Partition of Free Fatty Acids in Deacidification of Macaúba Pulp Oil by Liquid-Liquid Extraction using Ethanol/Water as Solvent

Transcription

1 Defect and Diffusion Forum Vols (2011) pp Online available since 2011/Apr/20 at (2011) Trans Tech Publications, Switzerland doi: / Partition of Free Fatty Acids in Deacidification of Macaúba Pulp Oil by Liquid-Liquid Extraction using Ethanol/Water as Solvent Renata Gomes Brito Mariano 1,a, Carolina Moura da Silva 1,b Sonia Couri 2,c, Regina Isabel Nogueira 3,d, Suely Pereira Freitas 1,e 1 Universidade Federal do Rio de Janeiro - Escola de Química. Av. Horácio Macedo, 2030, Bloco E - Sala E Ilha do Fundão, Rio de Janeiro, Brazil. CEP: Instituto Federal de Educação, Ciência e Tecnologia do Rio de Janeiro. Rua Senador Furtado, 121, Maracanã, Rio de Janeiro, Brazil. CEP: Embrapa Food Technology. Av. Das Américas, 29501, Guaratiba, Rio de Janeiro, Brazil. CEP: a renatagbm@yahoo.com.br; b carolmoura88@hotmail.com; c sonia.couri@gmail.com; d nogueira@ctaa.embrapa.br; e freitasp@eq.ufrj.br Keywords: Acrocomia aculeata; vegetable oil; de-acidification; liquid-liquid extraction; ethanol extration; Abstract: Free fatty acids (FFA) is one of the most frequently determined quality indices in fats and oils industry because it has the economic impacts on production [1]. The FFA in macaúba pulp oil is usually very high (10 to 40%) due to enzymatic activity in the raw material [2]. In this case, alkali neutralization is not economically recommended due to neutral oil loss by occlusion in soapstock. The differential solubility of fatty acids and triacylglycerides (TAG) in various organic solvents has formed the basis of several processes for deacidification of crude oils by liquid liquid extraction [3]. The aim of the present work was to evaluate the influence of raw oil acidity, ethanol/water ratio, water content in ethanol and temperature extraction on partition coefficients of FFA during macaúba oil deacidification. The macaúba crude oil presented high acid index, about 86 ± 1.2 mg KOH/g. The extraction data for the system macaúba oil + free fatty acids + ethanol + water was determined for oil:solvent mass ratios 1:1. Alcoholic solutions containing 2 to 10 % (w/w) of water were used to FFA extraction and the partition coefficients were determined after 24 hours at room temperature (293 K). The phase s separation was observed for water content in the mixture above 4%. After solvent evaporation, the FFA in oil phase was evaluated by AOCS standard method. Regarding the high acid oil, the single stage liquid-liquid extraction reduced the FFA in macaúba pulp oil in about 23 % while the three batch stages reduced up to 60%. At the selected parameters (ethanol 94 o GL, solvent/substrate ratio = 2), the partition coefficients in the firs stage to FFA (K FFA ) were 30.45, and 1.95, respectively to high acid, acid and low acid macaúba oils and the partition coefficients to TAG (K TAG ) were 7.07, 0.04 and 0.02, respectively to high acid, acid and low acid macaúba oils. These results show the potential of ethanol to vegetable oil deacidification. Besides reduced generation of environmental pollutants, this new approach could also lead to reduction in oil losses, especially for low acid oil. Introduction The crude oil extracted from oleaginous pulps is, mainly, a mixture of free fatty acids (FFA), mono-, di-, and triglycerides, pigments (carotenoids), sterols and tocopherols. Refining usually refers to the operations of degumming, neutralization or deacidification, bleaching and deodorization, to remove undesirable components and make the fat suitable for edible and industrial uses. FFA can be formed after the oil bearing materials have been harvested. Hydrolysis of ester bonds in lipids, All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of TTP, (ID: /04/11,21:03:09)

2 Defect and Diffusion Forum Vols resulting in the liberation of FFA, may be caused by enzyme action (lipase) or by heat and moisture. Any increase in the oil acidity must be avoided once the FFA is more susceptible to oxidation than the triglycerides. Deacidification represents a delicate and difficult step in the oil refining, since it determines the quality of the final product [4]. Chemical deacidification is the most commonly used method for oil deacidification. This method is carried out by the addition of an alkali, (sodium hydroxide), to degummed oil, thereby precipitating the FFA as soapstock. Finally it is removed from the neutral oil by centrifugation followed by washing steps. This chemical deacidification, promote a considerable oil loss due to the hydrolysis of neutral oil by caustic. Furthermore, the oil lost also occurs in the form of occlusion in soapstock, as much as 50% of its weight of neutral oil, reducing consequently the overall yield of refined product. The neutral oil loss depends of FFA content in the crude oil [5]. The physical deacidification offers many advantages if compared to alkali neutralization such as elimination of soapstock, as well as reduced effluent quantity and improved product yield. However, this method requires high energy consume and forms side reaction products, such as polymers and trans-isomers [6]. Additionally, steam stripping deacidification reduces the tocopherol content and destroys all carotenes present in palm oil. Ethanol, methanol and acetone have been recommended by Turkay & Civelekoglu [7] as solvents for vegetable oils deacidification. However, it has not been possible to obtain a complete separation, because the solubility of triglycerides increases in direct proportion to their FFA contents. Consequently, the deacidification with such solvents could be only partially successful. In Brazil, there is considerable interest in using ethanol as solvents for biodiesel production due to its abundance and very low toxicity. Furthermore, improving the quality of the processed oils by selected non-edible oils and by reducing energy consumption has been among the most active researches to make biodiesel a sustainable combustible. The macaúba is a native fruit of the Brazilian cerrado with potential to produce up to 6.5 ton of oil/ha. The macaúba fruit is a spherical nut with diameter between 3.0 and 4.5 cm and presents, in average, 25% of shell, 36% of pulp, and 39% of endocarp. Its pulp contains, in dry basis, about 65% of oil. Due to the high content of unsaturated fatty acids in the oil (65 ± 1%), the high moisture (48 ± 1.21%) and the lipoxigenase endogenous in the macaúba pulp, their oil presents, few weeks after fruit harvest, high content of FFA and is usually used to soap production. Furthermore, the porosity of macaúba shell makes possible the pulp contamination that promotes pulp oil oxidation [8]. The present work reports the effects of FFA in raw-oil, water content in the ethanolic solvent and temperature extraction during the macaúba pulp oil deacidification by liquid liquid extraction. Additionally, the number of batch stages and the kinetic of deacidification were analyzed at the selected conditions. Materials and Methods They were used three types of oils like raw material: high, acid and low acid macaúba oils, supplied by Paradigma oil mill, locate at Santa Luzia, Minas Gerais State, Brazil. The anhydrous ethanol (P.A.) was purchased from VETEC (Brazil). The mixture ethanol/water at different volume concentrations (2 to 10 o GL) was prepared by addition of distillated water using a Gay-Lussac alcoholmeter. Preliminary experiments make possible to select the range for water content in the solvent.

3 556 Diffusion in Solids and Liquids VI The effect of temperature extraction (30 and 40 o C) and water content in the ethanol (96º, 92º and 90 o GL) was carried out according an experimental factorial (3 2 ) planning, with triplicate for all points. The liquid-liquid extraction was carried out for 2 hours in a thermostatic bath at 30 rpm (New Brunswick Scientific). The homogeneous mixtures were cooled at 10 ± 1 o C for 2 hours resulting in well-defined two phases. The phase separation was carried out in a separation funnel at 21 ± 1 o C. The oil solvent was removed under convective air at 21 ± 1 o C and the FFA in the oil phases was determined. The software Statistica (Statsoft, v. 9.0) was used to analyze the results. After selected extraction temperature, the deacidification was carried out in three batch stages at different water content in ethanol (1:1 and 1:2). The kinetic of deacidification was carried out, after the selected operational conditions, during 2 hours at intervals of 30 minutes. For equilibrium data comparison, the extraction and cooling processes was carried out for 24 hours. After this time, the partition coefficient was calculated as the mass ratio of the solute in extract and the same solute in the raffinate, at equilibrium conditions. A central composite design was employed to evaluate, simultaneously, the effect of water content in the ethanol (90, 92 and 94 o GL); solvent/oil ratio (1; 1.5 and 2) and separation temperature (20, 22 and 24 o C), from acid macaúba oil deacidification at room temperature for 1 hour. Analysis: FFA was measured by titration with NaOH (1N) according AOCS standard method [9]. The total solvent was evaporated under a convective air at room temperature (20 ± 2 o C). The TAG contents were determined by difference. All experiments and analysis were performed at least in triplicate. Results The FFA in the macaúba raw-oil was found ± 0.57 %, 8.76 ± 0.03 % and 2.07 ± 0.01 %, respectively for high acid, acid and low acid oils, expressed as oleic acid. From the statistical data in Pareto chart (Fig. 1) it can be concluded that temperature and water content in ethanol has not a significant effect (p < 0.05) in the FFA reduction. However, the solvent/substrate ratio and initial acidity in raw-material presented, as expected, a significant effect in the partition coefficient to FFA (K FFA ) as can be seen in Fig. 2. Figure 1. Effect of temperature extraction and water in ethanol on the FFA (%) in raffinate phase (FFA expressed in solvent-free basis). Solvent:substrate ratio =1

4 Defect and Diffusion Forum Vols Figure 2. Effect of solvent/ substrate ratio and raw-oil acidity on FFA partition coefficient (K FFA ). Ethanol 96 o GL. Physical deacidification of macaúba high acid oil by liquid liquid multiple-stages batch extraction, using aqueous ethanol as solvent, allowed the FFA reduction. As the stages number increased of 1 to 3, the FFA reduction increased of 20 to 60% (Fig. 3). Regarding the high acid macaúba oil, after the second stage of extraction arise a severe reduction of neutral oil in the raffinate phase. At the opposite, for low acid oil the maximum oil losses was about 2 %. According literature data, the solubility of TAG in ethanol increases in direct proportion to their FFA contents [10]. Figure 3. FFA reduction as function of stages number and water content in ethanol. a) high acid oil; b) low acid oil. As can be observed in Fig. 4, the FFA in extract phase was very much superior to FFA in the raffinate phase (K FFA = 30,45). Regarding the high acid macaúba oil, the TAG in the extract is too very high (K TAG = 7.01). In this case, the liquid-liquid extraction is not a selective method. In contrast, the deacidification of acid oil (K TAG = 0.04) and low acid oil (K TAG = 0.02) using ethanol (94 o GL) presented very high selectivity to FFA in the solvent phase and TAG in the extract phase.

5 558 Diffusion in Solids and Liquids VI This result indicates that the FFA is almost completely removed to extract due to the high solubility of FFA in ethanol (94 o GL). Figure 4. Partition of FFA, TAG and solvent (keeping solvent: substrate ratio = 2 and ethanol at 94 o GL). SFB: solvent-free basis. The solvent/substrate ratio and water content in the ethanol are relevant factors on acid oil deacidification. As solvent/substrate increase (1:1 to 2:1) the FFA removal increase and the macaúba oil acidity decrease about 56% as compared with raw-oil. Conclusions The deacidification by liquid-liquid extraction from high acid oil has several drawbacks such as neutral oil losses and residual FFA in the raffinate phase. In this case, it was necessary to complete the neutralization by the conventional chemical method. In contrast, for acid and low acid oil, it can be suggested that the conditions corresponding to solvent oil mass ratios between 1:1 and 2:1, water content in the solvent between 6 and 8 % permit the macaúba oil deacidification with acceptable levels of neutral oil losses, (between 2 and 3 %). Further studies on this new approach are necessary, including an assessment of the economic viability of ethyl esters (biodiesel) production from extract phase using immobilized lipase or heterogeneous catalyst, to successfully replace the conventional technology.

6 Defect and Diffusion Forum Vols References [1] B.M. Bhosle and R. Subramanian: J. Food Eng. Vol. 69 (2005), p [2] R.F. Nascimento, A.I.S. Brígida, C.C.C.M. Silva; M.H. Rocha-Leão, M.A.Z. Coelho, S.P. Freitas. In: 10th International Chemical Engineering Conference, CHEMPOR (2008). [3] A.C. Bhattacharyya and D. K Bhattacharyya: J. Oil Tech. Assoc. India Vol. 15 (1983), p. 36. [4] A. J. C Anderson. In P. N. Williams, 2nd ed., Refining of oils and fats for edible purposes, p. 92, Pergamon Press (1962). [5] M. V Ruiz-Mendez, G. Marquez-Ruiz, M. C. Dobarganes: J. Am. Oil Chem. Soc. Vol. 73 (1996), p [6] R. Sengupta, D. K. Bhattacharyya: J. Am. Oil Chem. Soc. Vol. 69 (1992), p [7] S.Turkay, H. Civelekoglu: J. Am. Oil Chem. Soc. Vol. 68 (1991), p. 83. [8] R.F Nascimento, R.G.B. Mariano, A.I.S Brígida, M. H. Rocha-Leão, S.P. Freitas. Revista de Química Industrial, SBQ-Rio de Janeiro, 18, [9] AOCS. American Oil Chemists Society. Official methods and recommended practices of the American Oil Chemist s Society. 5 th ed. Champaign (2002). [10] B.M. Bhosle, R. Subramanian: J. Food Eng. Vol. 69 (2005), p. 481.

7 Diffusion in Solids and Liquids VI doi: / Partition of Free Fatty Acids in Deacidification of Macaúba Pulp Oil by Liquid-Liquid Extraction Using Ethanol/Water as Solvent doi: /