WO1997044302A1

WO1997044302A1 - Process for preparing a c2-6-alkanol containing from two to four hydroxyl groups

Info

Publication number: WO1997044302A1
Application number: PCT/US1997/007920
Authority: WO
Inventors: Celio Lume-Pereira; Meinolf Maria Weidenbach; Gene W. Bachman; Michael T. Holbrook; Mark E. Jones
Original assignee: The Dow Chemical Company
Priority date: 1996-05-24
Filing date: 1997-05-12
Publication date: 1997-11-27
Also published as: AU3121097A

Abstract

A process for preparing a C2-6-alkanol containing from two to four hydroxyl groups which comprises the steps of a) hydrogenating an olefinically unsaturated compound A containing from two to four halogens to produce a corresponding saturated compound B containing from two to four halogens; and b) contacting said saturated compound B with a base to produce said corresponding C2-6-alkanol containing from two to four hydroxyl groups.

Description

PROCESS FOR PREPARING A C₂₆-ALKANOL CONTAINING FROM TWO TO FOUR HYDROXYL GROUPS

The present invention relates to a process for preparing a C₂₆-alkanol containing from two to four hydroxyl groups.

There is a high interest in producing C_j.₆-alkanols which contain from two to four hydroxyl groups, preferably C₂₆-alkane diols, particularly 1 ,3-propane diol or 1 ,2-propane diol. 1 ,2-propane diol is, for example, useful for preparing unsaturated polyester resins. 1 ,3-propane diol is useful for producing polyester fibers, films and coatings and for producing polyurethanes. As reported in the article "Fine and Intermediate Chemicals Makers

Emphasize New Products and Processes," by Stephen C. Stinson, in NEWS FOCUS, July 17, 1995 C&EN (Chemical and Engineering News), pages 10 and 11 , Shell Chemical announced commercialization of a new polyester-polytrimethylene terephthalate that is based on 1 ,3-propane diol. In the same article it is reported that Degussa has installed production capacity for producing 1 ,3-propane diol of 2.2 million pounds per year. In this production plant 1 ,3-propane diol is produced by hydration of acrolein to 3-hydroxypropionaldehyde, which yields 1 ,3-propane diol on hydrogenation. The preparation of 1 ,3-propane diol from acrolein is described in more detail in European Patent Application EP-A 0 487 903. U.S. Patent Nos. 5,463,144; 5,463,145; and 5,463,146 disclose the preparation of 1,3-propane diol by reacting ethylene oxide with carbon monoxide to produce 3-hydroxypropanal and hydrogenating 3-hydroxypropanal to 1 ,3-propane diol. In view of the above-mentioned variety of uses for C₂₆-alkanols, such as 1 ,3-propane diol, it would still be desirable to provide a new process for preparing

C_{2 β}-alkanols containing from two to four hydroxyl groups, preferably

C_j.₆-alkane diols, most preferably 1 ,3-propane diol. It would be particularly desirable to provide a new process wherein these products can be efficiently prepared in large quantities.

The present invention relates to a process for preparing a

C. _β-alkanol containing from two to four hydroxyl groups which comprises the steps of: a) hydrogenating an olefinically unsaturated compound A containing from two to four halogens to produce a corresponding saturated compound B containing from two to four halogens; and

b) contacting the saturated compound B with a base to produce the corresponding C₂₆-alkanol containing from two to four hydroxyl groups.

Useful starting materials in the process of the present invention are olefinically unsaturated compounds A which contain from two to four, preferably two, halogens. The olefinically unsaturated compounds A preferably contain one or two, more preferably one carbon-carbon double bond. Of the halogens, bromine and chlorine are preferred, the latter being more preferred.

Preferred compounds A are C_{2 β}-dichloroalkenes, such as cis- or trans- dichloroethene or a mixture thereof, cis- or trans-1 ,2-dichloropropene or a mixture thereof, 2,3-dichloro-1-propene, cis- or trans-1 ,2-dichlorobutene or a mixture thereof, cis- or trans- 1 ,3-dichlorobutene or a mixture thereof, 2,3-dichlorobutene, cis- or trans-1 ,4-dichloro-2- butene, cis- or trans-1 ,2-dichloro-2-butene, the various isomers of the dichloropentenes, dichlorohexenes or mixtures thereof.

When a mixture of compounds A is used as a starting material, preferably a mixture of the corresponding trans- and cis-isomers is used which yields the same C₂₆- dichloroalkane upon hydrogenation. Other mixtures of compounds A are less preferred because additional separation steps are necessary to obtain a pure product. The following description refers to the hydrogenation of "an olefinically unsaturated compound A" although hydrogenation of mixtures is not excluded. Unless stated otherwise, the term "a compound A" includes the trans- and cis-isomer of compound A as well as mixtures of the trans- and cis-isomer.

C₂.₆-dichloroalkenes are known compounds and can be produced in a known manner. For example, they can be produced by chlorinating a C₄.₆-alkadiene, such as butadiene or hexadiene, to the corresponding C₄₆-dichloroalkene, such as 1 ,4-dichloro-2- butene or 1 ,2-dichloro-2-butene or the various isomers of dichlorohexene.

The most preferred compound A is 1 ,3-dichloropropene, specifically cis- or trans-1 ,3-dichloropropene or a mixture thereof. Another preferred compound A is 2,3- dichloropropene. Worldwide, more than 60,000 metric tons of 1 ,3-dichloropropenes and 2,3-dichloropropene are generated yearly as by-products in processes for producing allyl chloride by thermal chlorination of propene. These by-products have little commercial value and their conversion to compounds of higher value is highly desirable.

Compound A can be reacted according to the process of the present invention in the presence of impurities, such as hereinafter described. However, the impurities or their derivative compounds formed in the process should generally be separated from the product at some stage of the process. It is desirable to remove such impurities from compound A prior to the process of the present invention by one or more appropriate operations such as distillation, rectification, stripping, washing, extraction, absorption or adsorption. Such impurities are, for example, one or more of the following:

saturated linear or cyclic hydrocarbons without any other chemical functionality in the molecule;

unsaturated linear or cyclic hydrocarbons without any other chemical functionality in the molecule;

compounds containing triple bonds without any other chemical functionality in the molecule;

aromatic or heteroaromatic compounds which contain oxygen, nitrogen, sulfur, phosphorus or any other non-metal, metalloid or metal as heteroatom(s);

linear or cyclic ethers, amines, amides or compounds containing sulfur, phosphorus, other non-metals, metalloids or metals.

However, if compound A is mixed with compounds such as those mentioned above, their total weight should generally be not more than 30 percent, preferably not more than 10 percent, more preferably not more than 5 percent, most preferably not more than 1 percent, based on the total weight of the starting material used in step a) of the process.

In a preferred embodiment of the process of the present invention, an olefinically unsaturated compound A which contains from two to four halogens is mixed with the corresponding saturated compound B which contains from two to four halogens. In this embodiment of the process of the present invention the saturated compound B serves as a reaction diluent for controlling the reaction temperature, as will be described further below in more detail. The total amount of compound A and compound B generally is from 70 to 100 percent, preferably from 90 to 100 percent, more preferably from 95 to 100 percent, most preferably from 99 to 100 percent, based on the total weight of the starting material used in step a). The amount of compound A is preferably from 1 to 95 percent, more preferably from 5 to 75 percent, most preferably from 10 to 50 percent, based on the total weight of the starting material used in step a).

As indicated above, the most preferred compound A is cis- or trans-1 ,3-dichloropropene or a mixture thereof which is produced as a by-product in the production of allyl chloride. Besides 1 ,3-dichloropropene, 1 ,2-dichloropropane or other halogenated hydrocarbons are often present in the same by-product stream. After the hydrogenation step a) these by-products are generally separated from the produced 1 ,3- dichloropropane, preferably by distillation.

The hydrogenation step a) is preferably carried out by means of pure hydrogen or a hydrogen-rich gaseous stream which comprises minor amounts of other components, such as C_{) 6}-alkanes or C_2.6-alkenes or hydrogen halide. Preferably, the weight of hydrogen in the hydrogen-rich gaseous stream amounts to at least 70 percent, preferably at least 90 percent and more preferably at least 98 percent, based on the total molar amount of the hydrogen-rich stream. The molar ratio between the hydrogen and the total number of olefinic double bonds in compound A generally is at least 1 :1 , preferably from 1:1 to 100:1 , more preferably from 1 :1 to 20:1 , most preferably from 1 :1 to 10:1.

The hydrogenation step a) is generally conducted in the presence of a catalyst. Suitable hydrogenation catalysts are known in the art. Generally, they are based on one or more metals of Group Vlll of the Periodic Table of Elements, such as iron, cobalt, nickel, ruthenium, rhodium, palladium, osmium, iridium or platinum and are supported by carriers, such as activated carbon, silica, titanium dioxide, zirconium dioxide, or, preferably, alpha-, beta- or gamma-alumina. The hydrogenation catalyst may further contain one or more of the following components: alkali metals, alkaline earth metals, elements of group Ilia, IVa and Va and sulfur components. More detailed information about hydrogenation catalysts are, for example, published in Chemical Abstract No. 116:255051 which relates to studies on hydrogenation of carbon-carbon double bonds with nickel catalysts; in Chemical Abstract No. 88:169335 which relates to studies on hydrogenation of olefins catalyzed by polymer-bound palladium (II) complexes; in Chemical Abstract No. 85:176349 which discloses an active palladium catalyst for the hydrogenation of organic compounds, such as 1 -hexene or cyclohexene, and in GB-A-1 ,008,175 which discloses the use of a rhodium catalyst for hydrogenation of unsaturated halogenated hydrocarbons. The hydrogenation step a) is preferably kept at mild conditions which allows selective hydrogenation of the olefinic double bond(s) in compound A and at which the formation of undesired by-products by hydrodehalogenation or by polymerization is minimized. These conditions are described in more detail below. The described reaction conditions are the preferred conditions when the hydrogenation reaction has reached a steady state.

The hydrogenation is generally conducted at a pressure of from 1 to 100 bar, preferably from 2 to 70 bar, more preferably from 5 to 60 bar, most preferably from 10 to 50 bar. The hydrogenation is preferably conducted at a temperature of from 15°C to 150°C, more preferably from 20°C to 100°C, most preferably from 20°C to 60°C. Under these conditions compound A and the above-mentioned other compounds with which compound A is optionally mixed are usually liquid. It may be desirable to heat the reactor(s) used for the hydrogenation or to preheat the starting material at the beginning of the hydrogenation reaction if an initial elevated temperature is desired. However, when the reaction reaches its steady state, heating is stopped. Preferred liquid residence times on the catalyst are from 0.5 to 60 minutes, more preferably from 1 to 25 minutes, most preferably from 2 to 15 minutes. The reaction can be carried out in batches or, preferably, continuously in one or more reactors. The hydrogenation reactor, for example, contains a fluidized bed, or, preferably, a fixed bed or trickle catalyst bed. The reactor(s) may contain fillings which are inert under the conditions chosen in the reactor(s). Useful inert fillings are, for example, glass or ceramic beads.

The hydrogenation step a) is an exothermic reaction which requires careful temperature control when the reaction is carried out a on large scale. The reaction temperature can be controlled by diluting compound A with a non-polar organic solvent, such as carbon tetrachloride, pentane, hexane, heptane, cyclohexane or octane. However, this temperature control is not very convenient because the organic solvent has to be separated from the product in an additional step after the reaction.

A preferred way of controlling the reaction temperature is the use of a jacketed reactor or a heat exchanger for cooling the reactor. A preferred reactor type is, for example, a tube-and-shell reactor wherein the catalyst is located in the tubes and a heat-transfer fluid, such as water, water containing suitable additives, brine or any appropriate commercially available heat-transfer fluid. According to another preferred alternative, compound A is hydrogenated in step a) in the presence of a corresponding saturated compound B. This means that preferably the same saturated compound B is used as reaction diluent as the one that is produced in step a). Preferably at least a portion of the compound B that is produced in the hydrogenation step a) is recycled to the hydrogenation reaction. The recycled compound B serves as a reaction diluent which allows a continuous, efficient and inexpensive control of the reaction temperature. The reaction temperature can be controlled by controlling and regulating the amounts of unsaturated compound A and saturated compound B that are fed into the hydrogenation reactor. The weight ratio between compound B and compound A in the hydrogenation reaction is preferably from 1 :1 to 100:1 , more preferably from 2:1 to 50:1 , most preferably from 5:1 to 40:1. At the initial phase of the reaction, compound B is preferably provided from another source as long as not a sufficient amount of compound B is produced in the hydrogenation reaction. When using compound B as a reaction diluent, preferred reactor types are a heat-insulated drum without a jacket, an insulated tube or an insulated vessel which is internally equipped with multiple tubes. The yield of compound B in the reaction step a) generally is at least 90 percent, typically at least 95 percent, in most cases at least 99 percent of the theoretically achievable amount.

After completion of the hydrogenation step a), excess hydrogen, if present, is preferably separated from the liquid comprising compound B in a gas/liquid separation device. The temperature during the gas/liquid separation preferably is from -20°C to +100°C, more preferably from -10°C to +50°C, most preferably from +5°C to +30°C. The pressure generally is from 1 to 100 bar, preferably from 1 to 50 bar, more preferably from 1 to 10 bar. It may be advantageous to recycle excess hydrogen to the hydrogenation step a). If the reaction temperature in the hydrogenation step a) is controlled by diluting an unsaturated compound A with the corresponding saturated compound B, as described above, preferably from 20 to 99 percent, more preferably from 25 to 90 percent of the total volume of the liquid product comprising compound B is recycled to the hydrogenation step a). In addition to compound B, the liquid product obtained in step a) may contain minor amounts of by¬ products, such as hydrogen chloride or by-products which originate from the starting material used for the hydrogenation step a). The amount of hydrogen chloride is usually less than 2 percent, in most cases less than 1 percent, based on the weight of compound B. Hydrogen chloride, if present, is preferably not separated from compound B prior to process step b). Other by-products are preferably separated from compound B by distillation prior to process step b). In step b) of the process of the present invention a produced saturated compound B which contains from two to four halogens, preferably a C_{2 6}-dichloroalkane, more preferably 1 ,2-dichloropropane, most preferably 1 ,3-dichloropropane, is contacted with a base to produce the corresponding C₂₆-alkanol containing from two to four, preferably two, hydroxyl groups. Preferred products are 1 ,2-propane diol or, more preferably, 1 ,3-propane diol.

Step b) preferably comprises the steps of:

b1 ) contacting the saturated compound B with a carboxylic acid salt to produce an corresponding ester; and

b2) contacting the produced ester with an alkali metal hydroxide or an alkaline earth metal hydroxide.

In European Patent 0 227 997, step b1 ) is generally described. However, the European patent does not describe the most preferred embodiment of step b1 ) wherein an organic liquid reaction medium is used and the reaction medium is the same C₂₆-alkanol as the one produced in the process of the present invention. Step b1 ) will be described in more detail below.

Preferred carboxylic acid salts are salts of a carboxylic acid which contains from 1 to 6 carbon atoms, more preferably from 2 to 3 carbon atoms. The carboxylic acid is preferably saturated. Monocarboxylic acids are preferred over dicarboxylic acids. The salts of butanoic acid, pentanoic acid or hexanoic acid may be used. The salts of propanoic acid and, more preferably, of acetic acid are preferred. Preferred salts are the alkaline earth metal salts, such as the calcium salt, or the alkali metal salts, preferably the sodium or potassium salts. Most preferably, the sodium salt of acetic acid is used in step b1 ) of the process of the present invention.

The saturated compound B and the carboxylic acid salt are preferably contacted in an organic liquid under conditions to achieve esterification of the saturated compound B. Preferred is an organic liquid single phase reaction medium, that means an organic liquid that is capable of forming a single liquid phase with the saturated compound B and the carboxylic acid salt and in which the inorganic salt formed by the reaction of the saturated compound B with the carboxylic acid salt is insoluble. By organic "liquid" reaction medium is meant an organic reaction medium that is liquid under reaction conditions. As used herein regarding the inorganic salt and the organic liquid, the term "insoluble" means that generally less than 2 weight percent, preferably less than 1 weight percent, and more preferably less than 0.5 weight percent of the inorganic salt will dissolve in the organic liquid. Most preferably, the same C₂₆-alkanol containing from two to four hydroxyl groups is used as an organic reaction medium as the one produced in the process of the present invention. That means, when producing 1 ,3-propane diol in the process of the present invention, preferably 1 ,3-propane diol is used as the reaction medium in step b1); and when producing 1 ,2-propane diol in the process of the present invention, preferably 1 ,2-propane diol is used as the reaction medium in step b1). By using this preferred organic reaction medium, subsequent separation of the reaction product from the reaction medium can be avoided. Alternatively, the ester produced in step b1) can be used as a liquid reaction medium. As a further alternative, the carboxylic acid, in acid form, can be used as an organic reaction medium which corresponds to the carboxylic acid salt employed in the process step b1 ). Other inert organic liquid single phase reaction media may also be used. However, their use requires additional separation steps and is less preferred. The organic liquid reaction medium is employed in an amount sufficient to dissolve the quantity of the carboxylic acid salt and the saturated compound B which is required to start the reaction. The amount of the carboxylic acid salt is generally from 0.1 to 40 percent, preferably from 1 to 25 percent, more preferably from 5 to 20 percent and most preferably from 10 to 15 percent, based on the total weight of the carboxylic acid salt and the solvent. The ratio of the weight of the saturated compound B to the weight of the organic liquid reaction medium is generally not higher than 1 :1 , preferably not higher than 0.3:1 , more preferably not higher than 0.15:1.

In general, the carboxylic acid salt and the saturated compound B are employed in amounts such that the reaction mixture contains from 0.1 to 3.0, preferably from 0.25 to 2.5, more preferably from 0.8 to 2.0 equivalents of the carboxylic acid salt per reactive halogen atom of the saturated compound B. In the most preferred embodiment of the process of the present invention, 1 ,3-dichloropropane is used as compound B, sodium acetate as the carboxylic acid salt and 1 ,3-propane diol as organic reaction medium.

The presence of low amounts of water in the reaction mixture has also been found to significantly affect the rate and selectivity of reaction. Specifically, water in th? reaction mixture, although increasing the rate of reaction, will more significantly decre 3 the selectivity of the reaction. Most preferably, the amount of water in the reaction mixture is minimized, with no measurable amounts of water being most preferred. However, in commercial operation, it is often impractical and/or impossible to eliminate water from the reaction mixture, and up to two weight percent of water based on the total weight of the reaction mixture can be tolerated while still achieving desirable selectivity. More preferably, the reaction mixture contains less than 0.5, most preferably less than 0.3 weight percent of water.

In general, the reaction is carried out in a temperature range from 160°C to 250°C to give a reasonable rate of reaction. Most preferably, temperatures from 180°C to 220°C are employed to reduce the formation of by-products such as unsaturated, monohalogenated hydrocarbons, which occurs increasingly at elevated temperatures. The reaction is conducted for a period sufficient to obtain the desired conversion. In general, reaction times will vary from 0.1 to 12 hours, preferably from 0.5 to 8 hours.

In general, the reaction is conducted in a closed reactor under a pressure greater than or equal to the vapor pressure of the contents. In conducting the reaction, the saturated compound B, the carboxylic acid salt and the organic liquid are advantageously mixed, preferably continuously mixed during reaction. A good mixing of the components is preferred, since it has been found that better mixing will reduce the time required for the desired reaction.

In step b1 ) an ester containing from two to four, preferably two ester groups, is produced. In addition to the ester, an inorganic halide is produced in step b1), which is preferably separated from the reaction mixture by solid-liquid separation, such as centrifugation or filtration. The yield of the ester compound in step b1) generally is at least 80 percent, and typically at least 90 percent of the theoretically achievable amount.

In step b2) the produced ester is preferably contacted with an alkaline earth metal hydroxide or, more preferably, an alkali metal hydroxide to hydrolyze the ester groups to the respective alcohol groups. The hydroxide preferably contains the same alkaline earth metal ion or alkali metal ion as the carboxylic acid salt used in step b1). Potassium hydroxide and, more preferably, sodium hydroxide, are preferred. It has been found that the speed of the hydrolysis is increased if the hydroxide is employed as an aqueous solution. Preferably, the aqueous solution contains at least 5 weight percent of the hydroxide, based on the total weight of the solution. The upper amount of the hydroxide is generally limited by its solubility in water. More preferably, the aqueous solution contains from 20 to 70 weight percent of the hydroxide. Generally the hydroxide and the ester are used in such amounts that the reaction mixture contains from 1 to 2.5, preferably from 1 to 2, more preferably from 1 to 1.5 equivalents of hydroxide per ester group of the compound produced in step b1).

The ester, hydroxide and water are preferably mixed and subjected to an elevated temperature and pressure in a closed reactor. The initial reaction temperature preferably is from 25°C to 50°C, more preferably from 30°C to 40°C. It has been found that the reaction is exothermic and should generally be cooled to keep the reaction temperature below 180°C, preferably below 160°C, more preferably below 140°C. The pressure is generally higher than the vapor pressure of the reaction mixture. Preferably, the pressure is from 0.1 to 10 bar, more preferably from 1 to 5 bar, most preferably from 3.5 to 4.5 bar. The reaction time generally is from 0.1 to 5 hours, preferably from 0.1 to 0.5 hours. The yield of the alcohol in the reaction step b2) generally is at least 95 percent, typically at least 99 percent and in most cases even at least 99.9 percent of the theoretically achievable amount. During the hydrolysis of the ester groups, the alkali or alkaline earth metal salt of the corresponding carboxylic acid is produced as a by-product. Preferably at least a portion thereof is recycled to the process step b1 ).

The conversion of the saturated compound B to the corresponding C₂₆-alkanol containing from two to four hydroxyl groups via the esterification step b1) and the subsequent hydrolysis b2) is very advantageous because essentially no side reactions like dehydrohalogenation take place. The selectivity of the reaction to the desired C₂₆-alkanol is generally at least 80 percent, typically at least 85 percent and in most cases even at least 95 percent.

Alternatively to the above described steps b1 ) and b2), the saturated compound B produced in above described step a) can be directly hydrolyzed with an alkaline earth metal hydroxide or, preferably, an alkali metal hydroxide or with a weak base to the corresponding C₂._β-alkanol containing from two to four hydroxyl groups. The preferred saturated compound B is 1 ,2-dichloropropane or, more preferably, 1 ,3-dichloropropane.

A preferred alkaline earth metal hydroxide is calcium hydroxide. Preferred alkali metal hydroxides are potassium and sodium hydroxide. The hydroxide is preferably used in the form of an aqueous solution which contains at least 1 weight percent of the hydroxide, based on the total weight of the solution. The upper amount of the hydroxide is generally limited by its solubility in water. More preferably, the aqueous solution contains from 5 to 50 weight percent of a hydroxide, most preferably sodium hydroxide. Generally the hydroxide and the saturated compound B are used in such amounts that the reaction mixture contains from 0.1 to 3.0, preferably from 0.25 to 2.5, more preferably from 0.8 to 2.0 equivalents of hydroxide per halogen in the halogenated saturated compound B.

Instead of an alkali metal hydroxide or an alkaline earth metal hydroxide, a weak base such as an alkali metal carbonate or bicarbonate or an alkaline earth metal carbonate or bicarbonate can be used for hydrolysis of the halogenated saturated compound B. In this case the alkali metal carbonate or alkali metal bicarbonate are more preferred. The carbonate or bicarbonate is preferably used in the form of an aqueous solution which contains at least 1 weight percent of the carbonate or bicarbonate, based on the total weight of the solution. The upper amount of the carbonate or bicarbonate is generally limited by its solubility in water. More preferably, the aqueous solution contains from 5 to 50 weight percent of a carbonate or bicarbonate, most preferably sodium carbonate or sodium bicarbonate. Generally the carbonate or bicarbonate and the saturated compound B are used in such amounts that the reaction mixture contains from 0.1 to 3.0, preferably from 0.25 to 2.5, more preferably from 0.8 to 2.0 equivalents of the carbonate or bicarbonate per halogen in the saturated compound B.

The halogenated saturated compound B, the base and water are preferably mixed and subjected to an elevated temperature at atmospheric pressure in a reactor equipped with a reflux condenser or at a higher than atmospheric pressure in a closed reactor. The reaction temperature generally is from 40°C to 220°C, preferably from 50°C to 150°C. Preferably, the pressure is from 1 to 100 bar, more preferably from 1 to 70 bar. The reaction time generally is from 0.25 to 50 hours, preferably from 0.5 to 8 hours.

The yield of the alcohol in the reaction step b) generally is at least 30 percent, typically at least 50 percent and in most cases at least 70 percent of the theoretically achievable amount.

The present invention is further illustrated by the following examples which should not be construed to limit the scope of the present invention. Unless stated otherwise, all parts and percentages are given by weight.

Example 1

For performing hydrogenation of 1 ,3-dichloropropene, a vertically installed tube made of stainless steel was used as a hydrogenation reactor. The tube contains 100 g of a conventional supported group Vlll hydrogenation catalyst. The remaining volume of the tube was filled with glass beads. The reactor was operated in the trickle-bed mode. The pressure in the hydrogenation reactor was controlled by means of a pressure control valve. The reactor effluent was cooled in a jacketed cooler, wherein the coolant had a temperature between 15°C and 20°C. The effluent from the cooler was fed into a gas-liquid separator. Excess hydrogen and any hydrogen chloride and alkane or chlorinated alkane formed in the reaction as by-products left the gas-liquid separator via a gas outlet at the top. The pressure in the gas-liquid separator was controlled by means of a pressure-control valve which was connected to the gas outlet of the gas-liquid separator. The liquid product stream was collected at the bottom of the gas-liquid separator. A portion of the liquid product stream was recycled to the hydrogenation reactor and the remainder was fed into a product vessel.

Before starting the hydrogenation process for the first time, liquid

1 ,2-dichloropropane was fed into the hydrogenation reactor at a rate of 300 g/h. In subsequent start-up operations, 1 ,3-dichloropropane kept aside from previous operations was fed into the reactor instead of 1 ,2-dichloropropane. Feeding was continued until a sufficient liquid level was achieved in the gas-liquid separator. Thereafter, the liquid was recycled from the gas-liquid separator to the hydrogenation reactor at a rate of 600 g/h. Direct feeding of 1 ,2- or 1 ,3-dichloropropane into the hydrogenation reactor was stopped. Now hydrogen was fed into the reactor at a rate of 25 Nl/h. The pressure in the hydrogenation reactor was adjusted to 30 bar. The temperature in the reactor was 20°C. A liquid feed consisting of 50 percent of cis-1 ,3-dichloropropene and 50 percent of trans-1 ,3-dichloropropene was introduced into the reactor at a rate of 20 g/h. The liquid feed was continued until the temperature in the reactor had risen to 50°C and the liquid feed rate was adjusted such that the temperature in the reactor was maintained at this level. The liquid residence time on the catalyst bed defined as the quotient of the empty-tube volume occupied by the catalyst bed and the volumetric flow rate of the liquid over the catalyst bed was 20 minutes. As the hydrogenation reaction proceeds, the liquid level in the gas-liquid separator increased, but it was adjusted by means of a level-control valve which opened regularly to allow liquid to flow to the product vessel. The pressure in the gas-liquid separator was kept at 2 bar.

Analysis of the liquid product by gas chromatography showed that the product consisted of 1 ,3- and 1 ,2-dichloropropane when the reactor was started for the first time. After stationary conditions had been achieved, the product only consisted of 1 ,3-dichloropropane. If the process was started with 1 ,3-dichloropropane from previous operations, the product analysis showed that it only consists of 1 ,3-dichloropropane.

113 g of the 1 ,3-dichloropropane collected in the product vessel above were introduced in an autoclave of 750 mL volume, equipped with a mechanical stirrer and a jacket connected to a thermostat filled with a Dowtherm™ J (Trademark of The Dow Chemical Company) thermal fluid. Then 165 g of sodium acetate, 300 g of 1 ,3-propane diol, 30 g of acetic acid and 1 g of water were added. The autoclave was closed, the stirrer set into operation and the contents of the autoclave heated to 180°C. After 20 hours at this temperature, the autoclave contents were cooled to ambient temperature, the stirring stopped and the autoclave contents analyzed. The autoclave was found to contain 14 g of unreacted sodium acetate, 6 g of unreacted 1 ,3-dichloropropane, 7 g of 3- chloropropylacetate, 143 g of 1 ,3-propyleneglycol diacetate, 108 g of sodium chloride, 30 g of acetic acid and 300 g of 1 ,3-propane diol. The amount of water and some trace components were not included. This mixture was filtered to separate the solid sodium chloride. The filtrate was transferred to a 1000 mL round-bottom glass flask equipped with a dropping funnel, a stirrer and a reflux condenser. In the dropping funnel 240 g of a 30 percent aqueous solution of sodium hydroxide was placed. Under stirring, this sodium hydroxide solution was given into the round-bottom flask at ambient temperature. Thereafter, the contents of the flask were heated under reflux for about 30 minutes. The subsequent analysis of the cooled solution was found to contain 367.5 g of 1 ,3-propane diol, 7 g of 3-chloropropyl acetate, 160 g of sodium acetate and 34.5 g of acetic acid. The amount of water was not included. The amount of 1 ,3-propane diol obtained corresponds to a yield of 89 percent of theory.

Example 2

1 ,2-dichloropropane was produced by hydrogenation of 2,3-dichloropropene according to the process described in Example 1. 1 ,2-dichloropropane was hydrolyzed to 1 ,2-propane diol as follows:

196 g of water and 274 g of sodium bicarbonate were placed in a 750 ml autoclave equipped with a magnetic stirrer and a jacket connected to a thermostat filled with a thermal fluid. The stirrer was set into operation and the contents of the autoclave were heated to 200°C. Then 168 g of 1 ,2-dichloropropane were introduced into the autoclave at a rate of 2 g/minute by means of a positive displacement pump. A pressure rise to 50 bar was observed, indicating that gaseous C0₂ was formed. After 84 minutes, when feeding of 1 ,2-dichloropropane was completed, the reactor was cooled to ambient temperature, the stirring stopped and the autoclave contents analyzed. The autoclave was found to contain 22 g of 1 ,2-dichloropropane, 5 g of allyl alcohol, 91 g of 1,2-propanediol, 198 g of water, 57 g of sodium bicarbonate and 151 g of sodium chloride. The amount of C0₂ as gaseous product was found to be 57 g. From these numbers a 1 ,2-dichloropropane conversion of 86.9 percent and a 1 ,2-propanediol selectivity of 92.7 percent were calculated.

Claims

CLAIMS:

1. A process for preparing a C₂₆-alkanol containing from two to four hydroxyl groups comprising the steps of

a) hydrogenating an olefinically unsaturated compound A containing from two to four halogens to produce a corresponding saturated compound B containing from two to four halogens; and

b) contacting said saturated compound B with a base to produce said corresponding C₂₆-alkanol containing from two to four hydroxyl groups.

2. The process of Claim 1 wherein a C₂₆-alkane diol is produced from a C, -dichloroalkene.

3. The process of Claim 2 wherein 1 ,3-propane diol is produced from cis- or trans-1 ,3-dichloropropene or a mixture thereof.

4. The process of Claim 2 wherein 1 ,2-propane diol is produced from 2,3-dichloropropene.

5. The process of anyone of Claims 1 to 4 wherein a C₄₆-alkane diol is produced from a C₄₆-dichloroalkene which has been produced by chlorination of a C, -alkadiene.

6. The process of anyone of Claims 1 to 5 wherein in step a) said saturated compound B is used as a reaction diluent.

7. The process of anyone of Claims 1 to 6 wherein in step b) said saturated compound B is directly hydrolyzed with an alkali metal hydroxide, an alkaline earth metal hydroxide or a weak base to the corresponding C₂₆-alkanol containing from two to four hydroxyl groups.

8. The process of anyone of Claims 1 to 6 wherein step b) comprises the steps of

b1) contacting said saturated compound B with a carboxylic acid salt to produce an ester and b2) contacting the produced ester with an alkali metal hydroxide or an alkaline earth metal hydroxide.

9. The process of Claim 8 wherein step b1 ) is carried out in an organic liquid single phase reaction medium.

10. The process of Claim 9 wherein said organic liquid reaction medium is the same C₂₆-alkanol containing from two to four hydroxyl groups as the one produced in the process.