Process for preparing acrylic acid

Abstract

The present invention relates to a process for preparing acrylic acid from acetic acid and formaldehyde, which comprises (a) provision of a stream S1 comprising acetic acid and formaldehyde, where the molar ratio of acetic acid to formaldehyde in the stream S1 is in the range from 0.5:1 to 2:1; (b) contacting of the stream S1 with an aldol condensation catalyst comprising vanadium, phosphorus and oxygen to give a stream S2 comprising acrylic acid, where, in (b), the space velocity WHSV is in the range from 0.35 to 7.0 kg/kg/h.

Claims

1. A process for preparing acrylic acid from acetic acid and formaldehyde, the process comprising: (a) providing a stream S1 comprising acetic acid and formaldehyde, where a molar ratio of acetic acid to formaldehyde in the stream S1 is from 0.5:1 to 2:1; (b) contacting the stream S1 with an aldol condensation catalyst comprising vanadium, phosphorus and oxygen to give a stream S2 comprising acrylic acid, wherein, in (b), a space velocity WHSV, defined ([Formaldehyde]+[acetic acid])/[aldol condensation catalyst]/time, is from 0.75 to 3.5 kg/kg/h, wherein [formaldehyde], [acetic acid], and [aldol condensation catalyst] represent mass of formaldehyde, acetic acid, and the aldol condensation catalyst, respectively, and wherein the aldol condensation catalyst in (b) comprises SiO.sub.2 as a support material.

2. The process according to claim 1, wherein the content of vanadium in the aldol condensation catalyst as per (b) is from 2 to 20% by weight, the content of phosphorus is from 2 to 20% by weight, and the content of oxygen is from 20 to 60% by weight, in each case based on the total weight of the aldol condensation catalyst.

3. The process according to claim 1, wherein the aldol condensation catalyst in (b) additionally comprises at least one further element M selected from the group consisting of Bi, W, Sn, Ti, Fe, Mn, Cr, Cu, K, Cs, Li, Mg and Ca.

4. The process according to claim 1, wherein the aldol condensation catalyst in (b) is present as a powder.

5. The process according to claim 1, wherein the aldol condensation catalyst in (b) does not comprise other elements in addition to vanadium, phosphorus and oxygen, has a molar ratio of vanadium to phosphorus of from 1:2 to 1:1.75, and the content of vanadium is from 5.6 to 6.2% by weight, the content of phosphorus is from 6.8 to 7.4% by weight, the content of oxygen is from 55 to 60% by weight, and the content of silicon is from 28 to 30% by weight, in each case based on the total weight of the aldol condensation catalyst.

6. The process according to claim 1, wherein the aldol condensation catalyst in (b) does not comprise other elements in addition to vanadium, phosphorus and oxygen, has a molar ratio of vanadium to phosphorus of from 1:2 to 1:1.75, and the content of vanadium is from 6.3 to 6.9% by weight, the content of phosphorus is from 7.5 to 8.1% by weight, the content of oxygen is from 50 to 55% by weight, and the content of silicon is from 31 to 33% by weight, in each case based on the total weight of the aldol condensation catalyst.

7. The process according to claim 1, wherein the aldol condensation catalyst in (b) comprises vanadium, phosphorus, oxygen, tungsten, and bismuth, has a molar ratio of (V+W+Bi) to phosphorus of from 1.75:1 to 2.25:1, and the content of vanadium is from 10 to 13% by weight, the content of phosphorus is from 5 to 7% by weight, the content of oxygen is from 35 to 40% by weight, the content of bismuth is from 7 to 10% by weight, the content of tungsten is from 7 to 10% by weight, and the content of silicon is from 4 to 7% by weight, in each case based on the total weight of the aldol condensation catalyst.

8. The process according to claim 1, wherein the molar ratio of acetic acid to formaldehyde in the stream S1 is from 0.75:1 to 1.5:1.

9. The process according to claim 1, wherein the stream S1 comprises at least one further component in addition to acetic acid and formaldehyde.

10. The process according to claim 9, wherein the at least one further component is selected from the group consisting of water, oxygen and an inert gas.

11. The process according to claim 10, wherein the stream S1 comprises from 50 to 100% by volume of acetic acid, formaldehyde, water, oxygen and the inert gas.

12. The process according to claim 9, wherein the stream S1 additionally comprises propionic acid.

13. The process according to claim 1, wherein the contacting (b) is carried out at a temperature of the catalyst bed of more than 320 C.

14. The process according to claim 1, wherein the stream S2 comprises propionic acid of not more than 2500 ppm by weight based on the acrylic acid comprised in S2.

15. The process according to claim 10, wherein the stream S1 additionally comprises propionic acid.

Description

DESCRIPTION OF THE FIGURE

(1) FIG. 1 shows a plot of the propionic acid content in ppm by weight in stream S2 based on the acrylic acid comprised in S2 (ordinate) at the respective space velocity WHSV in kg/kg/h, defined as (mass(formaldehyde)+mass(acetic acid))/mass(aldol condensation catalyst)/time (abscissa). The ordinate extends from 0 to 14 000 ppm by weight of propionic acid in stream S2, based on the acrylic acid comprised in S2, and the abscissa extends from 0.00 to 8.50 kg/kg/h. The results for catalyst 1 (.circle-solid.), catalyst 2 (.square-solid.), catalyst 3 (.box-tangle-solidup.) and catalyst 4 (.diamond-solid.) are shown.

(2) The present invention is illustrated by the following examples.

EXAMPLES

(3) I. Analytical Methods

(4) I.1 Gas Chromatography

(5) The analysis of the gaseous product stream was carried out with the aid of an on-line GCMS system from Agilent. The instrument was equipped with a 10-way valve having two sample loops (500 microliters/1000 microliters) which were operated at 220 C. Detection was effected by means of a flame ionization detector (FID) and two thermal conductivity detectors. For the FID stream introduced through the front inlet, the following parameters were selected: injector temperature: 275 C.; split: 1:5. An FFAP column having a length of 30 m, an internal diameter of 0.32 mm and a film thickness of 0.5 microns (column inflow: 5 ml/min) was used. The sample was fed to the thermal conductivity detectors in parallel through the rear inlet with the aid of a Y-adapter (JAS). Here, the following parameters were selected: injector temperature: 275 C.; split 1:2. For the first thermal conductivity detector, a column of the volamine type having a length of 60 m, an internal diameter of 0.32 mm and a film thickness of 0.45 microns (column flow: 2 ml/min) was used. The second thermal conductivity detector had a column system comprising two columns. First column: RTX5 having a length of 30 m, an internal diameter of 0.32 mm, a film thickness of 1 micron (column flow: 5 ml/min). Second column: select permanent gases/CO.sub.2 HR having a length of 50 m, an internal diameter of 0.32 mm and a film thickness of 10 microns (column flow: 2 ml/min). All columns were operated using helium as carrier gas. The GC oven temperature program was as follows: 40 C. (2.5 min hold time) heating to 105 C. at a heating rate of 20 K/min (0 min hold time) heating to 225 C. at a heating rate of 40 K/min (2.75 min hold time)
I.2 Elemental Analysis

(6) The determination of the quantitative composition of the catalysts was carried out by means of wavelength-dispersive X-ray fluorescence analysis.

(7) II. Production of the Catalysts

(8) II.1 Catalyst 1 (BiWVPO)

(9) 100 g of citric acid were dissolved in 1000 ml of DI water (deionized water) and heated to 80 C. 67 g of bismuth(III) acetate, 117.6 g of 85% strength by weight aqueous phosphoric acid, 100 g of ethylene glycol, 116 g of Ludox AS-40 (40% by weight of colloidal SiO.sub.2 in water) were added; the mixture obtained was stirred at 80 C. for 30 minutes. 110 g of ammonium metavanadate (NH.sub.4VO.sub.3), 169 g of ammonium metatungstate ((NH.sub.4).sub.6H.sub.2W.sub.12O.sub.40.H.sub.2O) and 20 g of methylcellulose were added, and the suspension was stirred at 80 C. for 3 hours.

(10) Volatile constituents of the suspension were removed at 60 C. and 45 mbar on a rotary evaporator; the powder obtained was dried at 100 C. in a convection drying oven for 16 hours and subsequently calcined according to the following temperature program in a muffle furnace: 10 K/min to 160 C., hold for 2 h; 3 K/min to 250 C., hold for 2 h; 3 K/min to 300 C., hold for 6 h; 3 K/min to 450 C., hold for 6 h.

(11) The material obtained had a bismuth content of 8.5% by weight, a tungsten content of 31% by weight, a vanadium content of 11.4% by weight, a phosphorus content of 6.5% by weight and a silicon content of 5.6% by weight. The difference to 100% by weight corresponds to the oxygen content.

(12) The powder was tableted and then comminuted to 350-500 m.

(13) II.2 Catalyst 2 (VPO SiO.sub.2)

(14) Silica gel (Siliperl AF 125, 3-5 mm balls, manufacturer: BASF) was comminuted to the desired fraction (350-500 microns) and used as support material. 400 g of support material (Siliperl AF 125) was impregnated with 372 ml of 2.0 molar aqueous vanadium(III) citrate solution. The mixture was subsequently dried at 80 C. in a convection drying oven for 16 hours.

(15) 168.2 g of 85% strength by weight aqueous phosphoric acid was diluted with DI water to 372 ml, and the above-described, dried material was impregnated therewith. The mixture was subsequently dried at 80 C. in a convection drying oven for 16 hours.

(16) The mixture comprising V and P was calcined according to the following temperature program in a muffle furnace:

(17) 1 K/min to 260 C., hold for 2 h.

(18) The material obtained had a vanadium content of 5.9% by weight, a phosphorus content of 7.1% by weight and a silicon content of 29.3% by weight. The difference to 100% by weight corresponds to the oxygen content.

(19) II.3 Catalyst 3 (VPO SiO.sub.2)

(20) Silica gel (CARiACT Q-20C, 3-5 mm balls, manufacturer: Fuji Silysia) was comminuted to the desired fraction (350-500 microns) and used as support material. 100 g of support material (CARiACT Q-20C) was impregnated with 84 ml of 2.2 molar aqueous vanadium(III) citrate solution. The mixture was subsequently dried at 80 C. in a convection drying oven for 16 hours.

(21) 42 g of 85% strength by weight phosphoric acid were diluted with DI water to 84 ml, and the above-described, dried material was impregnated therewith. The mixture was subsequently dried at 80 C. in a convection drying oven for 16 hours.

(22) The mixture comprising V and P was calcined according to the following temperature program in a muffle furnace:

(23) 1 K/min to 260 C., hold for 2 h.

(24) The material obtained had a vanadium content of 6.6% by weight, a phosphorus content of 7.8% by weight and a silicon content of 32.0% by weight. The difference to 100% by weight corresponds to the oxygen content.

(25) II.4 Catalyst 4 (Sn Zeolite)

(26) Deboronized zeolitic material of the structure type BEA was firstly produced as described in Example 6, sections 6.1 and 6.2, of WO 2013/117537 A1. 50 g of this zeolitic material were placed together with 14.2 g of tin(II) acetate (Sn(OAc).sub.2) in a mixer (Microton MB550 mill) and the mixture was milled for 15 minutes at 14 000 rpm (revolutions per minute). After milling, the mixture was placed in a porcelain dish and calcined at 500 C. for 3 hours under nitrogen, followed by 3 hours in air (heating rate 2 K/min). The material obtained had a tin content of 13.1% by weight, a silicon content of 38% by weight and a total content of organic carbon (TOC) of less than 0.1% by weight.

(27) III. Catalytic Studies

(28) III.1 Procedure

(29) The catalytic studies were in each case carried out on a 1 ml pulverulent sample; a crushed material fraction having a particle size in the range from 0.315 to 0.5 mm was used for this purpose. The WHSV was varied either by adjusting the gas stream S1 or by diluting the respective catalyst as per II with steatite particles (total volume of catalyst and steatite constant at 1 ml). The catalysts as per II, optionally diluted with steatite particles, were positioned in tube reactors between two inert particle beds of crushed fused silica and the loaded tube reactors were installed in the catalysis apparatus (16-fold high-throughput screening plant).

(30) A stream composed of technical-grade formalin (49% formaldehyde solution in water), acetic acid (100% Bernd Kraft, 16873.4000), nitrogen (Praxair, purity 5.0) and oxygen (synthetix air, Praxair, purity max.10% rel.) was heated to 200 C. (composition of the stream: acetic acid 9% by volume, formaldehyde 9% by volume, water 15% by volume, oxygen 1.5% by volume, nitrogen 65.5% by volume; molar ratio of acetic acid to formaldehyde: 1:1) and thereby vaporized. The gaseous mixture was then brought into contact with an aldol condensation catalyst (350-500 m) as per II in the form of crushed material at 370 C. and 1.1 bar.

(31) The temperature was measured by means of a thermocouple in the isothermal zone of the reactor, i.e. the catalyst bed, at the beginning of the experiment and corresponds to the temperature at which the reactions were carried out. The product stream was diluted with argon (purity: 5.0) (Ar: product stream=22:1) and the composition was determined by gas chromatography.

(32) The data shown in FIG. 1 and discussed below demonstrate the result found, with the process of the invention being operated for 8 hours.

(33) III.2 Evaluation

(34) The concentration of propionic acid based on acrylic acid (Con.sub.PRA) in the stream S2 is calculated according to the following formula:
Con.sub.PRA=100*(NC.sup.P.sub.PRA/NC.sup.P.sub.ACA) NC.sup.P.sub.PRA=number of carbon atoms comprised in the form of propionic acid in the stream S2. NC.sup.P.sub.ACA=number of carbon atoms comprised in the form of acrylic acid in the stream S2.

(35) It was surprisingly found in the context of the present invention that when a specific WHSV is selected, a low content of propionic acid based on acrylic acid can be obtained in the stream S2 obtained directly from (b). Especially in the range from 0.75 to 3.5 kg/kg/h, particularly in the range from 1.0 to 3.5 kg/kg/h, the results for the catalysts 1, 2 and 3 display particularly low values for the content of propionic acid based on acrylic acid in the stream S2 obtained directly from (b).

LITERATURE CITED

(36) Vitcha and Sims, I & EC Product Research and Development, Vol. 5, No. 1, March 1966, pages 50 to 53 DE 169 27 850 A1 EP 0 616 998 A1