Gas-phase and liquid-gas-phase nitrilation process

Abstract

A process for the nitrilation of a fatty acid or of a fatty acid ester, which is optionally unsaturated, by reacting the fatty acid or fatty acid ester with ammonia in a reactor operating continuously in the gas phase or in the mixed gas-liquid phase in a temperature range of from 180 to 400 C., in the presence of a solid catalyst comprising at least one metal oxide, the metal of which belongs to column 8 of the periodic table, as a mixture with at least one metal oxide chosen from aluminum oxides, zirconium oxides, niobium oxides, tantalum oxides and tin oxides, the metal oxide(s), the metal of which belongs to column 8, being present in a volume ratio of 0.1 to 0.6 relative to the volume of the mixture of all the oxides.

Claims

1. A process for the nitrilation of a fatty acid or of a fatty acid ester to the corresponding nitrile, which is optionally unsaturated, by reacting the fatty acid or fatty acid ester with ammonia in a reactor operating continuously in a gas phase or in a mixed gas-liquid phase in a temperature range of from 180 to 400 C. to produce the corresponding nitrile, in the presence of a solid catalyst comprising: at least one metal oxide, the metal of which belongs to column 8 of the periodic table, as a mixture with at least one metal oxide selected from the group consisting of aluminum oxides, zirconium oxides, niobium oxides, tantalum oxides and tin oxides, the metal oxide(s), the metal of which belongs to column 8 of the periodic table, being present in a volume ratio of 0.1 to 0.6 relative to the volume of the mixture of all the oxides.

2. The process as claimed in claim 1, wherein the catalyst comprises at least as metal oxide, the metal of which belongs to column 8 of the periodic table, ferric oxide.

3. The process as claimed in claim 1, wherein the catalyst comprises ferric oxide (Fe.sub.2O.sub.3) and niobium pentoxide (Nb.sub.2O.sub.5).

4. The process as claimed in claim 1, wherein the catalyst comprises ferric oxide (Fe.sub.2O.sub.3) and aluminum oxide (Al.sub.2O.sub.3).

5. The process as claimed in claim 1, wherein the catalyst comprises ferric oxide (Fe.sub.2O.sub.3) and zirconium oxide (ZrO.sub.2).

6. The process as claimed in claim 1, wherein the fatty acid or fatty acid ester is selected from the group consisting of w-unsaturated acids or esters having the following formula:
CH.sub.2CH(CH.sub.2).sub.nCOOR in which n represents the integer 7 or 8 and R represents either a hydrogen atom or an alkyl radical comprising 1 to 4 carbon atoms.

7. The process as claimed in claim 1, wherein the reaction temperature ranges from 200 C. to 300 C.

8. The process as claimed in claim 1, wherein the reaction temperature ranges from 200 C. to 250 C.

9. The process as claimed in claim 1, which is operated in a gas phase and not in a mixed gas-liquid phase.

10. The process as claimed in claim 1, which is operated in a mixed gas-liquid phase.

11. The process as claimed in claim 1, wherein a fatty acid is reacted.

12. The process as claimed in claim 1, wherein an unsaturated fatty acid ester is reacted and with the absence of isomerization of the double bond thereof.

13. The process as claimed in claim 1, wherein the fatty acid or of a fatty acid ester contains 10 or 11 carbon atoms per molecule.

14. The process as claimed in claim 1, wherein a fatty acid ester is reacted and the fatty acid ester is methyl 9-decenoate or methyl 10-undecenoate.

15. The process as claimed in claim 1, wherein the pressure exerted in the reactor is 0.1 to 10 atmospheres.

16. The process as claimed in claim 1, wherein the ratio of NH.sub.3 to fatty acid or fatty acid ester is 1 to 50.

17. The process as claimed in claim 1, which is conducted in the gas phase and the contact time with the solid catalyst is 1 second to 300 seconds.

18. The process as claimed in claim 1, which is conducted in the mixed gas-liquid phase and wherein the average contact time of the liquid phase in the reactor is less than 1 hour.

19. The process as claimed in claim 1, where the catalyst comprises Fe.sub.2O.sub.3 and Al.sub.2O.sub.3, and wherein the volume ratio of Fe.sub.2O.sub.3 to the sum of Fe.sub.2O.sub.3 and Al.sub.2O.sub.3 is 0.1 to 0.6.

20. The process as claimed in claim 19, wherein the volume ratio of the Fe.sub.2O.sub.3 to the sum of the Fe.sub.2O.sub.3 and Al.sub.2O.sub.3 is 0.2 to 0.5.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIGS. 1 and 2 illustrate certain of the experimental results obtained, as described in the Examples.

DESCRIPTION OF CERTAIN EMBODIMENTS OF THE INVENTION

(2) For the purposes of the present invention, the term mixed gas-liquid phase is intended to mean a process using a mixed gas-liquid fluid including a liquid as continuous phase and a gas as dispersed phase or else a liquid as dispersed phase and a gas as continuous phase. In a mode in which liquid descends on a catalyst in a fixed bed, there is a trickle-bed configuration, in a mode in which liquid rises and catalyst is in a fixed bed, there is an immersed-bed configuration. The reactor can also operate in fluidized-bed mode. In this configuration, the catalyst is kept fluidized not only by the gas stream, but also by the vaporization of liquid feedstock which thus generates a large volume of gas.

(3) Nitrilation Process

(4) The invention relates to a process for the nitrilation of a fatty acid or of a fatty acid ester, which is optionally unsaturated, by reacting ammonia in a reactor operating continuously in the gas phase or in the mixed gas-liquid phase in a temperature range of from 180 to 400 C., in the presence of a solid catalyst comprising at least one metal oxide, the metal of which belongs to column 8 of the periodic table, as a mixture with at least one metal oxide chosen from aluminum oxides, zirconium oxides, niobium oxides, tantalum oxides and tin oxides, the metal oxide(s), the metal of which belongs to column 8, being present in a volume ratio of 0.1 to 0.6 relative to the volume of the mixture of all the oxides.

(5) The reagents of the nitrilation process according to the invention may be fatty acids or esters, which are optionally unsaturated, preferably -unsaturated.

(6) For the purpose of the present invention, the term fatty is intended to mean an acid or an ester comprising a saturated or unsaturated, linear carbon chain comprising from 8 to 36 carbon atoms. Preferably, the acids and esters that are of use for the process of the invention have the following formula:
CH.sub.2CH(CH.sub.2).sub.nCOOR

(7) in which

(8) n represents the integer 7 or 8 and

(9) R represents either a hydrogen atom or an alkyl radical comprising 1 to 4 carbon atoms.

(10) Preferably, the process of the invention uses, as feedstock, -unsaturated acids or esters comprising either 10 carbon atoms or 11 carbon atoms per molecule. The first, in particular methyl 9-decenoate, are sold in ester form by the company Elevance Renewable Sciences. The second, in particular methyl 10-undecenoate, are produced by the company Arkema in the context of its castor oil-based process, methyl undecylenate being obtained after pyrolysis.

(11) The nitrilation step is carried out in a reactor operating continuously, i.e. in which the reagents, whether they are of gas, solid or liquid origin, are introduced (and the products extracted) into the reactor continuously according to predetermined flow rates.

(12) The process for the nitrilation of the fatty acids and the esters is carried out at a reaction temperature ranging from 180 to 400 C., and preferably from 200 to 300 C. and more preferably from 200 to 250 C. The feedstock of the fatty acids or esters, which are optionally unsaturated, is vaporized and brought to a temperature ranging from 180 to 350 C. on contact with the ammonia, the introduction temperature of which ranges from 150 to 600 C.

(13) The pressure exerted in the reactor can range from 0.1 to 10 atmospheres (absolute), and preferably from 0.5 to 5 atm, and even more preferably from 1 to 3 atm.

(14) The NH.sub.3/fatty ester or NH.sub.3/fatty acid molar ratio of the reagents can range from 1 to 50, preferably from 3 to 30, and even more preferably from 5 to 20.

(15) In the Pure Gas Phase

(16) In a first embodiment, the two reagents can be introduced into the reactor in the gas state (pure gas phase).

(17) The fatty acid or the fatty acid ester, which is optionally unsaturated, is vaporized and brought to a temperature ranging from 180 to 350 C. on contact with the ammonia, the introduction temperature of which ranges from 150 to 600 C., and under a pressure ranging from 0.1 to 10 atmospheres (absolute), preferably from 0.5 to 5 atmospheres and even more preferably from 1 to 3 atmospheres.

(18) The flow rates for introducing the reagents are such that the contact time with the solid catalyst ranges from 1 second to 300 seconds. In this case, the contact time is determined by the ratio calculated as follows: {volume of catalyst (in liters)3600}/{[flow rate of the ester or of the acid (in moles/h)+flow rate of ammonia (in moles/h)]22.4}=contact time in seconds.

(19) In the Mixed Phase

(20) Preferably, the mixed-phase nitrilation process is carried out with the fatty acids.

(21) In the other embodiment (mixed phase), the ammonia is introduced in gas form, whereas the acid is introduced, after optional preheating, into the reactor in proximity to the catalytic bed, at least partly in liquid form at a flow rate determined so as to flow in the form of a film (trickle bed) on the heated catalytic bed on contact with which a fraction of the liquid is vaporized. The reaction or the series of reactions takes place on contact with the surface of the catalyst or in its immediate proximity. This trickle-bed technique is well known and widely used in the oil industry. The ammonia stream may be cocurrent or countercurrent with respect to the acid stream.

(22) The acid introduction flow rate is such that the average contact time of the liquid phase in the reactor is less than 1 hour, and preferably less than 30 minutes. This contact time is determined by the following calculation: volume of catalyst (in liters)/acid flow rate (in liquid liters at 25 C. per hour), that is to say the inverse of the liquid hourly liquid volume rate.

(23) In this embodiment, it is possible to work in cocurrent mode, i.e. the gas current and the liquid stream are descending, or in the countercurrent mode, with the gas stream being ascending and the liquid current descending. The latter variant is preferred in the process of the invention. The countercurrent version, with the gas ascending and the acid descending, may be particularly advantageous for limiting the hydrolysis of the nitrile formed. This is because, in this configuration, the ammonia is injected at the bottom, the water and the alcohol leave at the top, the acid enters at the top and the nitrile leaves at the bottom. At the bottom, therefore, the concentration of nitrile and of ammonia is high, and at the top the concentration of water and alcohol is high, and the concentration of ammonia is lower. It is therefore possible to shift the equilibria, in particular that of the hydrolysis of the nitrile, which restores the acid.

(24) Catalyst

(25) A subject of the present invention is also a solid catalyst.

(26) The solid catalyst according to the present invention comprises at least one metal oxide, the metal of which belongs to column 8 of the periodic table, as a mixture with at least one metal oxide chosen from aluminum oxides, zirconium oxides, niobium oxides, tantalum oxides and tin oxides,

(27) the metal oxide(s), the metal of which belongs to column 8, being present in a volume ratio of 0.1 to 0.6 relative to the volume of the mixture of all the metal oxides.

(28) Preferably, the metal oxide, the metal of which belongs to column 8 of the periodic table, is an iron oxide. It is in particular chosen from FeO, Fe.sub.3O.sub.4 and Fe.sub.2O.sub.3. The metal oxide, the metal of which belongs to column 8 of the periodic table, which is preferred is ferric oxide: Fe.sub.2O.sub.3.

(29) Preferably, the metal oxides chosen from aluminum oxides, zirconium oxides, niobium oxides, tantalum oxides and tin oxides are chosen from aluminum oxides, zirconium oxides and niobium oxides. More particularly, they are chosen from aluminum oxide: Al.sub.2O.sub.3, zirconium oxide ZrO.sub.2 and niobium pentoxide: Nb.sub.2O.sub.5.

(30) Some of these oxides exist in certain crystallographic forms. Preferably, the alumina used is the gamma-alumina sold by the company BASF under the commercial reference AL-3996, or else by the companies Axens and Sasol.

(31) The zirconia (ZrO.sub.2) used is sold by the companies Norpro-St Gobain, Daiichi Kigenso KK, and MEL, the niobium oxide (Nb.sub.2O.sub.5) is sold by the companies Starck and CBMM, and the iron III oxide hydrate is sold, inter alia, by the company Sigma-Aldrich (FeO(OH)) (catalyst grade, 30-50 mesh broken and sieved).

(32) Preferably, the solid catalyst according to the invention comprises the following combinations: ferric oxide: Fe.sub.2O.sub.3 on a niobium pentoxide support: Nb.sub.2O.sub.5; ferric oxide: Fe.sub.2O.sub.3 on an aluminum oxide support: Al.sub.2O.sub.3, and also ferric oxide: Fe.sub.2O.sub.3 on a zirconium oxide support: ZrO.sub.2.

(33) The volume ratio of the metal oxide, the metal of which belongs to column 8 of the periodic table, to the volume of the mixture of all the oxides ranges from 0.1 to 0.6.

(34) Preferably, the volume ratio of the iron oxide/alumina oxide pair ranges from 0.2 to 0.5.

(35) Preferably, the volume ratio of the iron oxide/zirconium oxide pair ranges from 0.2 to 0.5.

(36) Preferably, the volume ratio of the iron oxide/niobium oxide pair ranges from 0.2 to 0.5, in particular from 0.25 to 0.4.

(37) Preferably, the catalyst according to the present invention is characterized by a specific surface area ranging from 10 to 500 m.sup.2/g, and preferably from 40 to 300 m.sup.2/g, more preferentially from 40 to 250 m.sup.2/g, and in particular from 40 to 200 m.sup.2/g. The term BET Brunauer, Emmett and Teller) specific surface area is intended to mean the available surface area per gram of material. This measurement is based on an adsorption of gas at the surface of the solid studied. The measurement of the specific surface area is carried out according to ASTM standard D 3663-84.

(38) Preferably, the catalyst is characterized by a pore size distribution such that less than 20% of the pore volume is in the pores with a diameter of less than 2 nm, and preferably less than 3.5 nm, and even more preferably less than 7 nm. The pore sizes are calculated according to the methods ASTM D4222-83 (by adsorption of nitrogen) for the pore volume distribution measurement and ASTM D4641-87 for the pore size distribution calculation.

(39) The catalyst may be in the form of beads, of extruded objects, of pellets, in cylindrical or polylobe shape, or else in the form of a hollow cylinder with one or more holes, or else in the form of a cylinder having notches along ridges, this being so as to increase the ratio of the external surface area of the grain relative to the volume of the grain. This criterion is important for reducing diffusional limitations.

(40) Preferably, for a catalyst on a fixed bed, the grains have a size of 1 to 8 mm, and preferably of 3 to 5 mm in their industrial use, this dimension being according to their largest length.

(41) Preferably, for a catalysis on a fluidized bed, the grains have an average size of from 40 to 300 m, and preferably of from 80 to 150 m, in their industrial use; the catalysts are preferably in the form of microbeads.

(42) Catalyst Preparation Process

(43) Several methods may be suitable as method for preparing the catalysts: coprecipitation of a salt or of a salt mixture; blending of precursors generally in the form of salts, of oxides or of hydroxides; impregnation of one compound with another, for example impregnation of aluminum, zirconium or niobium oxides with a solution containing an iron oxide precursor; reactive milling, in which the two oxides are intimately mixed by high-energy milling, which results in the formation of a new compound; or else atomization. The precursors of the oxides in various forms can be used in particular in oxide, nitrate, carbonate, chloride, sulfate (including oxysulfate), phosphate, organometallic compound, acetate or acetylacetonate form. In the case in point, the preparation of a catalyst from zirconium oxysulfate results in a catalyst suitable for the process of the invention.

(44) The mixing of the metal oxides according to the present invention is preferably mechanical mixing.

(45) According to one embodiment of the invention, the oxides can be separately milled, finely, preferably so as to achieve particle sizes ranging from 1 to 8 mm according to their largest length, and then measured volumes are mixed and homogenized.

(46) Finally, the invention is directed toward the use of the catalyst as defined above, in a process for the nitrilation of fatty acids or of fatty acid esters.

(47) The following examples serve to illustrate the invention without, however, being limiting in nature.

EXAMPLES

(48) The catalysts used in the examples have the following characteristics, which appear in table 1 below:

(49) TABLE-US-00001 TABLE 1 BET surface area Pore volume Catalyst Supplier (m.sup.2/g) (cm.sup.3/g) Al.sub.2O.sub.3 BASF 196 0.5 ZrO.sub.2 Norpro St- 53 nc Gobain Fe.sub.2O.sub.3 Aldrich 141 nc

Example 1: Test on the Methyl Ester of Lauric Acid

(50) Tests were carried out on the methyl ester of lauric acid.

(51) Procedure:

(52) The assembly consisted of an evaporation chamber, where the ester in its liquid state is continuously fed via a peristaltic pump. A controlled stream of dry nitrogen entrains the ester and transports it to the catalytic bed below. The controlled stream of ammonia encounters the stream of ester+nitrogen flush with the catalytic bed which is a cylinder 8 mm in diameter by 30 mm held on a stainless steel frit. The stream exiting is condensed a first time at 150-170 C. in order to recover the lauric compounds (nitriles, amide, acid, methylamide, etc), and a second time at 12 C. then 77 C. (via a dry ice trap) in order to condense the light compounds. The condensate is removed and analyzed by GC-FID and GC-MS, on the basis of which the concentrations of ester, nitrile and optionally amide and N-methylamide are calculated.

(53) The metal oxides were separately finely milled and then measured volumes were combined and homogenized.

(54) Results:

(55) 1. With the Fe.sub.2O.sub.3 and Al.sub.2O.sub.3 Catalyst

(56) This experiment was carried out with 3 different catalysts: alumina alone: Al.sub.2O.sub.3, iron oxide alone: Fe.sub.2O.sub.3 and a mixture of Al.sub.2O and Fe.sub.2O.sub.3 in an Fe.sub.2O.sub.3/(Al.sub.2O.sub.3+Fe.sub.2O.sub.3) volume ratio of 0.5.

(57) The average residence time is 4.5 seconds. The results in terms of conversion and of nitrilation and relative to the formation of the N-methylated by-product, each expressed as molar percentage, appear in the tables below.

(58) 1.1 At 200 C.

(59) TABLE-US-00002 TABLE 2 Fe.sub.2O.sub.3/(Al.sub.2O.sub.3 + Fe.sub.2O.sub.3) volume ratio 0.0 0.5 1 Conversion 18.3 21.8 7.2 (mol %) Nitrilation 10.3 17.4 5.6 (mol %) N-Methylamide 0.0 0.0 0.3 content (mol %)

(60) It is noted that, even at 200 C., which is a relatively low temperature, the mixture of catalysts according to the invention results in an increase for the nitrilation of 71% calculated relative to the value obtained for the alumina alone, and of 211% calculated relative to the value obtained for the iron oxide alone. Furthermore, there is no formation of N-methylamide.

(61) In addition, compared with the results obtained for the alumina alone, it is noted that the difference obtained between the conversion content and the nitrilation content is much smaller, thus indicating the low by-product content obtained with the catalyst according to the invention.

(62) 1.2 At 250 C.

(63) The results appear in table 3 below and in FIG. 1.

(64) TABLE-US-00003 TABLE 3 Fe.sub.2O.sub.3/(Al.sub.2O.sub.3 + Fe.sub.2O.sub.3) volume ratio 0.0 0.5 1 Conversion 52.6 76.1 38.7 (mol %) Nitrilation 38.2 72.7 36.9 (mol %) N-Methylamide 1.2 1.0 0.4 content (mol %)

(65) These results show a clear improvement in the conversion content and the nitrilation content for the catalyst according to the invention compared with the catalysts alone.

(66) Consequently, for this very short residence time, the tendency obtained at 200 C. is retained, namely good results in terms of conversion and of nitrilation with a very small difference between these two values, indicating a low by-product content. The formation of N-methylamide is limited, and is lower for the mixture of metal oxides than for the pure compounds.

(67) Finally, FIG. 1 provides evidence of the expected effect linked to the values of the volume ratio according to the invention.

(68) 2 With the Fe.sub.2O.sub.3 et ZrO.sub.2 Catalyst

(69) This experiment was carried out with 3 different catalysts: zirconia alone: ZrO.sub.2, iron oxide alone: Fe.sub.2O.sub.3 and a mixture of ZrO.sub.2 and Fe.sub.2O.sub.3 in an Fe.sub.2O.sub.3/(ZrO.sub.2+Fe.sub.2O.sub.3) volume ratio of 0.5.

(70) The average residence time is 4.5 seconds. The results in terms of conversion and of nitrilation, each expressed as molar percentage, appear in the tables below.

(71) 2.1 At 200 C.

(72) The results appear in table 4 below.

(73) TABLE-US-00004 TABLE 4 Fe.sub.2O.sub.3/(ZrO.sub.2 + Fe.sub.2O.sub.3) volume ratio 0.0 0.5 1 Conversion 55.8 43.1 7.2 (mol %) Nitrilation 3.9 20.6 5.6 (mol %)

(74) It is noted that the mixture of catalysts according to the invention results in an increase for the nitrilation compared with the metal oxides used alone.

(75) 2.2 At 250 C.

(76) The results appear in table 5 below and in FIG. 2.

(77) TABLE-US-00005 TABLE 5 Fe.sub.2O.sub.3/(ZrO.sub.2 + Fe.sub.2O.sub.3) volume ratio 0.0 0.5 1 Conversion 71.1 95.7 38.7 (mol %) Nitrilation 60.2 93.4 36.9 (mol %)

(78) These results show a clear improvement in the conversion content and the nitrilation content for the catalyst according to the invention compared with the catalysts alone.

(79) Consequently, for this very short residence time, good results were observed in terms of conversion and of nitrilation with a very small difference between these two values, indicating a low by-product content.

(80) Finally, FIG. 2 provides evidence of the expected effect linked to the values of the volume ratio according to the invention.

Example 2: Test on the Methyl Ester of Undecenoic Acid

(81) Tests were carried out on the methyl ester of undecenoic acid and with the catalyst Fe.sub.2O.sub.3 on an Al.sub.2O.sub.3 support. The procedure is the same as that set out in example 1 above.

(82) Results:

(83) This experiment was carried out with 3 different catalysts: alumina alone: Al.sub.2O.sub.3, iron oxide alone: Fe.sub.2O.sub.3 and an Al.sub.2O.sub.3/Fe.sub.2O.sub.3 mixture in an Fe.sub.2O.sub.3/(Al.sub.2O.sub.3+Fe.sub.2O.sub.3) volume ratio of 0.5.

(84) The average residence time is 4.5 seconds. The results in terms of conversion and of nitrilation, each expressed as molar percentage, appear in the tables below.

(85) 2.1 At 200 C.

(86) The results appear in table 6 below.

(87) TABLE-US-00006 TABLE 6 Fe.sub.2O.sub.3/(Al.sub.2O.sub.3 + Fe.sub.2O.sub.3) volume ratio 0.0 0.5 1 Conversion 8.7 16.5 0.7 (mol %) Nitrilation 1.9 9.2 0.7 (mol %)

(88) It is noted that, even at 200 C., which is a relatively low temperature, the mixture of catalysts according to the invention results in a surprising and unexpected increase in the nitrilation.

(89) 2.2 At 250 C.

(90) The results appear in table 7 below.

(91) TABLE-US-00007 TABLE 7 Fe.sub.2O.sub.3/(Al.sub.2O.sub.3 + Fe.sub.2O.sub.3) volume ratio 0.0 0.5 1 Conversion 33.7 30.5 17.0 (mol %) Nitrilation 19.0 26.6 12.0 (mol %)

(92) These results show a clear improvement in the nitrilation content for the catalyst according to the invention compared with the catalysts alone.

(93) These tests demonstrate the unexpected and surprising effect provided by the specific catalyst of the invention which comprises a mixture of metal oxides as previously defined, in the ratio of mixtures also previously defined. The degrees of conversion observed, and also the amounts of nitriles observed, are greater than those obtained with the catalysts comprising just one metal oxide. This is all the more surprising since the contact time is very short, evaluated at a few seconds: 4.5 sec.