS Technologies (NanoPro™ SX). Preferably, one of these composite membranes known from DE 195 07 584, EP 1 741 481, and WO 2011 / 067054 is used.

[0049] This nanofiltration preferably occurs as a cross-flow filtration, preferably at a temperature in the range of 20 to 90°C, and particularly preferably in the range of 40 to 80°C. The transmembrane pressure is preferably in the range of 2 to 5 MPa. The pressure on the permeate side can be as high as 10 MPa. The pressure on the permeate side is preferably selected to be higher than the minimum pressure in steps a) and b) of the method to prevent the release of dissolved components on the permeate side.

[0050] The organic phase P2 can also be separated by nanofiltration into a permeate rich in tungsten polyoxometalates and a permeate poor in tungsten polyoxometalates. A stream S1 containing unreacted olefins and olefin oxides formed as intermediates is separated from the permeate by distillation, and this stream can be recycled to the reaction in step a). The distillation is preferably carried out at a pressure under which the olefins can be condensed together with the distillate by water cooling. Alternatively, the distillation can be carried out at a lower pressure, under which the olefin oxides and only a portion of the olefins can be condensed together with the distillate, and the remaining vapor can be compressed to condense the olefins. In this embodiment, high-boiling-point byproducts and degradation products of the phase transfer catalyst can be discharged together with the bottom product of the distillation, and when the reaction in step a) is carried out continuously, the enrichment of poorly water-soluble byproducts and impurities in the organic phase in step a) can be avoided. Removing stream S1 by distillation after this nanofiltration process prevents the olefin oxides formed as intermediates from reacting further with the catalyst system by heating, which would lead to byproducts.

[0051] If the reaction is carried out in the presence of a solvent in step a) of the method, preferably, after the distillation for separating stream S1, the bottom product of this distillation is fed into a further distillation, wherein the solvent is separated by distillation. The separated solvent can be recycled to step a). If the alkene is used as a mixture with the corresponding alkane, the distillation is preferably carried out such that a further stream is obtained in addition to stream S1, which is essentially composed of the alkene and the alkane, from which the alkene oxide is separated. The alkane can be completely or partially separated from this further stream, and the resulting alkene, separated from the alkane or depleted of the alkane, can preferably be recycled to step a). In this case, preferably, the same amount of alkane is removed and fed into step a) together with the mixture of alkene and alkane. For this purpose, the distillation of the permeate can be carried out in two stages, wherein in the first distillation stage, the further stream, which is mostly composed of alkenes and alkanes, is separated, and stream S1 is subsequently separated in the second distillation stage. However, the distillation is preferably carried out in a column with a side draw device, where stream S1 is removed as a side draw, and the further stream, which is substantially composed of alkenes and alkanes, is removed as the overhead product of the column.

[0052] The organic phase P2 can also be separated into stream S1 and stream S2 by distillation. Stream S1 contains unreacted olefins and olefin oxides formed as intermediates, and stream S2 is depleted of olefins and olefin oxides. Stream S2 can be separated into a permeate rich in heteropolytungstates and a permeate depleted in tungsten polyoxometalates by nanofiltration. Stream S1 can be recycled to the reaction in step a). This embodiment is preferably used if the reaction in step a) of the method is carried out in the presence of a solvent and then the distillation is performed so that the solvent is retained in stream S2. The solvent can then be removed from the permeate by nanofiltration, preferably by distillation, and can then be recycled to the reaction in step a).

[0053] Compared to the above embodiment, the embodiment of separating stream S1 before the nanofiltration has the advantage that a relatively small stream is separated by the nanofiltration, which reduces the size of the instrument and the energy consumption of the nanofiltration. If the alkene is used as a mixture with an alkane, in this embodiment a further stream is preferably separated from stream S1 by an additional distillation process, which is essentially composed of alkene and alkane, and from which alkene oxides are separated before stream S1 is recycled to step a). The alkane can be completely or partially separated from this further stream, and the resulting alkene, separated from the alkane or depleted of alkane, is preferably recycled to step a). In this case, it is preferable that the same amount of alkane is separated and fed into step a) together with the mixture of alkene and alkane. For a continuous method including step c), refer to US 10,214,471 B2, and in particular the figures and paragraphs already cited above.

[0054] In step d) of the method of the present invention, the 1,2-alkyldiol is separated from the aqueous phase P1 separated in step b). The 1,2-alkyldiol can be separated from the aqueous phase by distillation, preferably by a one- or two-stage distillation, wherein the aqueous phase is distilled off in the first stage and the 1,2-alkyldiol is distilled off from the bottom product of the first stage in the second stage.

[0055] Prior to the separation of 1,2-alkyldiol, the peroxide is preferably removed by a catalytic hydrogenation. This hydrogenation is preferably carried out using a supported hydrogenation catalyst comprising one or more metals from the group consisting of Ru, Rh, Pd, Pt, Ag, Ir, Fe, Cu, Ni, and Co on a support, wherein activated carbon, SiO₂, TiO₂, ZrO₂, Al₂O₃, and aluminum silicate are preferably used as supports. Preferably, the hydrogenation catalyst comprises ruthenium as the active metal. This catalytic hydrogenation is preferably carried out at a hydrogen partial pressure of 5 to 300 bar and a temperature of 80°C to 180°C, more preferably from 90°C to 150°C. The hydrogenation catalyst can be used as a suspension or as a fixed bed; a trickle-bed hydrogenation with a fixed-bed catalyst is preferred. This hydrogenation process prevents problems caused by the decomposition of hydrogen peroxide in the separation of 1,2-alkyldiol by distillation, and reduces the byproducts 1-hydroperoxy-2-alkanol, 2-hydroperoxy-1-alkanol and 1-hydroxy-2-alkanone formed in step a) to 1,2-alkyldiol, thereby improving the yield of 1,2-alkyldiol.

[0056] In step d) of the method of the present invention, the aqueous phase P1 is preferably separated by nanofiltration into a permeate rich in tungsten polyoxometalates and a permeate poor in tungsten polyoxometalates, wherein the permeate can be recycled to the reaction in step a), and the 1,2-alkyldiol is separated from the permeate. For the nanofiltration in step d), a nanofiltration membrane can be used, which retains the tungsten polyoxometalates in the permeate and allows water and 1,2-propanediol to pass through with the permeate. The nanofiltration is operable such that the solubility limit of the tungsten polyoxometalates in the permeate is not exceeded. For a continuous reaction in step a), preferably a plurality of aqueous systems are recycled to step a) along with the permeate, resulting in the concentration of 1,2-alkyldiol in the aqueous phase P1 being in the range of 10 to 30% by weight. By properly recycling water, the formation of 1,1'-oxydi-2-alkylol and trialkyldiol in step a) can be prevented, and the amount of water that must be separated from 1,2-alkyldiol by distillation can be kept at a low level.

[0057] In addition to or as an alternative to nanofiltration, tungsten polyoxometalates can be removed from the aqueous phase P1 by adsorption on a support material. For this adsorption, a cationic inorganic support material described in WO 2009 / 133053 A1, page 7, line 1 to page 8, line 29, is preferably used. Preferably, the adsorption on the support material and the recovery of the tungsten polyoxometalates adsorbed on the support material can be carried out using the methods described in WO 2009 / 133053 A1 and WO 2013 / 110419 A1. If the adsorption is used in addition to the nanofiltration in step d), the adsorption is preferably carried out after the nanofiltration to maintain a low demand on the support material.

[0058] The nanofiltration and adsorption on a carrier material can preferably be carried out before the hydrogenation process to avoid deactivation of the hydrogenation catalyst by tungsten polyoxometalates.

[0059] In a preferred embodiment of the method according to the invention, the aqueous phase P1 separated in step b) can be contacted with the liquid olefin in step d) to obtain an aqueous phase P3 and an organic phase P4, which can be recycled to the reaction in step a), and the 1,2-alkyldiol system is separated from the aqueous phase P3. In this embodiment, preferably, tungsten is not separated from the aqueous phase P1, and the tungsten system is separated from the aqueous phase P3, preferably by nanofiltration and / or adsorption as described above for the aqueous phase P1. Preferably, the aqueous phase P1 is contacted with the olefin in an additional mixing reactor, in which a residence time results in a conversion of at least 50% of the hydrogen peroxide present in the aqueous phase, preferably at least 80%. The contact in the additional reactor is preferably carried out at a temperature of 60 to 100°C and a pressure higher than the saturated vapor pressure of the olefin at the selected temperature. Preferably, the aqueous phase P1 can be contacted with the olefin in a continuously operating reactor, particularly in a circulating reactor. By using an additional continuously operating reactor, a high hydrogen peroxide conversion rate can be achieved with a relatively low overall reactor volume.

[0060] Here, the alkene is preferably supplied to the method only in step d) and reacted with the organic phase P4 in step a). If a tetraalkylammonium cation is used in the method, which transfers more than half of the tungsten from the aqueous phase to the liquid alkene in the presence of hydrogen peroxide, the aqueous phase P1 can also be contacted with the liquid alkene in a countercurrent extraction, preferably a countercurrent extraction column, and the tungsten present in the aqueous phase P1 is then recycled together with the organic phase P4 to step a). Example I. Phase Separation Experiment Examples 1 to 4 and Comparative Examples 5 to 7:

[0061] To investigate the phase separation behavior of the biphasic epoxidation reaction mixture, an aqueous solution A containing the 1,2-alkyldiol and an organic solution D containing the epoxidation catalyst and a tetraalkylammonium salt were prepared and mixed as described below. Phase separation after mixing was observed by recording the volumes of the separated organic and aqueous layers over time. These results are shown in Table 1 and Figures 1 to 7.

[0062] Preparation of aqueous solution A: A solution of 1,2-propanediol (20 wt.%) and phosphoric acid (0.06 wt.%) in HPLC-grade demineralized water is prepared by mixing.

[0063] Preparation of POM catalyst solution B: Sodium tungstate dihydrate (13.0 g, 39.4 mmol) was dissolved in HPLC-grade demineralized water (62.4 g). The solution was stirred at ambient temperature using a PFTE-coated magnetic stir bar while 85% phosphoric acid (20.8 g, 0.181 mol) was gradually added.

[0064] Preparation of tetraalkylammonium solution C: The respective tetraalkylammonium salt (see Table 1) (10 mmol) was dissolved in an organic solvent (60 g).

[0065] Preparation of tetraalkylammonium / POM catalyst solution D: All of the tetraalkylammonium solution C was combined with 22.2 g of the POM catalyst solution B. The resulting biphase solution was mixed for 15 minutes at ambient temperature using a PFTE-coated magnetic stir bar. The phases were allowed to separate and the organic phase was recovered and used in the phase separation experiment described below.

[0066] Experimental procedure for the phase separation experiment: Equal volumes of tetraalkylammonium / POM catalyst solution D and aqueous solution A were placed in a 100 mL graduated cylinder with a 29 / 42 ground glass connector at the top. The contents were mixed five times at ambient temperature by manually inverting the graduated cylinder and returning it to the upright position. Table 1: Phase separation experiment Example Tetraalkylammonium salt organic solvents Time to achieve complete phase separation Example 1 Tetraoctylammonium bromide Trimethylbenzene < 3 min Example 2 Tetraoctylammonium bromide Hydrosol A200 ND 1 < 3 min Example 3 Tetradecylammonium bromide Trimethylbenzene < 4 min Example 4 Tri-dodecylmethylammonium iodide Trimethylbenzene < 9 min Comparative Example 5 Adogen 464 2 Trimethylbenzene > 240 min Comparative Example 6 3 Adogen 464 Trimethylbenzene > 300 min Comparative Example 7 Adogen 464 Hydrosol A200 ND > 240 min 1 : A mixture of low-naphthalene C10-aromatic compounds (from DHC Solvent Chemie GmbH, Mülheim am Ruhr, Germany) 2 Methyltrialkyl (C8-C10) ammonium chloride (from Sigma Aldrich, Inc., St. Louis, Missouri, USA) 3 In this embodiment, the two phases are collected in a bottle and heated overnight in an oven at 50°C. The phases are then transferred to a graduated cylinder and mixed at 50°C. Phase separation at 50°C is recorded.

[0067] When a methyltrialkylammonium salt system is used as a phase transfer catalyst, the methyltrialkylammonium salt having a total of less than 32 carbon atoms in the alkyl groups, the formation of a stable emulsion is observed (Comparative Examples 5-7). Regardless of the temperature and organic solvent used, the aqueous phase and the organic phase system separate slowly. Surprisingly, when a tetraalkylammonium salt having the general formula N + R1R2R3R4 is used, the formation of a stable emulsion can be avoided and the time required to achieve complete phase separation can be significantly reduced, wherein when all R1, R2, R3, and R4 are the same, the tetraalkylammonium cation has a total of at least 32 carbon atoms in the alkyl groups, such as tetraoctylammonium or tetradecylammonium (Examples 1-3), or when at least one of R1, R2, R3, and R4 is different from the others, the tetraalkylammonium cation has a total of at least 37 carbon atoms in the alkyl groups, such as tri-dodecylmethylammonium (Example 4). II. Epoxidation / Hydrolysis Experiment: Example 8:

[0068] A 1-L jacketed glass reactor is used in the catalyst preparation process. It contains a double-bladed impeller agitator, a bottom discharge port, a temperature detector, a temperature-controlled heating / cooling tank, and nitrogen purging.

[0069] Preparation of POM catalyst solution E: Sodium tungstate dihydrate (13.85 g) and 66.35 g of HPLC-grade demineralized water were added to a glass container to form a clear solution. Next, 85% phosphoric acid (22.1 g) was added with stirring. Finally, 50 wt.% aqueous hydrogen peroxide (19.4 g) was added with stirring.

[0070] Preparation of Tetraoctylammonium Solution F: Tetraoctylammonium bromide (25.0 g) and mesitylene (253.6 g) were added to different bottles, and the mixture was stirred under heating until the tetraoctylammonium bromide was completely dissolved. This solution was loaded into the 1-L glass reactor and maintained at 50-60°C. The solution was washed five times with a 50 wt.% sulfuric acid solution, and then bromide-to-sulfate ion exchange was performed using water. The mass of these washing solutions is recorded in Table 2. Table 2: Aqueous washing steps during the preparation of epoxidation / hydrolysis catalysts Aqueous cleaning solution Water content of the filling (g) Removal of water-based mass (g) 50 wt.% sulfuric acid 258 235 50 wt.% sulfuric acid 261 267 50 wt.% sulfuric acid 257 259 50 wt.% sulfuric acid 250 238 50 wt.% sulfuric acid 253 259 Demineralized water 163 144.5

[0071] The ion-exchanged tetraalkylammonium solution F is retained in the reactor, and the POM catalyst solution E (120.5 g) is added to it. The mixture is stirred at 55°C for 30 minutes. Stirring is stopped and the phases are allowed to separate. The aqueous phase (136 g) is removed, and the organic solution (278.5 g) is collected and stored in a refrigerator until further use.

[0072] Epoxidation / Hydrolysis Reaction Procedure: A 1-L stainless steel reactor is used for this epoxidation / hydrolysis reaction, comprising a bottom discharge port, an internal temperature sensor, a heating jacket, an internal cooling coil, an inclined impeller, and sampling lines. A D1000 Isco pump is used for loading propylene.

[0073] The epoxidation / hydrolysis catalyst solution (78.9 g) was diluted with trimethylbenzene (11.1 g) and then loaded into the 1-L stainless steel reactor. Next, a 15 wt.% aqueous hydrogen peroxide solution containing phosphoric acid was added to the reactor (this solution was prepared by mixing 93.8 g of 30 wt.% hydrogen peroxide, 93.8 g of demineralized water, and 0.123 g of 85% phosphoric acid). The reactor was sealed and pressurized with nitrogen (100 psig (689 kPa)) for two cycles of filling and venting to fill the reactor. Then, liquid propylene (139 mL) was added to the reactor by volume. The reactor was heated to 70°C using an electric heating jacket, and the reaction mixture was stirred at 500 rpm. Liquid samples were collected throughout the 300-minute experiment, and the concentration of 1,2-propanediol in the aqueous phase was measured by gas chromatography. The 1,2-propanediol yield (mol%) was calculated according to the following formula: 1,2-propanediol yield (mol%) = Total 1,2-propanediol (mol) / Total hydrogen peroxide (mol) * 100 Where total 1,2-propanediol (mol) = [1,2-propanediol concentration (wt.%) * Mass of aqueous solution in reactor (g)] / Molecular weight of 1,2-propanediol (g / mol)

[0074] These results are summarized in Figure 8. The first sample was collected once the reactor temperature reached a temperature within 2°C of the desired set point; this first sample was marked as 0 minutes, and all subsequent samples were marked chronologically from this first sample. As shown in Figure 8, this catalyst composition produced a high yield (87.2%) of 1,2-propanediol. Example 9:

[0075] A 1-L jacketed glass reactor is used in the catalyst preparation process and contains a double-bladed impeller agitator, a bottom discharge port, a temperature detector, a temperature-controlled heating / cooling tank, and nitrogen purging.

[0076] Preparation of POM catalyst solution G: Sodium tungstate dihydrate (13.8 g) and 66.4 g of HPLC-grade demineralized water were added to a glass container to form a clear solution. Next, 85% phosphoric acid (22.1 g) was added with stirring. Finally, 50 wt.% aqueous hydrogen peroxide (19.4 g) was added with stirring.

[0077] Preparation of Tetraoctylammonium Solution H: Tetraoctylammonium bromide (25.15 g) and mixed xylene (254 g, reagent grade, Sigma Aldrich, St. Louis, Missouri, USA) were added to different bottles, and the mixture was stirred under heating until the tetraoctylammonium bromide was completely dissolved. This solution was loaded into a 1-L glass reactor and maintained at 55°C. The solution was washed five times with a 50 wt.% sulfuric acid solution, and then subjected to bromide-to-sulfate ion exchange with water. The mass of these washing solutions is recorded in Table 3.

[0078] Table 3: Aqueous cleaning steps during the preparation of epoxidation / hydrolysis catalysts. Aqueous cleaning solution Water content of the filling (g) Removal of water-based mass (g) 50 wt.% sulfuric acid 264 234 50 wt.% sulfuric acid 257 259 50 wt.% sulfuric acid 254 260 50 wt.% sulfuric acid 252 252 50 wt.% sulfuric acid 257 259 Demineralized water 187 193

[0079] The ion-exchanged tetraalkylammonium solution G was retained in the reactor, and the POM catalyst solution H (121.0 g) was added. The mixture was stirred at 55°C for 30 minutes. Stirring was stopped, and the phases were allowed to separate. The aqueous phase (112 g) was removed, and the organic solution (282 g) was collected and stored in a refrigerator until further use.

[0080] Epoxidation / Hydrolysis Reaction Procedure: A 1-L stainless steel reactor is used for this epoxidation / hydrolysis reactor and includes a bottom discharge port, an internal temperature sensor, a heating jacket, an internal cooling coil, an inclined blade impeller, and sampling lines. A D1000 Isco pump is used for loading propylene.

[0081] The epoxidation / hydrolysis catalyst solution (80.0 g) was diluted with mixed xylene (10.1 g) and then loaded into the 1-L stainless steel reactor. Next, a 15 wt.% aqueous hydrogen peroxide solution containing phosphoric acid (prepared by mixing 93.8 g of 30 wt.% hydrogen peroxide, 93.8 g of demineralized water, and 0.110 g of 85% phosphoric acid) was added to the reactor. The reactor was sealed and pressurized with nitrogen (100 psig) for two cycles of filling and venting. Then, liquid propylene (139 mL) was added to the reactor by volume. The reactor was heated to 70°C using an electric heating jacket, and the reaction mixture was stirred at 500 rpm. Liquid samples were collected throughout the 300-minute experiment, and the concentration of 1,2-propanediol in the aqueous phase was measured by gas chromatography. The yield of 1,2-propanediol was calculated as described above.

[0082] These results are summarized in Figure 9. The first sample was collected once the reactor temperature reached a temperature within 2°C of the desired set point; this first sample was marked as 0 minutes, and all subsequent samples were marked chronologically from this first sample. As shown in Figure 9, the catalyst composition produced a high yield (80.6%) of 1,2-propanediol. Example 10:

[0083] A 1-L jacketed glass reactor is used in the catalyst preparation process. It contains a double-bladed impeller agitator, a bottom discharge port, a temperature detector, a temperature-controlled heating / cooling tank, and nitrogen purging.

[0084] Preparation of POM catalyst solution I: Sodium tungstate dihydrate (13.8 g) and 66.5 g of HPLC-grade demineralized water were added to a glass container to form a clear solution. Next, 85% phosphoric acid (22.1 g) was added with stirring. Finally, 50 wt.% aqueous hydrogen peroxide (19.4 g) was added with stirring.

[0085] Preparation of tetraoctylammonium solution H: Tetraoctylammonium bromide (25.22 g) and toluene (254 g, HPLC grade - Fisher Scientific) were added to different bottles, and the mixture was stirred at 50°C until the tetraoctylammonium bromide was completely dissolved. This solution was loaded into the 1-L glass reactor and maintained at 50°C. The solution was washed seven times with a 50 wt.% sulfuric acid solution, and then subjected to bromide-to-sulfate ion exchange twice with water. The mass of these washing solutions is recorded in Table 4.

[0086] Table 4: Aqueous cleaning steps during the preparation of epoxidation / hydrolysis catalysts. Aqueous cleaning solution Water content of the filling (g) Removal of water-based mass (g) 50 wt.% sulfuric acid 212 188 50 wt.% sulfuric acid 228 223 50 wt.% sulfuric acid 215 217 50 wt.% sulfuric acid 199 209 50 wt.% sulfuric acid 198 198 50 wt.% sulfuric acid 211 208 50 wt.% sulfuric acid 211 211 Demineralized water 191 198 Demineralized water 199 194

[0087] The ion-exchanged tetraalkylammonium solution I was retained in the reactor, and the POM catalyst solution J (121.5 g) was added. The mixture was stirred at 55°C for 30 minutes. Stirring was stopped and the phases were allowed to separate. The aqueous phase (121 g) was removed, and the organic solution (278.5 g) was collected and stored in a refrigerator until further use.

[0088] Epoxidation / Hydrolysis Reaction Procedure: A 1-L stainless steel reactor is used for this epoxidation / hydrolysis reactor, which includes a bottom discharge port, an internal temperature sensor, a heating jacket, an internal cooling coil, an inclined blade impeller, and sampling lines. A D1000 Isco pump is used for loading propylene.

[0089] The epoxidation / hydrolysis catalyst solution (79 g) was diluted with toluene (11.1 g) and then loaded into the 1-L stainless steel reactor. Next, a 15 wt.% aqueous hydrogen peroxide solution containing phosphoric acid was added to the reactor (this solution was prepared by mixing 93.8 g of 30 wt.% hydrogen peroxide, 93.8 g of demineralized water, and 0.109 g of 85% phosphoric acid). The reactor was sealed and pressurized with nitrogen (100 psig) and vented for two cycles to fill the reactor with nitrogen (100 psig) and these filling and venting cycles. Then, liquid propylene (139 mL) was added to the reactor by volume. The reactor was heated to 70°C using an electric heating jacket, and the reaction mixture was stirred at 500 rpm. Liquid samples were collected during the entire 300-minute experiment, and the concentration of 1,2-propanediol in the aqueous phase was measured by gas chromatography, with the yield of 1,2-propanediol calculated as described above.

[0090] These results are summarized in Figure 10. The first sample was collected once the reactor temperature reached a temperature within 2°C of the desired set point; this first sample was marked as 0 minutes, and all subsequent samples were marked in chronological order from this first sample. As shown in Figure 10, the catalyst composition produced a high yield (87.2%) of 1,2-propanediol.

[0091] As shown in Examples 8 to 10, the catalyst system described above can surprisingly produce 1,2-propanediol with a significantly improved yield of 80% or more. [Simplified Explanation of the Diagram]

[0013] Figure 1 shows the phase separation of a two-phase mixture according to the present invention (Example 1), using tetraoctylammonium as the tetraalkyl cation and mesitylene as the organic solvent at ambient temperature, and recording the interface levels (mL) of the organic phase (upper curve) and the aqueous phase (lower curve) over time (minutes).

[0014] Figure 2 shows the phase separation of a two-phase mixture according to the present invention (Example 2), using tetraoctylammonium as the tetraalkyl cation and Hydrosol A200 ND as the organic solvent at ambient temperature, and recording the interface level (mL) of the organic phase (upper curve) and the aqueous phase (lower curve) over time (minutes).

[0015] Figure 3 shows the phase separation of a two-phase mixture according to the present invention (Example 3), using tetradecylammonium as the tetraalkyl cation and mesitylene as the organic solvent at ambient temperature, and recording the interface level (mL) of the organic phase (upper curve) and the aqueous phase (lower curve) over time (minutes).

[0016] Figure 4 shows the phase separation of a two-phase mixture according to the present invention (Example 4), using tri-dodecylmethylammonium as the tetraalkyl cation and mesitylene as the organic solvent at ambient temperature, and recording the interface levels (mL) of the organic phase (upper curve) and the aqueous phase (lower curve) over time (minutes).

[0017] Figure 5 shows the phase separation of a comparative two-phase mixture (Comparative Example 5), using methyltrialkyl (C8-C10)ammonium as the tetraalkyl cation and mesitylene as the organic solvent at ambient temperature, and recording the interface levels (mL) of the organic phase (upper curve) and the aqueous phase (lower curve) over time (minutes).

[0018] Figure 6 shows the phase separation of a comparative two-phase mixture (Comparative Example 6), using methyltrialkyl (C8-C10)ammonium as the tetraalkyl cation and mesitylene as the organic solvent at 50°C, and recording the interface levels (mL) of the organic phase (upper curve) and the aqueous phase (lower curve) over time (minutes).

[0019] Figure 7 shows the phase separation of a comparative two-phase mixture (Comparative Example 7), using methyltrialkyl (C8-C10)ammonium as the tetraalkyl cation and Hydrosol A200 ND as the organic solvent at ambient temperature, and recording the interface levels (mL) of the organic phase (upper curve) and the aqueous phase (lower curve) over time (minutes).

[0020] Figure 8 shows the mol% yield of 1,2-propanediol obtained by the epoxidation / hydrolysis method according to the present invention (Example 8) as a function of reaction time (minutes).

[0021] Figure 9 shows the mol% yield of 1,2-propanediol obtained in the epoxidation / hydrolysis method according to the present invention (Example 9) as a function of reaction time (minutes).

[0022] Figure 10 shows the mol% yield of 1,2-propanediol obtained in the epoxidation / hydrolysis method according to the present invention (Example 10) as a function of reaction time (minutes).

Claims

1. A method for preparing 1,2-alkyldiol from a corresponding alkene and hydrogen peroxide, comprising the steps of: a) reacting the alkene with hydrogen peroxide in the presence of at least one tungsten polyoxometalate and at least one tetraalkylammonium cation, wherein the reaction is carried out in a biphasic reaction mixture comprising an aqueous phase having a pH of up to 6 and an organic phase, wherein the tetraalkylammonium cation has the general formula N + R1R2R3R4, wherein R1, R2, R3, and R4 are alkyl groups, which may be independently the same or different, wherein... When R1, R2, R3, and R4 are all the same, the tetraalkylammonium cation has a total of at least 32 carbon atoms in the alkyl groups; or when at least one of R1, R2, R3, and R4 is different from the others, the tetraalkylammonium cation has a total of at least 37 carbon atoms in the alkyl groups; b) separating the biphase mixture from step a) into an aqueous phase P1 and an organic phase P2; c) optionally recycling any alkene oxide that may be present in the separated organic phase P2 back to reaction step a); and d) separating 1,2-alkyldiol from the aqueous phase P1 separated in step b).

2. The method of claim 1, wherein the 1,2-alkyldiol is 1,2-propanediol and the corresponding olefin is propylene.

3. The method of claim 1 or claim 2, wherein the tungsten polyoxometalate is a heteropolytungstate, preferably a polytungstophosphate.

4. The method of claim 3, wherein the polytungsten phosphate is produced in situ from phosphoric acid and a water-soluble basic tungstate, preferably from phosphoric acid and sodium tungstate, wherein the phosphoric acid and sodium tungstate are more preferably used in a molar ratio from 1:2 to 10:1, or even more preferably from 3:1 to 8:

1.

5. The method of any of the preceding claims, wherein R1 is CH3, and R2, R3 and R4 are each the same, and preferably the same C12-C18 alkyl, or wherein each of R1, R2, R3 and R4 is the same C8-C18 alkyl.

6. The method of any of the preceding claims, wherein the tetraalkylammonium cation comprises at least one of tetraoctylammonium, tetradecylammonium, tetra-dodecylammonium, tri-dodecylmethylammonium, or a mixture thereof.

7. The method of any of the preceding claims, wherein the at least one tetraalkylammonium cation is provided in the form of a quaternary tetraalkylammonium salt having an anion, different from the tungsten polyoxometalate, preferably in the form of a tetraalkylammonium sulfate or a tetraalkylammonium methyl sulfate.

8. The method of claim 7, wherein the molar ratio of the tetraalkylammonium cation to tungsten is preferably in the range of 1:1 to 3:

1.

9. The method of any of the preceding claims, wherein the reaction in step a) is carried out in the presence of at least one solvent containing an aromatic hydrocarbon, preferably an alkylated aromatic hydrocarbon having 7 to 12 carbon atoms.

10. The method of any of the preceding claims, wherein the pH of the aqueous phase in step a) is maintained in the range of 1.0 to 3.5, preferably in the range of 2.0 to 3.

0.

11. The method of any of the preceding claims, wherein the reaction mixture of step a) does not contain any compound having monoester functionality.

12. The method of any of the preceding claims, wherein the method is a continuous method and step a) is preferably carried out in a loop reactor.

13. The method of any of the preceding claims, wherein the tungsten polyoxometalate present in the organic phase P2 separated in step b) is recycled to the reaction in step a).

14. The method of any of the preceding claims, wherein step b) is carried out in a setter vessel, wherein the two-phase mixture is preferably passed through a coalescer element comprising structured or random filler.

15. The method of any of the preceding claims, wherein the aqueous phase P1 is separated in step d) by nanofiltration into a permeate rich in polyoxometalates and a permeate poor in polyoxometalates, the permeate being recycled to the reaction in step a), and 1,2-alkyldiol being separated from the permeate.

Images