Mesoporous basic catalysts: comparison with alkaline exchange zeolites (basicity and porosity). Application to the selective etherification of glycerol to polyglycerols.

Mesoporous basic catalysts: comparison with alkaline exchange zeolites (basicity and porosity). Application to the selective etherification of glycerol to polyglycerols.

91998 Elsevier Science B.V. All rights reserved. Preparation of Catalysts VII B. Delmon et al., editors. 895 M e s o p o r o u s basic catalysts: co...

397KB Sizes 0 Downloads 20 Views

91998 Elsevier Science B.V. All rights reserved. Preparation of Catalysts VII B. Delmon et al., editors.

895

M e s o p o r o u s basic catalysts: comparison with alkaline exchange zeolites (basicity and porosity). Application to the selective etherification of glycerol to polyglycerols. J-M. Clacens a, Y. Pouilloux a, J. Barrault a, C. Linares b and M. Goldwasserb ~Laboratoire de Catalyse en Chimie Organique. UMR 6503 ESIP, 40, avenue du recteur Pineau, 86022 Poitiers Cedex, France Tel.: +33/(0)5.49.45.40.24- Fax.: +33/(0)5.49.45.33.49 E-mail : [email protected] bFacultad de ciencias, escuela de quimica Apartado de correo 47102, Caracas, 1041-A, Los Chaguaramos, Caracas, Venezuela

The glycerol condensation (etherification) was studied in presence of alkaline exchange zeolites and mesoporous basic catalysts. It was observed that the diglycerol selectivity was increased when X zeolites were exchanged with cesium while the selectivity of Cs - ZSM5 was comparable to that of Na2CO3. Over mesoporous M La or M Mn materials, the formation of triglycerol and of tetraglycerol was more important. Such results could be due to changes of both the pore size and the basicity of the catalysts. 1. INTRODUCTION Glycerol is mainly a natural product issued from the methanolysis of vegetable oils. In Europe, due to the increasing use of methyl esters as fuel additives, one can expect an increase of glycerol production which could become a cheaper raw material for chemistry (1). For example, polyglycerols and specially polyglycerols-esters (PGEs) are gaining prominence in new products for tensioactive, lubricants, cosmetics, foods additives, .... Indeed PGEs exhibit multifunctional properties and a wide range of formulating options, if it is possible to control; (i) the length of the polyglycerols chain, (ii) the degree of esterification and (iii) the fatty acid molecular weight. These reactions are quite interesting goals for shape - selective catalytic processes. Previous works showed that the selectivity of the first step is not really controlled and that a mixture of di- to hexa-glycerol (linear or cyclic) is obtained (2). Then it is rather difficult to get a well-defined product and to predict the Hydrophilic Lipophilic Balance (HLB) after esterification. In our laboratory it was evidenced that the esterification of glycerol could be selective to ct monoglycerides over cationic resins (3). Nevertheless polyglycerols and

896 polyglycerols esters as well as acroleine were obtained as main by-products. Moreover we observed that the modification of the pseudo-pore size of these materials improved the selectivity to (PG + PGEs) but acroleine is always obtained over these acid catalysts. In order to obtain a selective formation of di-, tri-, tetra- or poly-glycerols, we investigated first the selective etherification of glycerol (figure 1) over solid bases and compared the results to those obtained with Na2CO3 catalysts. A first series of solid catalysts used for this reaction were mesoporous type materials (MCM-41) (4) or Metal-MCM-41. We also reported results obtained with microporous alkali exchanged zeolites and alkali supported over alumina. Moreover, the pore size was controlled in using different "templating systems" so as to know if there could exist a relationship between the polymerization degree and the size of the pores of the catalyst (5).

2

OH

HQ,v.~OH Glycerol

OH

OH

HO.v.,~ o V ~ o H

Diglycerol

OH

l

OH

+

OH

H20

H20

Triglycerol

Poly[glycerols Figure 1. Etherification of glycerol 2. EXPERIMENTAL PART

2.1. Catalyst preparation Cesium-X zeolites are obtained from exchange experiment (CsX{i)) and exchangeimpregnation (CsX0,a)) at 60~ with a 0.4 M solution of cesium acetate (0.02 g of zeolite / mL of solution) so as to get Cs20 clusters close to exchange cesium ions. We made this exchange twice during 24 hours with a 13X zeolite coming from Prolabo; then the mixture is washed (CsX0)) or not (CsX(~a)) with distilled water and dried at 80~ for 12 hours. The resulting solid is calcined under air at 550~ for 5 hours at a heating rate of 1~ / min. Cesium ZSM-5 zeolites are obtained by exchanging two different H-ZSM-5 zeolites (Degussa) with Si / AI ratios of respectively 28 (CsZSM5(28)) and 1000 (CsZSM5(1000)); with a 0.4 M solution of cesium nitrate at 80~ for 24 hours (0.02 g of zeolite / mL of solution). Then the zeolite is dried at 80~ for 12 hours and calcined under air at 550~ for 5 hours at a heating rate of 2~ / min. Mesoporous (M) and promoted mesoporous (M Me(n), where n -- Si / Me ratio) materials were prepared according to a procedure developed in our laboratory (we add dropwise 0.2

897 mole of a solution of sodium silicate (27% SiO2) to a solution containing 0.022 mole of NaOH, 23.31 moles of 1-I20, 0.005 mole of the metal nitrate salt and 0.043 mole of template (cetyltrimethylamonium bromide). The pH is adjusted at 10.5 with diluted HC1; then the formed gel is placed in a autoclave at 100~ for 24 hours. The resulting solid is filtered, washed with water, dried at 100~ overnight and calcined under air at 550~ overnight at a heating rate of l~ / min. K/Alumina was prepared according to the following method : we add dropwise to 3 g of alumina (with particle size not greater than 0.15 mm), a solution of potassium acetate in order to have 1.5 mmoles of K / g of alumina. The catalyst is first dried under air at ambient temperature for 12 h, dried under air at 100~ for 12 h and finally calcined under air at 500~ for 12 h with an air flow of 30 mL / h.g, and a heating rate of 5~ / min.

2.2. Characterization Characterization of the solids is done using X-ray diffraction, surface BET and thermodesorption analysis. Basicity was deduced from carbon dioxide adsorptionthermodesorption pulsed experiments done using the following procedures, (i) for the modified zeolites, the calcined materials were first dried at 100~ for 8 hours under air, then they were activated at 500~ for 5 hours and cooled to 20~ under a helium flow. The activated materials were then saturated with dry gaseous carbon dioxide at 20~ Physisorbed carbon dioxide was removed by purging the sample under a helium flow at 20~ until a stable baseline was monitored. The TPD was performed by heating the sample from 20 to 500~ with a heating rate of 10~ / min., (ii) for the mesoporous materials, the calcined materials were first activated at 550~ for 8 hours and cooled to 20~ under a helium flow. The activated materials were then saturated with dry gaseous carbon dioxide at 20~ Physisorbed carbon dioxide was removed by purging the sample under a helium flow at 20~ until a stable baseline was monitored. The TPD was performed by heating the sample from 20 to 550~ with a heating rate of 10~ / min. In some cases the quantification of CO2 adsorbed over the solids was quite difficult due to a low or a strong CO2 adsorption. Then we only reported in the table 2 the ratio S sites / W sites, where strong (S) and weak (W) sites correspond respectively to CO2 desorption at 250~ and 100~

2.3. Catalytic testing Glycerol etherification was carried out in a glass-reactor equipped with a mechanical stirrer in the presence of 2 wt. of catalyst, water formed during the reaction being eliminated and collected using a dean-Stark system. The reaction was performed at 240~ or 260~ for 8 or/and 24 hours with 50 g of glycerol under nitrogen. Reagents and products were analyzed after a silylation procedure (6) with a GPC equipped with an on-column injector, a FID and a polar column (HT5) supplied by SGE (L = 25 m, ID = 0.22 mm, thickness of the film = 0.10 txm).The molar percentage of each compound was determined by using standardization methods with methyl laurate as an internal reference.

898 3. RESULTS AND DISCUSSION The results reported in the table 1 and figures 2-5 show that Na2CO 3 is the more active and the less selective catalyst; the distribution of polyglycerols being rather large. Cs exchanged zeolites are quite selective to diglycerol and triglycerol while exchanged ZSM-5 are less active and selective. This is the result of a change of the size of the channels for ZSM-5 solids whatever their Si/A1 ratio. In that particular case, the reaction could proceed on the surface of zeolite particles rather than inside the channels. Over M mesoporous materials, the reaction rate of glycerol was slow but rather similar to that of Cs exchanged ZSM-5(28). When these solids were promoted with basic elements (Mg or A1), rather than with acidic ones (A1), one observed an important increase of the reaction rate without change of the selectivity (diglycerol + triglycerol). But the comparison of the results for a glycerol isoconversion shows that : (i) the formation of triglycerol is more important over promoted M catalysts specially over the M La (20) sample, (ii) the formation of tetraglycerol is increased over Mn or La promoted M samples. Table 1 Catalytic results Catalyst (Si/Me ratio)

8 hours selectivity (%) (24 hours) 8 hours conversion (%) Diglycerol Triglycerol Tetraglycerol (24 hours)

Na2CO3 K/AI203 CsX(i) CsX(i,a) CsZSM5(28) CsZSM5(1000) M M AI (20) M Mg (20) M La (20) M Mn (20)

96(-) 45(65) 36(82) 39(68) 16(21) 18(43) 12(31) 9(21) 15(65) 39(92) 12(91)

24(-) 81(59) 80(55) 85(68) 87(85) 92(79) 86(86) 74(91) 90(63) 82(26) 80(29)

35(-) 18(29) 18(29) 15(24) 13(15) 8(21) 14(14) 26(9) 10(27) 18(23) 14(27)

22(-) 1(10) 2(12) 0(7) 0(0) 0(0) 0(0) 0(0) 0(9) 0(27) 6(27)

others 29(-) 0(2) 0(4) 0(1) 0(0) 0(0) 0(0) 0(0) 0(1) 0(24) 0(17)

899 100 90 80

v

r-

70

-e-Na2CO3

60

-.e- K/AI203 --k- CsX(i)

Or) L_

5o

+

CsX(i,a)

> cO ~

4o

+

CsZSM5(28)

3o

--t-- CsZSM5(1000)

2O

10

0

0

5

10

15

20

25

Time (h) Figure 2. Conversion of glycerol over modified zeolites, K/A1203 and Na2CO3.

100

Diglycerol

70

--e- Na2CO3 - e - K/AI203

~o

.m

>

0

(l)

(!) or)

--A- CsX(i)

5o

--e- CsX(i,a)

4o

- ~ - CsZSM5(28)

Triglycerol

3O

0

0

1'0

2'0

30

- - ~ CsZSMS(1000)

40

5'0

6'0

Conversion (%)

7C)

8'0

9'0

100

Figure 3. Glycerol etherification over modified zeolites, K/A1203 and Na2CO3. Relationship between selectivity and conversion.

900

Na2CO3

7O v

60

"L~

50

cO

.-i-M - e - MAI(20) --&- MMg(20)

(!) E O r,J

40

MMn(20)

30

- 4 - MLa(20)

~'

5

10

Time (h)

1'5

20

25

Figure 4. Conversion of glycerol over modified mesoporous materials and Na2CO3.

100

=-.9, v

Diglycerol

7O

-,-Na2CO3 -a-M

60

- e - MAI(20)

5O

"6 1~ (D

- i - MMg(20)

4o

--4- MLa(20) 30

Triglycerol

0

10

20

30

--)K--MMn(20)

40

50

60

Conversion (%)

70

80

90

100

Figure 5. Glycerol etherification over modified mesoporous materials and Na,2CO 3. Relationship between selectivity and conversion.

901 The results reported in the table 2 show that basic sites are stronger with the cesium exchange X zeolite than with the potassium over alumina. We observed no basicity for the mesoporous promoted samples probably due to the too low CO2 adsorption temperature used during the TPD that would not allow the activation of the basic sites.

Table 2 Characterization results Catalyst (Si/Me ratio)

BET surface (m2/g)

Basic sites ratio (%) (strong / weak)

K/A1203 CsX(i) CsX(i,a) CsZSM5(28) CsZSM5(1000) M M A1 (20) M M s (20) M La (20) M Mn (20)

205 352 338 251 264 960 923 789 736 972

20 / 80 62 / 38 56 / 44

0/0 0/ 0 0/ 0 8/ 0 0/0

X-ray diffraction characterization of the samples was also done and confirm the structure of the exchange zeolites as well as those of the mesoporous promoted materials. BET analysis don't show unexpected results except perhaps a quite lower crystallization level for the mesoporous materials promoted with Mg and La. 4. CONCLUSION The results presented in this paper show that some mesoporous materials containing a alkaline earth, a transition element or a rare earth (i.e. La) are as active as zeolites materials exchanged with alkaline. Acid solids such M and M A1 have a much lower activity. M La (20) solids are among the most active for the glycerol etherification. The product distribution obtained with all these catalysts seem rather similar, nevertheless the change of the heteroelement could modify the selectivity as for M La or M Mn samples. Up to now we have no clear interpretation of these results in relation with the first characterizations and work is in progress.

Acknowledgments : The authors gratefully acknowledged support from the European Community "FAIR program".

902 REFERENCES

1. A.J. Kaufman and R.J. Ruebush, Proceedings of the world conference on oleochemicals into 21 st century, T-H. Applewhite Ed., American Oil Chemist Society (1991) 10. 2. K. Cottin, DEPSUP, Poitiers, 1996. 3. S. Abro, Y. Pouilloux and J. Barrault, 4th Symposium on Heterogeneous Catalysis and Fine Chemicals, Basel, 1996. 4. J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonovicz, C.T. Kresge, K.D Schmitt, C.T.W. Chu, D.H. Olson, E.W. Sheppard, S.B. McCuller, J.B. Higgins, J.L. Schlenker, J. Amer. Chem. Soc., 114 (1992) 10834. 5. To be published 6. M.R. Sahasrabudhe, J.A.O.C.S., 44 (1996) 376.