Method for hydrogenating carboxylic acids in order to form alcohols

Abstract

Process for the continuous hydrogenation of a carboxylic acid (I) to an alcohol (II) by means of hydrogen at a temperature of from 100 to 300 C. and a pressure of 10 to 33 MPa abs in a reactor tube through which axial flow occurs and which has a fixed-bed catalyst which is fixed therein and comprises at least one element from the group consisting of Re, Co and Cu, and in which the carboxylic acid (I) to be hydrogenated is present in a liquid mixture (III) comprising the carboxylic acid (I), water and alcohol (II), where the mixture (III) has an acid number of from 0.2 to 25 mg KOH/g and comprises at least 15% by weight of water and at least 20% by weight of alcohol (II) and the flow velocity of the flowing liquid calculated on the basis of the geometric cross-sectional area of the empty, catalyst-free reactor tube is from 10 to 50 m/h.

Claims

1. A process for the continuous hydrogenation of a carboxylic acid of the general formula (I)
Y.sup.1X.sup.1COOH(I) where X.sup.1 is a (CH.sub.2).sub.n group with n from 1 to 10 or a CHCH group and Y.sup.1 is H-, HOOC or HOCH.sub.2, or a mixture thereof, with retention of the number of carbon atoms to give an alcohol of the general formula (II)
Y.sup.2X.sup.2CH.sub.2OH(II) where X.sup.2 is a (CH.sub.2).sub.n- group with n from 1 to 10 and Y.sup.2 is H or HOCH.sub.2, by means of hydrogen at a temperature of from 100 to 300 C. and a pressure of from 10 to 33 MPa abs in a reactor tube through which axial flow occurs and which has a fixed-bed catalyst which is fixed therein and comprises at least one element from the group consisting of Re, Co and Cu, wherein the carboxylic acid (I) to be hydrogenated is present in a liquid mixture (III) comprising the carboxylic acid (I), water and alcohol (II), where the mixture (III) a) has an acid number of from 0.2 to 25 mg KOH/g, b) comprises at least 15% by weight of water, c) comprises at least 20% by weight of alcohol (II) and d) the flow velocity of the flowing liquid calculated on the basis of the geometric cross-sectional area of the empty, catalyst-free reactor tube is from 10 to 50 m/h.

2. The process according to claim 1, wherein the mixture (III) has an acid number of from 0.5 to 10 mg KOH/g.

3. The process according to claim 1, wherein the sum of the contents of carboxylic acid (I), water and alcohol (II) in the mixture (III) is 100% by weight.

4. The process according to claim 3, wherein the sum of the contents of carboxylic acid (I), water and alcohol (II) in the mixture (III) is from 40 to 100% by weight.

5. The process according to claim 1, wherein the flow velocity of the flowing liquid calculated on the basis of the geometric cross-sectional area of the empty, catalyst-free reactor tube is 20-40 m/h.

6. The process according to claim 1, wherein the process is carried out without recirculation of hydrogenated mixture.

7. The process according to claim 1, wherein the mixture (III) comprises from 10 to 1000 ppm by weight of alkali metal selected from the group consisting of Na and K.

8. The process according to claim 1, wherein succinic acid, 4-hydroxybutyric acid, maleic acid, glutaric acid, 5-hydroxyvaleric acid, adipic acid, 6-hydroxycaproic acid or a mixture thereof is used as carboxylic acid (I).

9. The process according to claim 1, wherein a supported catalyst comprising from 0.1 to 10% by weight of Re on a support selected from the group consisting of graphite, activated carbon, ZrO.sub.2, Al.sub.2O.sub.3, SiO.sub.2 and TiO.sub.2, a precipitated catalyst comprising from 1 to 90% by weight of Co, a precipitated catalyst comprising from 0.5 to 60% by weight of Cu or a precipitated catalyst comprising from 15 to 85% by weight of Co and from 5 to 20% by weight of Cu, where the sum of the contents of Co and Cu does not exceed 100% by weight, is used as fixed-bed catalyst.

10. The process according to claim 1, wherein the mixture (III) is obtained by continuous prehydrogenation of a solution comprising carboxylic acid (I) and water, with the solution having an acid number of from 50 to 900 mg KOH/g.

11. The process according to claim 10, wherein the prehydrogenation is carried out with recirculation and from 50 to 98% by weight of the prehydrogenated solution is recirculated to the prehydrogenation.

Description

EXAMPLES

Examples 1 to 4 (Hydrogenation of Succinic Acid to 1,4-Butanediol)

(1) Production of an Re/Pd Catalyst

(2) (5% of Re, 5% of Pd on Oxidized Activated Carbon)

(3) 500 g of activated carbon having a particle size of 3070 mesh were in each case admixed with an excess (supernatant solution) of concentrated nitric acid (69-71% strength HNO.sub.3) and stirred at 80 C. for about 18 hours. After cooling, the product was isolated by filtration, washed a number of times with an excess of water and dried at 120 C.

(4) 1.5 kg of the oxidized activated carbon produced as described above were in each case then mixed with 7.2 kg of an aqueous solution comprising 114 g of NH.sub.4ReO.sub.4 and 1.09 kg of an aqueous Pd(NO.sub.3).sub.2 solution having a Pd content of 7.26% by weight and the slurry obtained was evaporated to dryness and subsequently dried at 120 C. and tableted to give 33 mm pellets.

(5) Production of a Prehydrogenated Feed by Hydrogenation of Maleic Acid

(6) 3.00 liters of the Re/Pd catalyst pellets produced by the abovementioned method were then introduced into a 10 m long reactor tube having an internal diameter of 2 cm and firstly activated therein. For this purpose, the catalyst was heated in a stream of hydrogen at 1 C. per minute to 200 C. and then kept in the stream of nitrogen at 200 C. for 5 hours.

(7) A stream of 0.5 kg/h of a 33% strength by weight solution of maleic acid in water was then hydrogenated continuously over this catalyst at 20 MPa abs, from 150 to 170 C. and a product recirculation rate of 5 kg/h with introduction of 250 standard I/h of hydrogen and the reaction output was collected. After 4 weeks of continuous operation, the process was stopped and the collected reaction output was analyzed.

(8) The feed which had been prehydrogenated in this way had a water content of about 75% by weight. As organic components, it comprised, according to gas-chromatographic analysis, calculated on a water-free basis, about 90.7% by weight of 1,4-butanediol, about 2.4% by weight of tetrahydrofuran, about 2.5% by weight of n-butanol, about 0.5% by weight of gamma-butyrolactone and also further reaction products of maleic acid. The acid number of the prehydrogenated feed was 3.8 mg KOH/g and was predominantly attributable to succinic acid.

(9) Hydrogenation of Succinic Acid in the Prehydrogenated Feed to 1,4-Butanediol

(10) To carry out the hydrogenation of the succinic acid, 3.00 liters of the Re/Pd catalyst pellets produced by the abovementioned method were introduced into a 10 m long reactor tube having an internal diameter of 2 cm and firstly activated. For this purpose, the catalyst was heated in a stream of hydrogen at 1 C. per minute to 200 C. and then kept in the stream of hydrogen at 200 C. for 5 hours.

(11) The prehydrogenated feed was subsequently passed continuously in the downflow mode over the catalyst at a reactor inlet temperature of 170 C., 20 MPa abs together with 70 standard I/h of hydrogen per kg of prehydrogenated feed. In the present trial, the amount of prehydrogenated feed fed in was gradually increased, with sampling being carried out in each case about 24 hours after setting of an inflow amount.

(12) The results are shown in table 1. The water content was determined by the Karl-Fischer method and the content of 1,4-butanediol was determined gas-chromatographically.

(13) Examples 1 to 4 show the dependence of the acid number and also of the content of target alcohol on the flow velocity of the liquid in the tube reactor.

(14) At a low flow velocity of only 5 m/h (example 1), the carboxylic acid present is hydrogenated largely completely, which is shown by a very low acid number of <0.5 mg KOH/g, but appreciable amounts of the target alcohol 1,4-butanediol are also destroyed by hydrogenation.

(15) Thus, the content of 1,4-butanediol decreases from an original 90.7% by weight to 88.0% by weight, in each case calculated on a water-free basis.

(16) A high flow velocity of 60 m/h (example 4) does give a high calculated throughput through the reactor, but allows only partial hydrogenation of the carboxylic acid present. Thus, the output still has an acid number of 1.5 mg KOH/g, which corresponds to hydrogenation of only about 60% of the carboxylic acid present.

(17) In comparison, examples 2 and 3 with a flow velocity of 10 and 30 m/h, respectively, show both virtually complete hydrogenation of the carboxylic acid present, as indicated by an acid number in the output of <0.5 mg KOH/g, and also a significant increase in the target alcohol 1,4-butanediol, as indicated by a significantly increased content of 1,4-butanediol of 94.5 and 94.3% by weight, respectively.

(18) The hydrogenation according to the invention gave a virtually carboxylic acid-free product stream with a significant increase in target alcohol from a feed stream comprising alcohol and carboxylic acid.

Examples 5 to 8 (Hydrogenation of 6-Hydroxycaproic Acid to 1,6-Hexanediol)

(19) Production of a Co/Cu/Mn/Mo Catalyst

(20) (66% of CoO, 20% of CuO, 7.3% of Mn.sub.3O.sub.4, 3.6% of MoO.sub.3, 0.15% of Na.sub.2O, 3% of H.sub.3PO.sub.4)

(21) The Co/Cu/Mn catalyst precursor was produced by two-stage precipitation of a starting mixture composed of 38.3 kg of an aqueous cobalt nitrate solution comprising 12.6% by weight of cobalt, 6.53 kg of an aqueous copper nitrate solution comprising 15.3% by weight of copper, 2.78 kg of an aqueous manganese nitrate solution comprising 12.6% by weight of manganese and 0.199 kg of 75.3% strength by weight phosphoric acid with 20% strength by weight soda solution. Starting mixture was fed continuously in an amount corresponding to 1.5 kg of metal oxide/h into a first stirred vessel (useful capacity 6 I) at 50 C. and admixed while stirring with the amount of soda solution required to maintain a pH of 8.5 (measured using a glass electrode). The incomplete precipitation mixture is transferred in its entirety into a second vessel and then after-precipitated at a pH of from 6.5 to 7.5 (optionally with addition of further soda solution) over a period of 2 hours. The suspension obtained was filtered and the solid was washed and dried.

(22) This gave a basic carbonate having a BET surface area of about 120 m.sup.2/g. This carbonate was then decomposed to the oxide at a temperature in the range from 420 to 540 C. in a stream of air and washed free of residual alkali with deionized water. 4 kg of the washed and dried oxide were then admixed in a kneader with 652 g of an ammoniacal Mo solution which had been produced by dissolution of technical-grade molybdenum oxide hydrate in aqueous ammonia solution and had a calculated MoO.sub.3 content of 25.5% by weight and mixed by kneading. During kneading, the phosphorus discharged by means of the washing processes was replaced by fresh phosphoric acid and 285 g of a 65.3% strength by weight nitric acid and 1300 g of deionized water were introduced and the mixture was kneaded intensively for 2.5 hours. The kneaded composition was then shaped to give extrudates having a diameter of 4 mm and a length of from 3 to 9 mm, dried, and calcined at 500 C. for 6 hours. The extrudates had a bulk density of 1700 g/l.

(23) Hydrogenation of 6-Hydroxycaproic Acid to 1,6-Hexanediol

(24) 3.00 liters of the Co/Cu/Mn/Mo catalyst described were then introduced into a 10 m long reactor tube having an internal diameter of 2 cm. The feed to be hydrogenated was obtained by prehydrogenation of a mixture which comprises 17% by weight of adipic acid, 16% by weight of 6-hydroxycaproic acid, 2% by weight of glutaric acid, 1.5% by weight of 5-hydroxypentanoic acid, 1% by weight of formic acid, 1% by weight of 1,4-cyclohexanediol, 1% by weight of 1,2-cyclohexanediol and 0.3% by weight of cyclohexanol/cyclohexanone and is formed as by-product in the oxidation of cyclohexane to cyclohexanol/cyclohexanone and has been obtained by scrubbing with water over a Co/Cu/Mn/Mo catalyst as described above. The prehydrogenated feed had a water content of about 52.5% by weight. As organic components, it comprised, according to gas-chromatographic analysis, calculated on a water-free basis, 61.1% by weight of 1,6-hexanediol and further reaction products of the abovementioned by-product stream. The acid number of the prehydrogenated feed was 6.5 mg KOH/g and was predominantly attributable to 6-hydroxycaproic acid and traces of adipic acid. Further components in the prehydrogenated feed were, inter alia, 1-hexanol, 1-pentanol, 1,2-cyclohexanediol, 1,3-cyclohexanediol, 1,4-cyclohexanediol, 1,5-pentanediol, 1-pentanol and 1,4-butanediol.

(25) To carry out the hydrogenation of the 6-hydroxycaproic acid and of the adipic acid, the prehydrogenated feed was passed continuously in the downflow mode over the catalyst at a reactor inlet temperature of 230 C., 25 MPa abs together with 50 standard liters/h of hydrogen per kg of prehydrogenated feed. In the present trial, the amount of prehydrogenated feed fed in was gradually increased, with sampling being carried out in each case about 24 hours after setting of an inflow amount.

(26) The results are shown in table 2. The water content was determined by the Karl-Fischer method and the content of 1,4-butanediol was determined gas-chromatographically.

(27) At a low flow velocity of only 5 m/h (example 5), the carboxylic acids present are hydrogenated largely completely, which is indicated by a very low acid number of <0.5 mg KOH/g, but appreciable amounts of the target alcohol 1,6-hexanediol are also destroyed by hydrogenation. Thus, the content of 1,6-hexanediol decreases from an original 61.1% by weight to 60.2% by weight, in each case calculated on a water-free basis.

(28) A high flow velocity of 60 m/h (example 8) does give a high calculated throughput through the reactor but allows only partial hydrogenation of the carboxylic acids present. Thus, the output still has an acid number of 2.3 mg KOH/g, which corresponds to hydrogenation of only about 65% of the carboxylic acids present.

(29) In comparison, examples 6 and 7 with a flow velocity of 10 and 30 m/h, respectively, show virtually complete hydrogenation of the carboxylic acids present, as indicated by an acid number in the output of <0.5 mg KOH/g, and also a significant increase in target alcohol 1,6-hexanediol, as indicated by a significantly increased content of 1,6-hexanediol of 64.0 and 64.1% by weight, respectively.

(30) A virtually carboxylic acid-free product stream with a significant increase in target alcohol was obtained by the hydrogenation according to the invention from a feed stream comprising alcohol and carboxylic acid.

Example 9

(31) In example 9 (comparative example), the long-term behavior of the hydrogenation of 6-hydroxycaproic acid to 1,6-hexanediol at a flow velocity of the flowing liquid calculated on the basis of the geometric cross-sectional area of the empty, catalyst-free reactor tube of 60 m/h was examined. For this purpose, example 8 was firstly repeated and left under these conditions for a period of 1000 hours. During this time, the acid number in the output slowly increased from 2.3 to 4.5 mg KOH/g. Correspondingly, the carboxylic acid conversion decreased from about 65 to about 30%. At the same time, small amounts of Co and Mn totaling up to 10 ppm by weight were found in the output.

(32) Comparative example 9 shows a tremendous deterioration in the performance within only 1000 hours,

Example 10

(33) In example 10, the long-term behavior of the hydrogenation of 6-hydroxycaproic acid to 1,6-hexanediol at a flow velocity of the flowing liquid calculated on the basis of the geometric cross-sectional area of the empty, catalyst-free reactor tube of 30 m/h was examined. For this purpose, example 7 was firstly repeated and left under these conditions for a period of 3000 hours. During this time, the acid number in the output slowly increased from <0.5 to just 0.6 mg KOH/g. The carboxylic acid conversion after 3000 hours was thus still above 90%. Co and Mn were detected in only a small total amount of <4 ppm by weight.

(34) Example 10 shows only a slight increase in the acid number and in the Co and Mn content in the output even after 3000 hours of operation at a flow velocity of 30 m/h.

(35) TABLE-US-00001 TABLE 1 Cross-sectional Feed Output Feed throughput .sup.#1 Acid number 1,4-BDO .sup.#2 Water Acid number 1,4-BDO .sup.#2 Water Example [l/h] [m/h] [mg KOH/g] [% by wt] [% by wt] [mg KOH/g] [% by wt] [% by wt] 1 (comparison) 1.6 5 3.8 90.7 75.0 <0.5 88.0 75.5 2 (invention) 3.1 10 <0.5 94.5 75.2 3 (invention) 9.4 30 <0.5 94.3 75.1 4 (comparison) 19 60 1.5 93.0 75.1 .sup.#1 Flow velocity of the flowing liquid calculated on the basis of the geometric cross-sectional area of the empty, catalyst-free reactor tube. .sup.#2 1,4-Butanediol, calculated on a water-free basis.

(36) TABLE-US-00002 TABLE 2 Cross-sectional Feed Output Feed throughput .sup.#1 Acid number 1,6-HDO .sup.#2 Water Acid number 1,6-HDO .sup.#2 Water Example [l/h] [m/h] [mg KOH/g] [% by wt] [% by wt] [mg KOH/g] [% by wt] [% by wt] 5 (comparison) 1.6 5 6.5 61.1 52.5 <0.5 60.2 53.5 6 (invention) 3.1 10 <0.5 64.0 53.0 7 (invention) 9.4 30 <0.5 64.1 52.9 8 (comparison) 19 60 2.3 61.0 52.7 .sup.#1 Flow velocity of the flowing liquid calculated on the basis of the geometric cross-sectional area of the empty, catalyst-free reactor tube. .sup.#2 1,6-Hexanediol, calculated on a water-free basis.