Industrial Crops and Products 43 (2013) 207212

Contents lists available at SciVerse ScienceDirect

Industrial Crops and Products

journal homepage: www.elsevier.com/locate/indcrop

Study of the methyl crambe (Crambe abyssinica Hochst) and soybean biodiesel oxidative stability
Willian Tenfen Wazilewski , Reinaldo Aparecido Bariccatti, Gislaine Iastiaque Martins, Deonir Secco, Samuel Nelson Melegari de Souza, Helton Aparecido Rosa, Luiz Incio Chaves
UNIOESTE Western Paran State University, Post-Graduation Program, Master Course of Energy in Agriculture, Rua Universitria, 2069, CEP: 85.819-130 Bairro Faculdade, Cascavel, Paran, Brazil

a r t i c l e

i n f o

a b s t r a c t
The quality of the biodiesel is of most importance for the success in the marketing of this fuel. Among the qualitative parameters there is the kinematic viscosity that can be affected due to the degradation of biofuel. Such degradation occurs on account of the weather changing and also because of some catalytic agents. Because of that the stability to oxidation was denominated as a factor which relates the resistance level of the product to the different oxidizing agents. The present work aimed to study the oxidative stability of the biodiesel produced of the crambe and the soybean vegetal oil. The techniques of infrared spectroscopy, refractive index and kinematic viscosity were used to monitor the level of oxidation of the biodiesels. The biodiesels were submitted to a thermal stress in an incubator, at 130 C for a nal period of 22 h of study, and were contaminated with iron and bronze lings. The viscometric studies showed that the crambe biodiesel is more stable than the soybean biodiesel at the end of their exposure to the thermal stress, as well as that the addition of 5% and 10% of crambe biodiesel lowers the viscometric alteration in about 43.9% regarding the pure soybean biodiesel. By observing the index of refraction it was veried that the alteration in the soybean biodiesel contaminated with bronze was enhanced when compared to the pure biodiesel and the biodiesel contaminated with iron. As for the pure crambe biodiesel and the biodiesel contaminated with iron, the alteration in the index of refraction is minimum, it is only signicant for the biodiesel contaminated with bronze, but lower, however than the one observed for the soybean/bronze biodiesel. 2012 Elsevier B.V. All rights reserved.

Article history: Received 5 June 2012 Received in revised form 21 July 2012 Accepted 24 July 2012 Keywords: Crambe abssynica Hoechst Biodiesel Index of refraction Oxidative stability

1. Introduction The environmental impacts caused by the use of fuels made out of fossil sources, and the concerns regarding the Crude Oil Inventories have been motivating studies and researches for renewable sources of biofuels, mostly for the replacement of diesel. According to Hill et al. (2006) in order to be a viable alternative the chain of the biofuel must have a positive energy balance, present environmental benets, to be economically competitive and to be produced on a large scale without interfering in the food supply chain. Brazil has natural conditions which are particularly favorable for the production of oilseeds. Despite the protectionisms, this production registers a strong expansion since it is competitive in the worldwide markets, contributing in harnessing the productive resources of the country. The development of the area planted with temporary oilseeds in Brazil has been persistently increasing, as it is

Corresponding author. E-mail address: tw.willian@gmail.com (W.T. Wazilewski). 0926-6690/$ see front matter 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.indcrop.2012.07.046

one of the key points to overcome the demand for the raw material necessary to the production of biodiesel (Osaki and Batalha, 2011). The used biofuel in the partial or in the total replacement of diesel in internal combustion engines, diesel cycle, can be obtained through the pyrolysis reaction, the cracking process and the transesterication process, the last one being the most used, of animal fat and vegetable oil. In a chemical matter, the biodiesel is nothing more than a mixture of alkyl esters of straight-chains, once its production also results in a co-product, the glycerol (Lbo et al., 2009). As a result of its origin, the biodiesel, when compared to the fossil diesel, presents a degradability much higher in the environment, and for that same reason it is susceptible to oxidative and microbial degradation from its synthesis. The susceptibility to oxidation is a relevant aspect inside the existence cycle of the biodiesel, once that the fatty acids triglycerides, especially the unsaturated ones, present re-actives sites which are susceptible to oxidation (Monyem and Gerpen, 2001). One of the processes to which they are subjected is the oxidative degradation, also known as rancidication, and it causes serious implication to the consumer market.

208

W.T. Wazilewski et al. / Industrial Crops and Products 43 (2013) 207212 Table 1 Analyzed properties of the raw materials used for the biodiesels synthesis. Soybean oil Acidity index (mg of KOH kg Density (g cm3 ) Viscosity (mm2 s1 )
1

According to Tang et al. (2008), among the negative implications of the oxidative degradation of the biodiesel, the increase of viscosity can be highlighted, the elevation of acidity and the formation of gums and undesirable polymeric compounds. Having in view that the main factors that affect the biodiesel characteristics are: the length of the chain (number of carbon atoms), the number of unsaturated chains and the occurrence of other chemical functions (Santos et al., 2011), it is clear then, with this, that the properties of the raw materials are the ones that will result in a biofuel of a good or of a bad quality. The oil obtained through the Crambe abyssinica grain pressing, an oilseed belonging to the Cruciferae family, features in its fatty acid prole more than 56% of erucic acid. This acid contains 22 carbon atoms and only one double bond (Laghetti et al., 1995). The soybean oil presents 54% of linoleic acid and 24% of oleic acid, both of which present 18 carbon atoms and, respectively, two and one double bonds (Sanibal and Mancini, 2004). Because of this drastic difference in the fatty acids prole, theoretically, the biodiesels obtained from these vegetable oils will show different stabilities to oxidation. Currently, the technique used by the international regulations in the oxidation study is a method that was originally proposed by Hadorn and Zurcher (1974), also known as the Rancimat method, which consists in exposing the sample to an air ow (10 l h1 ) at 110 C. In this method, as the reactions of formation of oxidation compounds are intensied, a conductivity increase is veried (Velasco et al., 2004). Although a more accurate way of evaluating the oxidation level is by an infrared spectroscopy, through which it is possible to verify a carbonylic grouping formation, that result from the biodiesel oxidation (Furlan et al., 2010). The aim of this paper is to compare the oxidative stability of biodiesels, produced from crambe and soybean oil, by measuring kinematic viscosity, refractive index and transmittance (infrared spectroscopy) in samples subjected to thermal stress and metal contamination.

Crambe oil 0.68 0.9055 49.4

0.47 0.9138 31.5

2. Material and methods The experiment was carried out in a laboratory belonging to the Western Paran State University Cascavel Campus/PR. The crambe crude oil was obtained from a research company which is located in Maracaju/MS. The oil was extracted from the grain by mechanical pressing, without addition of any solvent. As for the soybean oil, it was obtained from an agro-industry located in the city of Cascavel PR and it was left in an incubator at 60 C for 12 h to remove any eventual excess of water. The biodiesels were obtained by the transesterication reaction via methylic route with a basic catalyst KOH. The reaction was based on the stoichiometric ratio between the triglycerides and the chosen alcohol, as the amount of catalyst and alcohol was established as a percentage with regard to the volume of oil. For the synthesis of the crambe biodiesel, a percent ratio of 25% alcohol (methanol) and of 0.8% catalyst (KOH) on the account of the oil volume (500 ml) was used. As for the soybean biodiesel, a higher percentage of catalyst (1%) and the same amount of alcohol (25%) were used. The oils were placed under constant mechanical stirring and they were heated at 60 C, then the mixture alcohol + catalyst (potassium methoxide) was added in order to initiate the transesterication reaction. The temperature and the stirring were maintained for a period of 30 min to ensure the maximum efciency of the reaction. At the end of the reaction time, all of the contents were transferred to two separation funnels, leaving the mixtures in decantation for 24 h to ensure the total separation of the glycerin. After the separation of the biodiesel, it passed through a washing process, to remove the excess of catalyst and of any

eventual triglycerides or free fatty acids that did not react. The washing process was made with distilled water at a temperature of 80 C in amounts of one third of the initial oil volume, and repeating as many times it was necessary so that the phenolphthalein indicator did not show any reaction at the washing water. So the biodiesels were left in an incubator at 60 C for 20 h to remove the excess of water, and then they were left to reach room temperature in a desiccator. To determine the oxidative stability of the biodiesels, they were submitted to thermal stress in an incubator at 130 C, leaving the samples for different periods of time and collecting aliquots to obtain parameters for the construction of comparative graphics of stability to the undergo stress. Also, iron and bronze lings were used to contaminate the biodiesel and to catalyze the effect, this allows to compare them with the samples free of metal contamination. The lings of iron (low carbon content) and bronze (90% of cooper) were homogenized by using sieves of 2 and 0.1 mm, so that the used metal particles were not inferior to 0.1 mm or superior to 2 mm. The proportions of metals used to contaminate the samples were of 50% based on the weight of biodiesel contained in each beaker. To evaluate a possible anti-oxidant effect of the crambe biodiesel in the soybean biodiesel, mixtures were made with percentages of 5%, 10% and 20% (m/m) of crambe biodiesel added in the soybean biodiesel. These blends were also submitted to thermal stress in an incubator at 130 C. The parameters that were evaluated to monitor the oxidation process were: the kinematic viscosity, the index of refraction and the absorption prole in the mid infrared region. The treatments were repeated and the measures of viscosity and refraction index were done with duplicates. The kinematic viscosity was obtained through a Cannon-Fenske No. 150 capillary viscometer in thermostatic bath at 40 C. An Abb refractometer was used to determine the refraction index, following the method proposed by AOCS (American Oil Chemists Society), in which it is indicated to use the device at a constant temperature of 40 C to make the readings, and performed in duplicates (Jorge et al., 2005). To obtain the absorption prole in the mid infrared region, a Perkin Elmer infrared spectrometer was used, the one which was adjusted to collect a scanning in the spectral bands between 4000 and 2800 cm1 supported by KBr windows and a separator of 0.50 mm.

3. Results and discussion Table 1 compares the physical and the chemical properties of the crambe oil with the ones of the soybean oil. The rst presented a kinematic viscosity of 49.4 mm2 s1 superior to the one of the soybean oil (31.5 mm2 s1 ), this can be explained due to the high concentration of fatty acids with long chains in the crambe oil, containing 22 carbon atoms (erucic acid). The soybean oil does not present expressive amounts of fatty acids with chains superior to 18 carbon atoms, and in association with this, it has a higher number of unsaturated bonds which provides the decrease of viscosity (Garcia, 2006).

W.T. Wazilewski et al. / Industrial Crops and Products 43 (2013) 207212

209

Fig. 1. Kinematic viscosity of the soybean biodiesel, contaminated with metals, according to the time of exposure to thermal stress.

Fig. 3. Kinematic viscosity of biodiesel of crambe, soybean and their blends according to the time of exposure to thermal stress.

The thermal stress studies were replicated and showed same tendency for all measures, because of this just one of the treatments kinematic viscosity graphics were shown. The oxidation of the biodiesel can affect the acidity index, the viscosity and the cetane number (Dunn, 2008). Lacoste and Lagardere (2003) correlated the viscosity variation with the induction period of oxidation obtained by the Rancimat to several biodiesels, conrming that the viscosity can be used as an oxidation indicator. In Fig. 1 it is shown the viscosimetric behavior of the soybean oil in thermal stress in the absence and in the presence of iron and bronze. According to the results showed in Fig. 1, the soybean biodiesel has a thermal stability for about 16 h, after this time period its viscosity increases, indicating a beginning of oxidation. The presence of iron does not noticeably alter the viscosity, indicating that within the studied period this metal does not cause a catalysis in the oxidation of the soybean biodiesel. An opposite situation is veried for the bronze, this metal practically reduced the induction period for the biodiesel oxidation in half-time, indicating it is a catalyst to the oxidation of the soybean biodiesel. Initially, the performed viscosimetric studies with biodiesel made out of crambe oil, as seen in Fig. 2, indicates that this biodiesel has a thermo-oxidative stability higher than the one of the soybean biodiesel (Fig. 1), requiring 20 h to cause an alteration in the viscosity that is similar to the alteration caused in the soybean biodiesel within just 16 h. The effect of the metals in the viscosity alteration observed for the soybean alteration was also veried in the crambe biodiesel. In Fig. 2 it is shown that the iron does not noticeably alter the viscosity of the biodiesel, but the bronze causes a signicant alteration in the viscosity, decreasing the time of the oxidative stability of this biofuel. Schober and Mittelbach (2004) veried that the induction time of biodiesel oxidation depends on the biodiesel that is being studied and on the presence of antioxidant, considering that the best antioxidant to a given biodiesel does not guarantee that it will be the best antioxidant for another biodiesel. Studies of the oxidative

stability of soybean biodiesel conducted by Ferrari et al. (2005) and by Guzman et al. (2009) showed that the oxidative stability of soybean biodiesel signicantly alters when the oil passes through a purication process. The time of oxidative induction for the rst author alters from 277 to 132 min when it is deodorized and for the second author, alters from 2.68 to 0.17 h when the oil is distilled, thus the presence of antioxidant compounds in oil in natura are removed by the purication process by altering its oxidative stability, this assumption was also proposed by Conceico et al. (2009). So small changes in the composition and in the purity of the oil cause signicant alterations in the thermo-oxidative stability of biodiesel. In Fig. 3 there is the viscosimetric study of the biodiesel of soybean, crambe and its blend (soybean/crambe) with 5%, 10% and 20% of crambe biodiesel mixed in the soybean biodiesel. It was veried that after 16 h the alteration of the viscosity of the biodiesels and their blends followed a linear function, with the following equations done using the last three points of the viscosity value from 16 to 22 h of thermal stress: Soybean biodiesel : y = 0.2554x + 1.4658 (R2 = 0.9913) Crambe biodiesel : y = 0.028x + 6.283 (R2 = 0.9848) Blend (Soybean 80/Crambe 20) : y = 0.1432x + 2.84093 (R2 = 0.997)

Blend (Soybean90/Crambe10) : y = 0.1565x + 2.4214 (R2 = 0.998)

Blend (Soybean 95/Crambe 5) : y = 0.1708x + 2.3099 (R2 = 0.976) These equations can be associated to oxidation kinetics in which the inclination indicates the viscosity variation per hour, were the range of the independent variable went from 0.028 up to 0.2554. According to these values, the blend containing 20% of crambe biodiesel reduces the viscosimetric alteration in about 43.9% regarding the pure soybean biodiesel, so, it causes a decrease in the oxidation kinetic of the soybean biodiesel in 43.9% and this effect is smaller, though in the same order of magnitude for the blends of 5% and 10% of crambe biodiesel. In Table 2 it is shown the refraction index of the soybean biodiesel and its mixture with iron and bronze. According to the obtained values, the thermo-oxidative behavior of the witness is similar to the one observed for the biodiesel in the presence of iron, reproducing what was observed also for the viscosity. For

Fig. 2. Kinematic viscosity of the crambe biodiesel, contaminated with metals, according to the time of exposure to thermal stress.

210

W.T. Wazilewski et al. / Industrial Crops and Products 43 (2013) 207212

Table 2 Index of refraction of soybean biodiesel as a function of the time of exposure to thermal stress in an incubator (130 C) in the presence of metal contamination. Time (h) Witness Average 0 3 6 9 12 16 20 S.D. = standard deviation. 1.44975 1.44990 1.45005 1.45035 1.45045 1.45090 1.45225 S.D. 0.00035 0.00057 0.00035 0.00078 0.00078 0.00127 0.00290 Iron Average 1.44975 1.45005 1.45015 1.45045 1.45045 1.45070 1.45150 S.D. 0.00035 0.00035 0.00021 0.00064 0.00064 0.00085 0.00198 Bronze Average 1.44985 1.45030 1.45065 1.45155 1.45200 1.45355 1.45570 S.D. 0.00021 0.00085 0.00092 0.00177 0.00255 0.00346 0.00594

the mixture that contains bronze, it was observed an alteration in the refraction index which is higher than the ones veried for the witness or the mixture with iron, thus this metal has a higher tendency to oxidize the soybean biodiesel than the iron. Not only the average value of the refraction index was affected, but also its standard deviations. That indicates that the thermo-oxidative mixture is more complex than the non-oxidized ones, even these values signicantly differ once comparing the mixtures of biodiesel/bronze and the other mixtures. The values of the refraction index for the biodiesel made out of crambe oil are showed in Table 3. In this table, it is possible to verify that the alterations at the values are minimums for the crambe biodiesel, even up to 20 h at 130 C, assuring the higher stability of this biodiesel in comparison to the soybean oil. Similar to the observed results in the viscosimetric study, the presence of iron causes little alteration in the crambe biodiesel. The bronze has a bigger effect in the alteration of the refraction index of the crambe biodiesel, but this variation is smaller than the one observed with the bronze in the presence of soybean biodiesel. It means that the oxidative effect of the bronze is smaller in the crambe oil once that this biodiesel can be an inhibitor of catalyst of the metallic oxidation. The values of refraction index for the blends of biodiesel made out of soybean and crambe oil are showed in Table 4. In this table it is observed that the mixture of biodiesel made out of soybean/crambe is established confronting it with the pure soybean biodiesel. According to these results, the alteration in the refraction index of the soybean biodiesel is 0.0025 and for its blends is 0.0017, 0.0015 and 0.0013 for 5%, 10% and 20% for the crambe biodiesel, respectively. For the soybean biodiesel, it would be a stabilization of respectively 32.0%, 40.0% and 48.0%. The infrared spectroscopy was used by Furlan et al. (2010) for the studies of biodiesel oxidation. According to theses authors, the main groups that are formed in the oxidation are hydroxyperoxides, alcohol, acids, aldehydes and ketones, signicantly altering the infrared spectrum. In the region between 3500 and 3400 cm1 there is the band associated to the second harmonic of the carbonyl, so this region was used to monitor the formation of this group. As longer the time of submission to thermal stress is, so higher the alteration of this region of the spectrum will be, this way reducing the transmittance in this region, due to the formation of the grouping of carbonyl which absorbs just in this spectral region. In Fig. 4, it is shown the spectral behavior for the soybean biodiesel in the presence and in the absence of metals, and in Fig. 5 the same is shown for the crambe biodiesel. The crambe biodiesel showed a lower variation on its transmittance, going from 70% to 35% in 20 h of thermal stress (Fig. 5), while the soybean biodiesel went from 70% to only 8% in the same period of time under thermal stress (Fig. 4). Hence, in the same period of time there was a larger formation of carbonyl groups in the soybean biodiesel than in the crambe biodiesel. The presence of bronze enhances the formation of carbonyl groups, reecting the transmittance variation, i.e. the

Fig. 4. Infrared spectrum of the soybean biodiesel as a function of the time of exposure to thermal stress.

transmittance of the soybean biodiesel in the presence of bronze subjected to 16 h of heating (results are not shown) is similar to the one with 20 h of heating without the mentioned metal (Fig. 4). The increase of the carbonyl band in the solution in which the bronze is present is so intense that the transmittance of the device is close to zero (Fig. 4), no longer responding in the region between 3450 and 3550 cm1 . It is observed that as longer the time of submission to thermal stress is, so greater the alteration in the region of the spectrum associated to the formation of carbonyl also is. Although the crambe biodiesel contaminated with bronze presented a variation slightly different when compared to the pure crambe biodiesel in the region from 3400 to 3500 cm1 . The pure crambe biodiesel passed through a transmittance alteration from 65% to about 35% and the bronze

Fig. 5. Infrared spectrum of crambe biodiesel as a function of the time of exposure to thermal stress.

W.T. Wazilewski et al. / Industrial Crops and Products 43 (2013) 207212 Table 3 Index of refraction of crambe biodiesel as a function of the time of exposure to thermal stress in an incubator (130 C) in the presence of metal contamination. Time (h) Witness Average 0 3 6 9 12 16 20 S.D. = standard deviation. 1.44960 1.44970 1.44940 1.44985 1.44995 1.45015 1.45045 S.D. 0.00014 0.00028 0.00057 0.00007 0.00007 0.00021 0.00035 Iron Average 1.44960 1.44985 1.44995 1.45000 1.45025 1.45030 1.45050 S.D. 0.00014 0.00007 0.00007 0.00000 0.00035 0.00028 0.00057 Bronze Average 1.44960 1.45040 1.45085 1.45115 1.45165 1.45155 1.45285 S.D.

211

0.00014 0.00042 0.00007 0.00021 0.00092 0.00064 0.00134

Table 4 Index of refraction of blends of soybean and crambe biodiesel as a function of the time of exposure to thermal stress in an incubator (130 C). Time (h) Soybean 100% Average 0 16 18 20 22 1.4502 1.4514 1.4519 1.4474 1.4527 S.D. 0.0002 0.0022 0.0029 0.0094 0.0026 Soybean 95% Average 1.4502 1.4509 1.4510 1.4516 1.4519 S.D. 0.0002 0.0016 0.0018 0.0018 0.0017 Soybean 90% Average 1.4502 1.4510 1.4510 1.4514 1.4517 S.D. 0.0002 0.0014 0.0014 0.0016 0.0016 Soybean 80% Average 1.4501 1.4507 1.4507 1.4512 1.4514 S.D. 0.0003 0.0019 0.0019 0.0016 0.0016 Soybean 0% Average 1.4497 1.4497 1.4501 1.4503 1.4505 S.D. 0 0 0 0 0.0002

S.D. = standard deviation.

contaminated crambe biodiesel passed from 65% to about 30%, when submitted to 20 h of thermal stress. However, for the crambe biodiesel in the presence of bronze, the band in 3650 cm1 is intensied when compared to the other spectrums, this band can be associated to the formation of groups OH of alcohol that, when diluted, absorb between 3600 and 3670 cm1 , thus this metal favors the formation of groups OH in the crambe biodiesel. The variation of the infrared spectrum in the region of carbonyl grouping (35003400 cm1 ), was smaller once in the presence of iron, indicating an interference which is less accentuated of this metal in the oxidative stability. Both, in the soybean biodiesel, Fig. 4, as in the crambe biodiesel, Fig. 5, the iron caused a spectral alteration which is lower than the one observed in the pure biodiesel, thus in both contaminated biodiesels, this metal induced a lower variation of the spectrum corresponding to the carbonyl grouping, when compared to the pure biodiesel, even in the nal time of exposure to thermal stress, 20 h. Knothe and Dunn (2003), using mixtures of copper and methyl oleate, veried that the index of oxidative stability of ester is reduced by half in the presence of copper, while the iron does not induce a reduction so accentuated in the oxidative stability of the methyl oleate. The results here obtained are similar to the ones obtained by Knothe and Dunn, however the effect of the iron in the soybean or crambe biodiesel is small and this metal caused a smaller formation of carbonyl groups (Figs. 4 and 5). 4. Conclusion According to the results analyzed in this work, the crambe biodiesel is more stable than the soybean biodiesel. The mixture of soybean biodiesel with crambe increases its stability when compared to the pure soybean biodiesel. Among the studied metals, the bronze has an accentuated capacity of increasing the viscosity, the index of refraction, the carbonyl groups and the OH groups with thermal stress, indicating that this metal oxides the soybean and the crambe biodiesel, the latter biodiesel suffers a less accentuated oxidation.

For the iron, there was not such an accentuated thermooxidative effect as in the one obtained by bronze, that is associated to the presence of copper in the bronze alloy according to the references. Finally, the crambe biodiesel stabilizes the viscosimetric alteration of soybean biodiesel when they are mixed in different proportions. References
Conceico, M.M., Dantas, M.B., Rosenhaim, R., Fernandes, V.J., Santos, I.M.G., Souza, A.G., 2009. Evaluation of the oxidative induction time of the ethylic castor biodiesel. J. Therm. Anal. Calorim. 97, 643646. Dunn, R.O., 2008. Antioxidants for improving storage stability of biodiesel. Biofuel Bioprod. Bior. 2, 304318. Ferrari, R.P., Oliveira, V.S., Scabio, A., 2005. Oxidative stability of biodiesel from soybean oil fatty acid ethyl esters. Sci. Agric. (Piracicaba, Braz.) 62, 291295. Furlan, P.Y., Wetzel, P., Johnson, S., Wedin, J., Och, A., 2010. Investigating the oxidation of biodiesel from used vegetable oil by FTIR spectroscopy: used vegetable oil biodiesel oxidation study by FTIR. Spectrosc. Lett. 43, 580585. Garcia, C.M., 2006. Vegetable oil transesterication. Diss. (Chemistry Masters Degree). Campinas State University, Campinas, SP, p. 120. Guzman, R., Tang, H., Salley, S., Ng, S.K.Y., 2009. Synergistic effects of antioxidants on the oxidative stability of soybean oil- and poultry fat-based biodiesel. J. Am. Oil Chem. Soc. 86, 459467. Hadorn, H., Zurcher, K., 1974. Zur bestimmung der oxydationsstabilitat von olen und fetten. Deutsche Lebensmittel Rund 70, 5765. Hill, J., Nelson, E., Tilman, D., Polasky, S., Tiffany, D., 2006. Environmental, economic, and energetic costs and benets of biodiesel and ethanol biofuels. Proc. Natl. Acad. Sci. U.S.A. 103, 1120611210. Jorge, N., Soares, B.B.P., Lunardi, V.M., Malacrida, C.R., 2005. Alteraces fsicoqumicas dos leos de girassol, milho e soja em frituras. Quim. Nova 28, 947951. Knothe, H., Dunn, R.O., 2003. Dependence of oil stability index of fatty compounds on their structure and concentration and presence of metals. J. Am. Oil Chem. Soc. 80, 10211026. Lacoste, F., Lagardere, L., 2003. Quality parameters evolution during biodiesel oxidation using Rancimat test. Eur. J. Lipid Sci. Technol. 105, 149155. Laghetti, G., Piergiovanni, A.R., Perrino, P., 1995. Yield and oil quality in selected lines of Crambe abyssinica grow in Italy. Ind. Crop. Prod. 4, 203212. Lbo, I.P., Ferreira, S.L.C., Cruz, R.S., 2009. Biodiesel: parmetros de qualidade e mtodos analticos. Quim. Nova 32, 15961608. Monyem, A., Gerpen, J.H.V., 2001. The effect of biodiesel oxidation on engine performance and emissions. Biomass Bioenergy 20, 317325. Osaki, M., Batalha, M.O., 2011. Produco de biodiesel e leo vegetal no Brasil: Realidade e desao. Organiz. Rurais Agroindust. 13, 227242.

212

W.T. Wazilewski et al. / Industrial Crops and Products 43 (2013) 207212 Tang, H., Wang, A., Salley, S.O., Simon, K.Y., 2008. The Effect of natural and synthetic antioxidants on the oxidative stability of biodiesel. J. Am. Oil Chem. Soc. 85, 373382. Velasco, J., Andersen, M.L., Skibsted, L.H., 2004. Evaluation of oxidative stability of vegetable oils by monitoring the tendency to radical formation: a comparison of electron spin resonance spectroscopy with the Rancimat method and differential scanning calorimetry. Food Chem. 85, 623632.

Sanibal, E.A.A., Mancini, J.F., 2004. Perl de cidos graxos trans de leo e gordura hidrogenada de soja no processo de fritura. Cinc. Tecnol. Aliment. 24, 2731. Santos, V.M.L., Silva, J.A.B., Stragevitch, L., Longo, R.L., 2011. Thermochemistry of biodiesel oxidation reactions: a DFT study. Fuel (Guildford) 90, 811817. Schober, S., Mittelbach, M., 2004. The impact of antioxidants on biodiesel oxidation stability. Eur. J. Lipid Sci. Technol. 106, 382389.