Hydroformylation of triglycerides in a self-emulsifying medium

Abstract

The invention relates to a method for the hydroformnylation of triglycerides by homogeneous catalysis in the presence of at least one substituted cyclodextrin, said method comprising a step a) of combining, under agitation, at least one catalyst, waiter, at least one unsaturated triglyceride and said substituted cyclodextrin, in the presence of gaseous hydrogen and carbon monoxide, said step being carried out in reactive conditions allowing the formation of an emulsion during the agitation and a decanting of the products once the agitation has stopped.

Claims

1. A process for the hydroformylation of triglycerides by homogeneous catalysis in the presence of at least one substituted cyclodextrin, the process comprising: a step a) of bringing together, with stirring, at least one catalyst, water, at least one unsaturated triglyceride and the substituted cyclodextrin, in the presence of hydrogen gas and of carbon monoxide; and wherein the step is carried out under reaction conditions allowing the formation of an emulsion during the stirring and decanting of the reaction products once the stirring has stopped.

2. The process as claimed in claim 1, wherein the substituted cyclodextrin is an , or -cyclodextrin.

3. The process as claimed in claim 1, wherein the substituted cyclodextrin is a methylated cyclodextrin.

4. The process as claimed in claim 3, wherein the methylated cyclodextrin has an average degree of substitution ranging from 0.5 to 2.

5. The process as claimed in claim 1, wherein the substituted cyclodextrin is a hydroxypropylated cyclodextrin.

6. The process as claimed in claim 5, wherein the hydroxypropylated cyclodextrin has an average degree of substitution ranging from 0.5 to 2.0.

7. The process as claimed in claim 1, that wherein step a) is carried out at a temperature ranging from 40 C. to 120 C.

8. The process as claimed in claim 1, wherein the cyclodextrin/triglyceride mole ratio is chosen in a range of from 0.2 to 4.

9. The process as claimed in claim 1, wherein step a) is carried out under pressure of a gas comprising molecular hydrogen and carbon monoxide and in that the gas pressure is greater than or equal to 80 bar.

10. The process as claimed in claim 1, wherein the catalyst is a complex comprising a transition metal combined with a sulfonated phosphine.

11. The process as claimed in claim 1, wherein the triglyceride is a mixture of triglycerides, which mixture does not comprise more than 10 mol % of polyunsaturated fatty chains in the triglycerides.

12. The process as claimed in claim 1, wherein the reaction conditions are close to, or are those of, the catalysis carried out in the presence of cyclodextrin.

13. The process as claimed in claim 1, wherein the decanting allows the separation of an organic phase containing the reaction products and that the step of recovering the reaction products is carried out by separation of the reaction medium from the phase.

14. The process as claimed in claim 4, wherein the methylated cyclodextrin has an average degree of substitution of 0.8.

15. The process as claimed in claim 6, wherein the hydroxypropylated cyclodextrin has an average degree of substitution of 0.8.

16. The process as claimed in claim 7, wherein step a) is carried out at a temperature ranging from 60 C. to 80 C.

17. The process as claimed in claim 8, wherein the cyclodextrin/triglyceride mole ratio is 2.9.

18. The process as claimed in claim 10, wherein the sulfonated phosphine is selected from the group consisting of TPPTS and a sulfonated phenyl(s)/biphenyl(s) phosphine.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention will be understood more clearly on reading the appended figures, which are provided by way of examples and are in no way limiting in nature, in which:

(2) FIG. 1: is a phase diagram at 80 C. of a water/triolein/RAME--CD system two minutes after the stirring has stopped.

(3) FIG. 2: shows phase diagrams of the water/triolein/RAME--CD system at 25 C., 80 C. and 100 C., two minutes after the stirring has stopped.

DETAILED DESCRIPTION OF THE INVENTION

Example 1

Production of the Phase Diagram for the Water/Triolein/Substituted Cyclodextrin (CD) System at t=2 Min

(4) This diagram is produced by varying the weight proportions between the components. The results are presented in FIG. 1. The proportions are calculated in the following way:

(5) $\%\left(i\right)=\frac{w_i}{w_{\mathrm{total}}}=\left(\frac{w_i}{w_{\mathrm{CD}}+w_{\mathrm{water}}+w_{\mathrm{triolein}}}\right)$
w.sub.i being the weight of the compound of which it is desired to determine the proportion.

(6) Since emulsions are temperature-dependent, the diagram was produced at a temperature of 80 C. (temperature used for the catalytic tests). The procedure used is the following:

(7) Predetermined amounts of RAME--CD and of triolein (cf. Table 1a, tubes 1 to 4) are placed in a test tube equipped with a magnetic stirrer. A certain amount of water is then added to each tube. The medium is left to stir (magnetic bar at 1400 revolutions/min) in an oil bath at 80 C. for 20 min. The stirring is then stopped. Two minutes after the stirring has stopped, the test tube is removed from the bath for observation. Once the phases have been identified, a further amount of water is added, thus changing the proportions of the various constituents.

(8) The operations described above are repeated until obtaining of the results presented in table 1b which are plotted on the phase diagram and make it possible to determine the following domains A, B, C and D:

(9) Domain A: two distinct phases, an aqueous phase and an oily phase

(10) Domain B: emulsion (oil-in-water)+an aqueous phase

(11) Domain C: phase comprising a solid in suspension

(12) Domain D: emulsion (oil-in-water)

(13) TABLE-US-00001 TABLE 1a Total weight water Weight CD Weight oil (g) (g) (g) Tube 1 0.1 0.36 Tube 2 0.3 0.46 Tube 3 0.5 0.27 Tube 4 0.7 0.14

(14) TABLE-US-00002 TABLE 1b Example 1 Total weight water (g) 0.1 0.2 0.3 0.4 0.6 0.8 1 1.4 1.8 2.2 3.6 Domains A A A A A B B B B B B C D B B B B B B B B B C C B B B B B B B B B C C D D D B B B B B B

Example 2

Reaction for Hydroformylation of a Triglyceride, Triolein, in a Self-Emulsifying System According to the Invention

(15) The reaction scheme is the following:

(16) ##STR00003##

(17) 6.2 ml of an aqueous solution containing 3.05 mmol (4 g) of partially methylated -cyclodextrins (RAME--CD) having an average molar mass of 1310 g.Math.mol.sup.1 and an average degree of substitution of 1.8 per glucoside unit, 0.029 mmol (7.5 mg) of (acetylacetonato)dicarbonyl rhodium(I) and 0.15 mmol (85 mg) of tri(meta-sulfonatophenyl)phosphine trisodium salt are placed in a 25 ml autoclave containing 1.9 mmol (1.8 ml) of triolein (i.e. 5.8 mmol of internal CC double bonds).

(18) This system corresponds to the proportions given in FIG. 1 on the phase diagram. The autoclave is heated to 80 C., pressurized under 50 bar of a CO/H.sub.2 mixture in equimolar proportion, and the stirring is regulated at 1400 revolutions/minute for 18 hours.

(19) After the stirring has stopped, an emulsified phase containing the oil (initial products and reaction products) in the water and an aqueous phase are observed. Five minutes after the stirring has stopped, the emulsified phase has completely decanted, allowing the extraction of the organic phase and the recycling of the aqueous phase containing the catalyst and the CDs.

(20) A conversion of 26% and a selectivity of 78% in favor of the aldehyde are determined by NMR of the reaction products after decanting of the latter. This conversion corresponds to a conversion of the unsaturations contained in the triolein and not a conversion of the triolein itself.

(21) Aldehyde selectivity: number of moles of aldehyde formed/number of moles of initial olefins that have been converted.

Examples 3 to 7

Influence of the Amount of Substituted CD on the Hydroformylation Reaction of Example 2

(22) These examples were carried out under the same experimental conditions in terms of time, pressure and temperature and also with the same catalytic agents as example 2. They show the influence of the amount of CD on the hydroformylation reaction.

(23) The relative positions of examples 4 to 7 on the phase diagram previously produced are given in FIG. 1. Experiment 3 is an example without cyclodextrin and is given by way of comparison. It can be shown that domain B is a zone which makes it possible to have a significant conversion and selectivity. Indeed, example 7, located in domain D, shows a decrease in the selectivity of the hydroformylation. The specific experimental conditions and also the results are collated in table 2 below.

(24) TABLE-US-00003 TABLE 2 Weight of RAME-- Conversion Aldehyde RAME--CD CD/triglycerides of double selectivity Examples (g) mole ratio bonds (%) (%) 3 0 /// 0 /// 4 0.5 0.20 5 62 5 2 0.82 19 84 6 7 2.88 39 78 7 9 3.71 35 44

Examples 8 to 12

Influence of the Temperature on the Hydroformylation Reaction of Example 2

(25) The following examples were carried out under the same experimental conditions and with the same catalytic agents as example 2. These examples show the influence of the reaction temperature on the activity of the catalytic system inherent in the invention. Since temperature influences the emulsion, several phase diagrams produced at temperatures of 25 C., 80 C. and 100 C. (at t=2 min) are represented in FIG. 2. The reaction takes place in domain B for these three temperatures. The temperature conditions and also the results are collated in table 3:

(26) TABLE-US-00004 TABLE 3 Temperature Conversion of double Aldehyde selectivity Example ( C.) bonds (%) (%) 8 25 21 75 9 40 47 90 10 60 49 86 11 100 21 48 12 120 11 28

(27) These results show a temperature application range of between 40 C. and 80 C. with an optimum temperature at 60 C. for optimal conversion and optimal selectivity. Considering that the usual hydroformylation temperature is generally in the region of 80 C., this result is unexpected and particularly advantageous since it makes it possible to carry out the reaction under milder conditions while minimizing energy expenditure.

Examples 13 and 14

Influence of the Pressure on the Hydroformylation Reaction of Example 2

(28) The following examples were carried out under experimental conditions identical to example 2. The following examples reflect the influence of the pressure of CO/H.sub.2 syngas on conversion and selectivity. The results of these examples are collated in table 4:

(29) TABLE-US-00005 TABLE 4 Conversion of Aldehyde Example Pressure (bar) double bonds (%) selectivity (%) 13 20 12 25 14 80 58 87

(30) These results show the effect of the pressure on the activity of the system inherent in the invention.

Examples 15 to 21

Comparison Between the Substituted CDs Used in the Process According to the Invention and Other Substituted CDs

(31) The following examples were carried out with various native or substituted CDs under the same experimental conditions as example 2 and after verification that these reaction conditions are those located in domain B and therefore make it possible to obtain an emulsion. They make it possible to show the influence of the cyclodextrin and of the chemical modifications made thereto on the conversion (Cony) and the aldehyde selectivity (Selec). The results are collated in table 5.

(32) TABLE-US-00006 TABLE 5 Modification Weight.sub.CD T P Conv Selec Example DS.sup.c (%).sup.d ( C.) (bar) (%) (%) Decanting 15 -CD // 5 80 50 >2 /// No 16 -CD // 7 80 50 3 /// No 17 RAME--CD.sup.a 1.8 34 80 80 58 87 Yes 18 RAME--CD 1.8 34 80 80 32 76 Yes 19 HP--CD.sup.b 0.6 34 80 80 29 81 Yes 20 HP--CD.sup.b 0.8 34 80 80 98 90 Yes 21 HP--CD.sup.b 0.6 34 80 80 95 89 Yes 22 HP--CD.sup.b 0.6 34 80 80 86 90 Yes .sup.acorresponds to a partially methylated cyclodextrin .sup.bcorresponds to a cyclodextrin partially substituted with a 2-(hydroxypropyl) group .sup.cDS: degree of substitution per glucoside unit .sup.dweight proportion of the cyclodextrin in question in the reaction medium

Examples 23 to 27

Process for Hydroformylation on a Mixture of Triglycerides of Natural Origin: Commercial Olive Oil

(33) The following examples were carried out under the same conditions as example 2. As in the previous examples, it was verified beforehand that the system is in a domain of type B by producing the phase diagrams. Decanting after reaction is observed for the commercial oil, and for the triolein. These vegetable oils are not pure, but the following results make it possible to state that the present invention is suitable for the hydroformylation of complex mixtures of triglycerides. The results are collated in table 6.

(34) TABLE-US-00007 TABLE 6 Pressure Temperature Time Conversion Selectivity Example Cyclodextrin (bar) ( C.) (hours) (%) (%) 23 RAME--CD 50 80 18 44 56 (1.8).sup.a 24 RAME--CD 80 80 18 66 70 (1.8).sup.a 25 RAME--CD 100 50 6 35 78 (1.8).sup.a 26 RAME--CD 150 50 6 58 81 (1.8).sup.a 27 HP--CD 80 80 18 78 85 (0.6).sup.a .sup.aThe number between parentheses represents the degree of substitution per glucoside unit of the cyclodextrin in question.

(35) The commercial oil used for these examples was analyzed by trans-esterification of the triglycerides in methanol and by gas chromatography in order to determine the various fatty acid derivatives making up the vegetable oil. The results are collated in table 7:

(36) TABLE-US-00008 TABLE 7 Fatty acid.sup.a Molar proportion in olive oil (%) Palmitoleic acid (C.sub.16 :1) 0.7 Palmitic acid (C.sub.16 :0) 11.1 Stearic acid (C.sub.18 :0) 3.2 Oleic acid (C.sub.18 :1) 76.7 Linoleic acid (C.sub.18 :2) 4.1 Linolenic acid (C.sub.18 :3) 1.7 Arachidic acid (C.sub.20 :0) 2.5 .sup.aThe abbreviations between parentheses make it possible to schematize the type of fatty acid. The first number represents the total number of carbon atoms of the fatty acid, the second number indicating the number of unsaturations on the carbon-based chain.

Examples 28 to 31

Influence of the Nature of the Phosphine

(37) The following examples were carried out according to the same experimental conditions as example 2, except for the nature of the phosphine used in combination with rhodium and also the CO/H.sub.2 syngas pressure, which is 80 bar. In these examples, a sulfonated monophosphine with a phosphine/metal mole ratio of 5/1 is used. The phosphines used and also the effects of the latter on the hydroformylation reaction are given in table 8:

(38) TABLE-US-00009 TABLE 8 Conversion Selectivity Example Phosphine Phosphine structure (%) (%) 28 Disulfonated tris(o-toluyl) phosphine 91 88 29 Trisulfonated biphenyldiphenyl phosphine 58 85 30 Trisulfonated bis(biphenyl)phenyl phosphine 72 86 31 Trisulfonated tris(biphenyl) phosphine 51 82

(39) The phosphines that are particularly advantageous in the process of the present invention are TPPTS and sulfonated phenyl(s)-biphenyl(s) phosphines, i.e. sulfonated phosphines comprising phenyl groups and/or biphenyl groups. The examples listed in table 8 make it possible to show the validity of the present invention for a range of sulfonated aromatic phosphines.

Example 32

Influence of the Temperature for HP--CD

(40) The presence of an optimum temperature for HP--CD is established using the procedure of example 2 under the following specific reaction conditions: % by weight of the various constituents: HP--CD: 34%, water: 54%, oil: 17%, 80 bar CO/H.sub.2, 18 hours, 1 mmol of triolein (3 mmol of double bonds), mole ratio: double bond/Rh=200; phosphine/Rh mole ratio=5. The catalyst and the phosphine are those of example 2. A phase separation clearly takes place. The results presented in table 9 were obtained using the analytical methods previously described.

(41) TABLE-US-00010 TABLE 9 Conversion of Aldehyde Temperature ( C.) double bonds (%) selectivity (%) 30 28 68 50 53 79 60 69 85 80 91 90 100 55 65 120 13 23

Example 33

Influence of the Weight Concentration of HP--CD

(42) The role of the weight concentration of HP--CD in the reaction is established using the procedure of example 2 under the following specific reaction conditions: water: 3.4 ml, pressure 80 bar CO/H.sub.2, stirring time 18 hours, 1 mmol of triolein (3 mmol of double bonds), double bond/Rh mole ratio=200, phosphine/Rh mole ratio=5. The catalyst and the phosphine are those of example 2. A phase separation clearly takes place except when the CD concentration is outside the desired phase separation domain. The results presented in table 10 were obtained using the analytical methods previously described.

(43) TABLE-US-00011 TABLE 10 % by weight of Conversion of double Aldehyde CD bonds (%) selectivity (%) Decanting 44 87 89 No 34 91 90 Yes 20 61 84 Yes 12 43 70 Yes 7 23 59 Yes

(44) These results indicate that, since phase separation conditions which allow decanting are used, the higher the amount of cyclodextrin, the higher the degrees of conversion and of selectivity.

Example 34

Hydroformylation Process Using HP--CD on Olive Oil, Validation of the Temperatures and of the Weight Concentrations of the Mixture

(45) The role of 1) the temperature, 2) the weight concentration of HP--CD and 3) the weight concentration of water (at constant HP--CD/oil ratio) is established using the procedure of example 2.

(46) 1) For the temperature, the specific reaction conditions are the following: % by weight: HP--CD: 34%, water: 54%, oil: 17%, pressure 80 bar CO/H.sub.2, stirring time 18 hours, 1 mmol of commercial olive oil, double bond/Rh mole ratio=200, phosphine/Rh mole ratio=5. The catalyst and the phosphine are those of example 2. A phase separation clearly takes place. The results presented in table 11 were obtained using the analytical methods previously described.

(47) TABLE-US-00012 TABLE 11 Temperature Conversion of double Aldehyde ( C.) bonds (%) selectivity (%) 30 28 68 50 50 79 60 55 80 80 78 85 100 48 65 120 12 23

(48) Thus, an optimum temperature of around 80 C. is established.

(49) 2) For the weight concentration of HP--CD, the specific reaction conditions are the following: water: 3.4 ml, pressure 80 bar CO/H.sub.2, stirring time 18 hours, 1 mmol of commercial olive oil, double bond/Rh mole ratio=200, phosphine/Rh mole ratio=5. The catalyst and the phosphine are those of example 2. A phase separation clearly takes place except when the CD concentration is outside the desired phase separation domain. The results presented in table 12 were obtained using the analytical methods previously described.

(50) TABLE-US-00013 TABLE 12 % by weight of Conversion of double Aldehyde CD bonds (%) selectivity (%) Decanting 44 81 76 No 34 78 85 Yes 20 51 72 Yes 12 38 70 Yes 7 23 56 Yes

(51) These results indicate that, since phase separation conditions which allow decanting are used, the higher the amount of cyclodextrin, the higher the degrees of conversion and of selectivity.

(52) 3) For the weight concentration of water (constant HP--CD/olive oil weight ratio equal to 2), the specific reaction conditions are the following: 2.3 g of HP--CD, 1.1 ml of commercial olive oil, pressure 80 bar CO/H.sub.2, stirring time 18 hours, double bond/Rh mole ratio=200, phosphine/Rh mole ratio=5. The catalyst and the phosphine are those of example 2. A phase separation clearly takes place except when the amount of water is outside the desired phase separation domain. The results presented in table 13 were obtained using the analytical methods previously described.

(53) TABLE-US-00014 TABLE 13 % by weight Conversion of double Aldehyde of water bonds (%) selectivity (%) Decanting 24 3 43 No 39 19 54 No 52 78 85 Yes 66 74 83 Yes 79 66 77 Yes 88 55 65 Yes

(54) These results indicate that, since phase separation conditions which allow decanting are used, the optimum amount of water is less than 79% by weight.

Example 35

Hydroformylation Process Using CRYSMEB on Triolein

(55) CRYSMEB is a -cyclodextrin methylated in position 2 with a low degree of substitution (DS.sub.CRYSMEB=0.8 compared with DS.sub.RAMEB=1.8). This cyclodextrin was tested beforehand under the emulsion condition. It makes it possible to obtain emulsions that are slightly more stable (slightly slower decanting) but all the same allow decanting of the oil after a resting time without stirring of about a few minutes.

(56) The presence of an optimum temperature for CRYSMEB is established using the procedure of example 2 under the following specific reaction conditions: % by weight of the various constituents: CRYSMEB 20%, water: 71%, oil: 9% (1 mmol of triolein (3 mmol of double bonds)), pressure 80 bar CO/H.sub.2, stirring time 6 hours, mole ratio: double bond/Rh=200; phosphine/Rh mole ratio=5. The catalyst and the phosphine are those of example 2. A phase separation clearly takes place. The results presented in table 14 were obtained using the analytical methods previously described.

(57) TABLE-US-00015 TABLE 14 Conversion of double Aldehyde Temperature bonds (%) selectivity (%) 40 69 87 60 96 88 80 96 94 100 74 84 120 36 70

(58) Thus, an optimum temperature in the region of 60 to 80 C. is established.

(59) The role of the weight concentration of CRYSMEB in the reaction is established using the procedure of example 2 under the following specific reaction conditions: water: 8.0 ml, temperature: 80 C., pressure 80 bar CO/H.sub.2, stirring time 6 hours, 1 mmol of triolein (3 mmol of double bonds), double bonds/Rh=200, phosphine/Rh=5. The catalyst and the phosphine are those of example 2. A phase separation clearly takes place. The catalyst and the phosphine are those of example 2. A phase separation clearly takes place except when the CD concentration is outside the desired phase separation domain. The results presented in table 15 were obtained using the analytical methods previously described.

(60) TABLE-US-00016 TABLE 15 % by weight Conversion of double Aldehyde of CD bonds (%) selectivity (%) Decanting 25 94 94 No 20 96 94 Yes 11 79 97 Yes 6 61 85 Yes 3 41 82 Yes 0 <1 /// Yes

(61) The study according to the weight proportion of CD shows that the more the amount of CD increases in the medium, the more the conversion and the selectivity increase. The increase in the CD concentration would allow the formation of more inclusion complexes by shifting the equilibrium in favor of the formation of the latter. Since the number of surfactant complexes increases, the emulsion that results therefrom would be more stable, and more homogeneous with a maximum interface. However, it should be noted that, beyond a certain weight of CD, the conversion no longer changes and the selectivity begins to decrease. Furthermore, the emulsion increasingly stabilizes with the proportion of CD, which is of course to be avoided.

Example 36

Process for the Hydroformylation of Sunflower Oil

(62) The process is carried out using sunflower oil and three cyclodextrins previously described. The experimental protocol is the same as that of example 2, under the following specific conditions:

(63) [a] Conditions: CD (2.3 g), oil (1 ml, 1 mmol), Rh((CO)).sub.2(acac) (3.9 mg, 0.015 mmol), TPPTS (42 mg, 0.075 mmol), water (5 ml), CO/H.sub.2 pressure=80 bar, temperature=80 C., stirring time 18 h.

(64) [b] Conditions: CD (2.3 g), oil (1 ml, 1 mmol), Rh((CO)).sub.2(acac) (3.9 mg, 0.015 mmol), TPPTS (42 mg, 0.075 mmol), water (8 ml), CO/H.sub.2 pressure=80 bar, temperature=80 C., stirring time 6 h.

(65) The results obtained were compiled with those previously obtained for olive oil, for the purposes of comparison, in table 16:

(66) TABLE-US-00017 % of Conversion Initial % of isomerization Number of of CC Aldehyde polyun- of the poly- initial CC double bonds selectivity saturate in unsaturated Entry CD Oil double bonds (%) (%) the oil chains 1.sup.a RAME--CD Olive 2.78 48 71 5.8 2.sup.a HP--CD Olive 2.78 78 85 5.8 3.sup.a HP--CD Sunflower 2.75 51 80 7.7 39 4.sup.b CRYSMEB Olive 2.78 86 86 5.8 5.sup.b CRYSMEB Sunflower 2.75 54 80 7.7 52

(67) Number of initial CC double bonds: average number of CC double bonds initially present in the triglycerides making up the oil comprising all types of unsaturations (monounsaturated+polyunsaturated). Number of unsaturations of the plant oils determined by .sup.1H NMR.

(68) Initial % of polyunsaturates in the oil: mole percentage of fatty esters having at least two double bonds on the carbon-based chain (linoleic, linolenic derivatives) in the oil in question. Percentage obtained after methanolic transesterification of the triglycerides and GC and GC-MS analysis of the methyl esters obtained.

(69) Percentage of isomerization of the polyunsaturated chains: number of moles of conjugated isomer formed/number of moles of initial polyunsaturated chains. It should be noted that only the polyunsaturated derivatives can be isomerized.

(70) The invention is not limited to the embodiments presented, and other embodiments will clearly become apparent to those skilled in the art.