Info

100 20 0

120 36 50

samples. Table 15.3 also shows the nature of the surface acid sites as determined from IR spectra obtained after adsorption of pyridine at room temperature and evacuation at 423 K. As it is well known, Cs-HPA contains only Bronsted sites [18]. Sample ZnO/SiO2 contained only Lewis sites, while the relative concentration of Bronsted and Lewis acid sites (L/B) on Al-MCM-41 and H-BEA were 1.8 and 1.2, respectively. Results of Table 15.2 and Table 15.3 reveal that zeolite H-BEA contains a higher density of stronger acid sites compared with A-MCM-41.

15.3.3 Catalytic Results

Figure 15.5 shows that the selective formation of menthol from citral requires bifunctional metal/acid catalysts with the ability of not only promoting coupled hydrogenation/isomerization reactions of the citral-to-menthols pathway, but also minimizing the parallel hydrogenation reactions of (a) citral to nerol/geraniol or 3,7-dimethyl-2,3-octenal and (b) citronellal to citronelol or 3,7-dimethyloctanal. In other words, from a kinetic point of view:

The formation rate of citronellal from citral must be much higher than the hydrogenation rates of citral to nerol/geraniol and 3,7-dimethyl-2,3-octenal

The formation rate of pulegols from citronellal must be much higher than the hydrogenation rates of citronellal to citronelol and 3,7-dimethyloctanal

Consequently, the individual steps involved in the reaction pathway leading to menthols from citral were studied separately to select the metallic and acid functions of the bifunctional catalyst.

The monometallic catalysts of Table 15.2 were tested for the liquid-phase hydrogenation of citral (T = 393 K, P = 1013 kPa, Wcat = 1 g ) using isopropanol as solvent. Pd/SiO2 and Ni/SiO2 selectively hydrogenated the conjugated C=C bond of the citral molecule, initially giving more than 90% selectivity to citronellal (Table 15.2). This result showed that, on both catalysts, the citral hydrogenation to citronel-lal is clearly favored compared with parallel hydrogenations leading to nerol/geraniol and 3,7-dimethyl-2,3-octenal. The citronellal selectivity then decreased with reaction time because citronellal is, in turn, hydrogenated to citronelol or 3,7-dimethyloctanal. In contrast, the maximum selectivity to citronellal was never higher than 60% on the other catalysts, which formed significant amounts of nerol/geraniol isomers. Overall, the results in Table 15.2 are consistent with previous works on citral hydrogenation showing that Ni and Pd favor C=C bond hydrogenation [19, 20], while Co and Ir are more selective for C=O bond hydrogenation [20, 21].

The solid acids of Table 15.3 were tested in the liquid-phase isomerization of citronellal to isopulegols (T = 343 K, PN2 = 506.5 kPa, W = 0.200 g, citronellal:tol-uene (ml) = 2:150) using toluene as solvent. Isopulegol yields are shown in Figure 15.6 as a function of Wt / nClal , where W is the catalyst weight, t the reaction time, and n°cu the initial moles of citronellal. The local slope of each product in Figure 15.6 gives its rate of formation at a specific value of reactant conversion and contact time. In all cases, pulegol isomers were the only products detected.

Figure 15.6 shows that the citronellal cyclization rate was clearly higher on zeolite Beta and Al-MCM-41 as compared with both ZnO/SiO2 and Cs-HPA. The exact nature of the surface-active sites required for efficiently catalyzing the cycliza-tion of citronellal to isopulegols is still debated. While several authors [22, 23] reported that the reaction is readily catalyzed on Lewis acids, others [24] correlated the cyclization activity on acid zeolites with accessible Bronsted acid sites. Chuah et al. [25] found that catalytic materials containing strong Lewis and weak Bronsted acidity show good activity and selectivity for cyclization of citronellal to isopulegol. The solid acids listed in Table 15.3 contain either Lewis (ZnO/SiO2), Bronsted (Cs-HPA), or both Lewis and Bronsted acid sites (Al-MCM-41 and H-BEA). The superior activity showed by zeolite Beta and Al-MCM-41 samples for the formation of isopulegols are consistent with the assumption that Lewis/weak Bronsted dual sites are required to efficiently catalyze the citronellal cyclization.

Based on the above results, three bifunctional catalysts containing one of the metals most selective for hydrogenating citral to citronellal (Pd or Ni) and one of the solid acids more active for converting citronellal to isopulegols (zeolite Beta or Al-MCM-41) were prepared: Pd(1%)/H-BEA, Ni(3%)/H-BEA, and Ni(3%)/Al-MCM-41. These bifunctional catalysts were tested for the conversion of citral to

FIGURE 15.6 Cyclization of citronellal to isopulegols: isopulegol yield as a function of parameter Wt / nClal (343 K, 506.5 kPa nitrogen, W = 0.200 g, citronellal:toluene = 2:150 [ml]).

FIGURE 15.6 Cyclization of citronellal to isopulegols: isopulegol yield as a function of parameter Wt / nClal (343 K, 506.5 kPa nitrogen, W = 0.200 g, citronellal:toluene = 2:150 [ml]).

menthols (T = 343 K, PT = 506.5 kPa, W = 1 g) using toluene as solvent. In all cases, citral and citronellal were totally converted after 5 h of reaction. Figure 15.7 shows the evolution of the yield of total menthols as a function of reaction time. It is observed that the menthol yield was only about 20% on Pd/H-BEA catalyst at the end of the catalytic test. This poor selectivity for menthol synthesis reflected the high activity of Pd/H-BEA to hydrogenate the C=C bond of citronellal, thereby forming considerable amounts of 3,7-dimethyloctanal. Pd/H-BEA also formed significant amounts of undesirable products (35%) via secondary decarbonylation and hydrogenolysis reactions. In contrast, Figure 15.7 shows that the menthol selectivity on Ni/H-BEA was 81% at the end of the catalytic test. None of the by-products formed from hydrogenation of citral or citronellal was detected on Ni/H-BEA, suggesting that this bifunctional catalyst satisfactorily combines the hydrogenation and isomerization functions needed to selectively promote the reaction pathway leading from citral to menthols. However, formation of secondary compounds, formed probably via decarboxylation and cracking reactions on the strong acid sites of zeolite Beta, was significant. The best catalyst was Ni/Al-MCM-41, which yielded ca. 90% menthol. The observed yield improvement for menthol on Ni/Al-MCM-41 is explained by considering that the moderate acid sites of Al-MCM-41 do not promote the formation of by-products via side cracking reactions. Table 15.4 shows the distribution of menthol isomers obtained at the end of catalytic runs. (±)-Neoisomenthol was never detected in the products. On Ni-based catalysts, the menthol mixture was composed of 70 to 73% of (±)-menthols, 15 to 20% of (±)-neomenthol, and 5 to 10% of (±)-isomenthol. On Pd/Beta the racemic (±)-menthol mixture represented only about 50% of total menthols.

FIGURE 15.7 Menthol synthesis from citral: total menthol yields as a function of time (343 K, 506.5 kPa total pressure, W = 1 g, citral:toluene = 2:150 [ml]).

Time (min)

FIGURE 15.7 Menthol synthesis from citral: total menthol yields as a function of time (343 K, 506.5 kPa total pressure, W = 1 g, citral:toluene = 2:150 [ml]).

TABLE 15.4

Menthol Synthesis from Citral: Menthol Isomer Distribution

Menthol Isomer Distribution (%) IR of Pyridine

Catalyst Pressure (kPa) (±)-Menthols (±)-Neomenthol (±)-Isomenthol

Ni/Al-MCM-41 2026.0 71.1 20.0 8.9 Note: 343 K, W = 1 g, citral:toluene = 2:150 (ml).

Finally, an additional test was performed on Ni/Al-MCM-41 by increasing the hydrogen pressure to 2026 kPa. Results are presented in Figure 15.8 and Table 15.4. Figure 15.8 shows the evolution of product yields and citral conversion as a function of time. Citral was totally converted to citronellal on metallic Ni crystallites, but the concentration of citronellal remained very low because it was readily converted to pulegols on acid sites of mesoporous Al-MCM-41 support. Pulegols were then totally hydrogenated to menthols on metal Ni surface sites. The menthols yield reached 94% at the end of the test, showing the beneficial effect of increasing PH2, probably because it diminishes the formation of undesirable products via secondary reactions. In contrast, menthol isomer distribution was not changed by increasing the hydrogen pressure (Table 15.4).

In summary, the results presented here show that the liquid-phase synthesis of menthols from citral was successfully achieved using proper bifunctional catalysts.

100 200 Time (min)

FIGURE 15.8 Menthol synthesis from citral on Ni(3%)/Al-MCM-41. Product yields and citral conversion as a function of time. Isopulegols (O), (±)-menthol (□), (±)-neomenthol (▲), (±)-isomenthol (0), others (□) (343 K, 2026 kPa total pressure, W = 1 g, citral:toluene = 2:150 [ml]).

100 200 Time (min)

FIGURE 15.8 Menthol synthesis from citral on Ni(3%)/Al-MCM-41. Product yields and citral conversion as a function of time. Isopulegols (O), (±)-menthol (□), (±)-neomenthol (▲), (±)-isomenthol (0), others (□) (343 K, 2026 kPa total pressure, W = 1 g, citral:toluene = 2:150 [ml]).

In fact, at PH2 = 2026 kPa, Ni(3%)/Al-MCM-41 yields 94% menthols directly from citral and produces about 72% of racemic (±)-menthol into the menthol mixture.

15.4 BIODIESEL FROM LOW-VALUE RAW MATERIALS 15.4.1 Introduction

Interest in the use of renewable resources for fuel and feedstocks has increased lately due to the rising price of crude oil and concerns over production of carbon dioxide. Biodiesel is a nontoxic and biodegradable renewable fuel obtained from vegetable oils or animal fats that possesses physical and fuel properties similar to those of conventional oil-derived diesel fuel [26]. However, biodiesel has important advantages compared with petroleum diesel. For example, it is oxygenated, contains no sulfur, reduces unburnt and particulate matter in the exhaust, and does not cause a net increase of carbon dioxide in the atmosphere because it contains photosynthetic organic carbon [27, 28]. Commercial biodiesel is produced from renewable resources including rapeseed, sunflower, palm, or soybean oils, which are essentially edible in nature and contain almost 90 to 95% of fatty acid triglycerides [29]. These are converted to biodiesel by transesterification with short-chain alcohols, typically methanol or ethanol, using homogeneous basic catalysis [30].

For economic reasons, biodiesel has not been produced in Argentina on a large scale since December 2001. However, there are small-scale commercial units that produce biodiesel from low-value sources such as recycled frying oil, acid tallow, and brown greases. These inedible materials are usually very acid and contain free fatty acids that, in presence of a base and water, form soaps. Formation of soaps hampers the use of conventional base-catalyzed transesterification processes for efficient production of biodiesel. On the other hand, acid catalysts transform fatty acids and triglycerides into methyl esters [31], which react very slowly with methanol in acid media. Consequently, a process based only in acid catalysis cannot be carried out at a commercial scale. Thus, technological processes for obtaining biodiesel require several transforming stages when starting from very acid sources [32]. In Argentina, the research group directed by Querini (INCAPE, Santa Fe) has explored the use of a two-step process that consecutively employs acid and basic catalysts to efficiently obtain biodiesel from low-value acid raw materials [33]. The main aspects of these studies related to the processes developed for Argentine companies that manufacture biodiesel are presented.

15.4.2 Materials and Methods

Different raw materials of high acidity (A) were employed: beef tallow (A = 17), oil of Paraguayan coconut (A = 12), recycled vegetable oil (A = 35), and brown fat (A = 55). The acidity represents the amount of free acids, expressed as grams of oleic acid per 100 grams of material, and was determined by titration, dissolving the raw material in a mixture of toluene/ethanol and using NaOH and phenolphtha-lein as indicator. The amount of water in the samples was determined by the method of Dean Stark (ASTM D95). The biodiesel properties — total and free glycerin content, viscosity, flash point, pour point, cloud point, and sulfur and methanol contents — were determined following ASTM 6515 methods of analysis. The fatty acid ester composition was determined by gas chromatography.

The transesterification reaction for obtaining biodiesel from raw materials with acidity higher than 3 involved the following steps:

Reaction with methanol and acid catalysis, using sulfuric acid

Product separation

Reaction with methanol and alkaline catalysis

Separation and purification of biodiesel

Reactions were carried out in a batch reactor with reflux at temperatures between 333 and 343 K. In general, the reaction time was long enough to reach equilibrium conversions.

15.4.3 Catalytic Results

Biodiesel is an alkyl ester (R-COO-CH3) mixture that is obtained by transesterifi-cation of triglycerides and free fatty acids (R-COOH) with short-chain alcohols such as methanol. The reaction scheme involved when the process uses methanol and is catalyzed by liquid bases is:

Triglyceride + 3 CH3OH c NaOH > Glycerine + 3 R-COO-CH3 (15.3)

Triglyceride + H2O < NaOH > Glycerine + R-COO-Na

Strongly acidic materials contain a high concentration of free fatty acids that are transformed to soaps (R-COO-Na) and water via Reaction 15.5. Water formed in Reaction 15.5 promotes in turn the production of additional soaps via Reaction 15.4, which consumes the catalyst and reduces catalyst efficiency. Thus, for materials of high acidity, the scheme of Reactions 15.3 to 15.5 leads to low biodiesel yields. Besides, the presence of soap hinders biodiesel separation from the glycerine fraction [34]. Conversion of fatty acids via Reaction 15.5 can be avoided by pretreating the raw material with a strong acid, according to the following reaction scheme:

R-COOH + CH3OH c H2S°4 > R-COO-CH3 + H2O (15.6)

Triglyceride + CH3OH < H2S°4 > R-COO-CH3 + glycerine (15.7)

Acid-catalyzed Reactions 15.6 and 15.7 convert fatty acids to biodiesel and water. Then, two phases are separated: a triglyceride-rich phase, and a methanol-rich phase containing most of the water formed through Reaction 15.6. Finally, the triglyceride-rich phase is transformed to biodiesel using basic catalysis according to Reaction 15.3.

To obtain high biodiesel yields when using strongly acid raw materials, it is required to efficiently convert the free fatty acids in a first step according to Reactions 15.6 and 15.7. Results obtained by Querini's research group [33] on this initial acid-catalyzed process for reducing the free fatty acids content for maximum biodiesel production are presented below.

The equilibrium constant (K) of fatty acid esterification is calculated using Equation 15.8:

where A0 and Af represent the initial and final acidity, respectively; k (p, PM) groups properties of the system; and VMeOH and Vtrig are the initial volume of methanol and triglyceride in the reaction, respectively.

Figure 15.9 shows the equilibrium constant values of the esterification of fatty acids obtained for several raw materials. Fatty acids of raw materials of Figure 15.9, excepting coconut, are all formed mainly by C18, followed by C16, and in smaller proportion by C14 and C20. The typical chain lengths of these fatty acids are similar then, and as a consequence their equilibrium constants are approximately the same. In contrast, the fatty acid distribution of coconut oil exhibits a predominant proportion of C12, followed by C10 and C8, which results in a smaller carbon chain length as compared with the other materials of Figure 15.9 and thus

Acidity

FIGURE 15.9 Esterification equilibrium constants determined for raw materials of different acidity.

Acidity

FIGURE 15.9 Esterification equilibrium constants determined for raw materials of different acidity.

a lower esterification equilibrium constant. Determination of K values is useful for selecting the methanol/raw-material ratio needed in each individual acid-catalyzed stage in order to diminish the raw material acidity to the values required for the final base-catalyzed step.

Figure 15.10 shows the evolution of the acidity as a function of time for a sample obtained by breakage of a vegetable oil emulsion. Figure 15.10 shows two different experiences carried out under the same experimental conditions using methanol (60%v) and sulfuric acid (0.17% H2SO4). Similar A vs. t curves were obtained, thereby indicating that the acid-catalyzed reaction can be satisfactorily reproduced.

Figure 15.11 shows the effect of H2SO4 concentration on the acidity decrease rate when using a chicken fat sample as raw material; sulfuric acid concentrations of 0.14 and 0.21% were used. The initial acidity of the sample (A0 = 65) dropped

FIGURE 15.10 Evolution of acidity as a function of time for a degumming residue of sunflower oil (T = 333 K, 0.17% H2SO4, CH3OH/raw material = 6).

Time (min)

FIGURE 15.10 Evolution of acidity as a function of time for a degumming residue of sunflower oil (T = 333 K, 0.17% H2SO4, CH3OH/raw material = 6).

Time, min

FIGURE 15.11 Effect of H2SO4 concentration on the acidity decrease rate (chicken fat, T = 333 K, CH3OH/raw material = 6).

Time, min

FIGURE 15.11 Effect of H2SO4 concentration on the acidity decrease rate (chicken fat, T = 333 K, CH3OH/raw material = 6).

to about 50 after dilution with methanol. As expected, the activity decline was faster when using higher acid concentrations, but similar A values were obtained at the end of the catalytic runs, suggesting that the equilibrium was reached for both cases. After phase separation, a second acid-catalyzed step was carried out using the same acid concentrations as in the initial step. It is observed in Figure 15.11 that at the end of this second reaction stage the acidity of the triglyceride-rich phase was below 2, thereby indicating that this phase can now be efficiently converted to biodiesel via a final base-catalyzed step.

Esterification of fatty acids with ethanol 96% in the presence of sulfuric acid was also investigated. The use of ethanol 96%, less expensive than absolute ethanol, is an attractive alternative for the acid-catalyzed reaction because the presence of water does not promote the formation of soap, as explained above. Figure 15.12 shows results obtained when a beef tallow sample (A = 4.9) was reacted in acid media with ethanol 99.5% and 96%, respectively. It is observed that quasi-stable acidity values are reached in both cases after 60 min, suggesting that fatty acids conversion is limited by reaction equilibrium. Because quasi-equilibrium conditions are reached, the final A value was higher when using ethanol 99.5%, as predicted by Reaction 15.4. But it should be noted that even with ethanol 96%, the acidity of a raw material can be diminished by transforming free fatty acids into biodiesel.

In summary, low-value raw materials of high acidity can be efficiently transformed to biodiesel via a two-step reaction process involving a first acid-catalyzed stage for converting free fatty acids and a final base-catalyzed for converting triglycerides into biodiesel. Design of the initial acid-catalyzed process is based on

FIGURE 15.12 Esterification with ethanol: effect of the water content (beef tallow, 0.21% H2SO4, T = 333 K, C2H5OH/raw material = 6).

Time (min)

FIGURE 15.12 Esterification with ethanol: effect of the water content (beef tallow, 0.21% H2SO4, T = 333 K, C2H5OH/raw material = 6).

the determination of the equilibrium constant of fatty acid esterification, which can be conveniently used to select the alcohol/raw-material ratio and the number of required stages. Biodiesel that meets all the requirements of standards has been produced from brown fats of acidity higher than 50 and from emulsions derived from degumming of different types of vegetable oils.

REFERENCES

1. Commarieu, A., et al., Fries rearrangement in methane sulfonic acid, an environmental friendly acid, J. Mol. Catal. A: Chemical, 182, 137, 2002.

2. Uwaydah, I. et al., U.S. Patent 5,696,274, 1997.

3. Fritch, J., Fruchey, O., and Horlenko, T., U.S. Patent 4,954,652, 1990.

4. Vogt, A., Kouwenhoven, H., and Prins, R., Fries rearrangement over zeolitic catalysts, Appl. Catal., 123, 37, 1995.

5. Jayat, F., Sabater Picot, M.J., and Guisnet, M., Solvent effects in liquid phase Fries rearrangement of phenyl acetate over a HBEA zeolite, Catal. Lett., 41, 181, 1996.

6. Mueller, J. et al., U.S. Patent 4,508,924, 1985.

7. Freese, U., Hinrich, F., and Roessner, F., Acylation of aromatic compounds on H-Beta zeolites, Catal. Today, 49, 237, 1999.

8. Subba Rao, Y.V. et al., An improved acylation of phenol over modified ZSM-5 catalysts, Appl. Catal. A: General, 133, L1, 1995.

9. Jayat, F. et al., Acylation of phenol with acetic acid: effect of density and strength of acid sites on the properties of MFI metallosilicates, in Stud. Surf. Sci. Catal, Vol. 105, Chon, H., Ihm, S.-K., and Uh, S., Eds., Elsevier, Amsterdam, 1997, p. 1149.

10. Padró, C.L. and Apesteguía, C.R., Gas-phase synthesis of hydroxyacetophenones by acylation of phenol with acetic acid, J. Catal, 226, 308, 2004.

11. Heidekum, A., Harmer, M.A., and Holderich, W.F., Highly selective Fries rearrangement over zeolites and Nation in silica composite catalysts: a comparison, J. Catal., 176, 260, 1998.

12. Neves, I. et al., Acylation of phenol with acetic acid over HZSM-5 zeolite, reaction scheme, J. Molec. Catal., 93, 169, 1994.

13. Clark, G.S., Menthol, Perfumer Flavorist, 23, 33, 1998.

14. Fleischer, J., Bauer, K., and Hopp, R., German Patent, DE 2,109,456, 1971.

16. Misono, M. and Nojiri, N. Recent progress in catalytic technology in Japan, Appl. Catal., 64, 1, 1990.

17. Trasarti, A.F., Marchi, A.J., and Apesteguía, C.R., Highly selective synthesis of menthols from citral in a one-step process, J. Catal., 224, 484, 2004.

18. Kozhevnikov, I.V., Catalysis by heteropoly acids and multicomponent polyoxometa-lates in liquid-phase reactions, Chem. Rev., 98, 171, 1998.

19. Aramendía, M.A. et al., Selective liquid-phase hydrogenation of citral over supported palladium, J. Catal., 172, 46, 1997.

20. Maki-Arvela, P. et al., Liquid phase hydrogenation of citral: suppression of side reactions, Appl. Catal. A: General, 237, 181, 2002.

21. Singh, U.K. and Vannice, M.A., Liquid-phase citral hydrogenation over SiO2-sup-ported group VIII metals, J. Catal., 199, 73, 2001.

22. Ravasio, N. et al, Intramolecular ene reactions promoted by mixed cogels, in Stud. Surf. Sci. Catal., Vol. 108, Blaser, H.U., Baiker, A., and Prins, R., Eds., Elsevier, Amsterdam, 1997, p. 625.

23. Milone, C. et al., Isomerisation of (+)citronellal over Zn(II) supported catalysts, Appl. Catal. A: General, 233, 151, 2002.

24. Fuentes, M. et al., Cyclization of citronellal to isopulegol by zeolite catalysis, Appl. Catal., 47, 367, 1989.

25. Chuah, G.K. et al., Cyclisation of citronellal to isopulegol catalysed by hydrous zirconia and other solid acids, J. Catal, 200, 352, 2001.

26. Dunn, R.O. and Knothe, G., Alternative diesel fuels from vegetable oils and animal fats, J. Oleo. Sci., 50, 415, 2001.

27. Maa, F. and Hannab, M.A., Biodiesel production: a review, Bioresource Technol., 70, 1, 1999.

28. Bondioli, P. et al., Storage stability of biodiesel, JAOCS, 72, 699, 1995.

29. Dmytryshyn, S.L. et al., Synthesis and characterization of vegetable oil derived esters: evaluation for their diesel additive properties, Bioresource Technol., 92, 55, 2004.

30. Lang, K. et al., Preparation and characterization of biodiesel from various bio-oils, Bioresource Technol., 80, 53, 200.

31. Freedman, B., Butterfield, R.O., and Pryde, E.H., Transesterification kinetics of soybean oil, JAOCS, 63, 1375, 1986.

32. Alcantara, R. et al., Catalytic production of biodiesel from soybean oil, used frying oil and tallow, Biomass Bioenergy, 18, 515, 2000.

33. Pisarello, M.L. et al., Biodiesel: producción a partir de materias primas de alta acidez, in Proc. XIX Iberoamerican Symposium on Catalysis, Mérida, México, 2004, p. 63.

34. Demirbas, A., Biodiesel fuels from vegetable oils via catalytic and non-catalytic supercritical alcohol transesterifications and other methods: a survey, Energy Convers. Manage., 44, 2093, 2003.

Was this article helpful?

0 0

Post a comment