Influence of refining on the properties of buriti oil ( Mauritia flexuosa L

Vegetables oils are renewables sources of ready availability with vast commercial employability. The application of these oils, generally, requires refining procedures for the removal of substances that make it unfeasible, such as phosphatides, proteins and free fatty acids. The objective of this study was to evaluate the effect of acid and aqueous degumming on the physical-chemical properties of crude oil of Mauritia flexuosa L.f. (buriti), palm tree of the family Arecaceae, very commercially exploited and well adapted to the Cerrado biome. Physical and chemical analysis were performed as acidity index, peroxide index, refraction, viscosity, density, gas chromatography coupled to the mass spectrometer and infrared spectroscopy before and after the refining process. The acidity, density and refraction indices didn´t change significantly during the treatments; however, significant differences were observed for the peroxide and viscosity parameters after the acid degumming process. Chromatographic and infrared spectroscopy results didn´t reveal changes in the constitution of constituent esters. The acid degumming was more efficient for the removal of undesirable substances, with significant variation (p < 0.05) for the peroxide index after this process. However, M. flexuosa oil presented good stability, keeping indexes acceptable in view of the current resolutions for its use.


Introduction
Vegetal oils have a great commercial applicability and the attractiveness of this resource is a renewable source of ready availability (Adekunle, 2015). Due to its low cost, vegetal oil is a raw material with numerous advantages, like biodegradability and low toxicity, characterizing its significant employability in the industrial sector (Saurabh et al., 2011). In addition, the global concern with the growing demand for energy and with the finitude and risks of the fossil matrix has driven the most diverse studies on renewable energy sources, such as, for example, the use of oils of vegetable origin for the synthesis of biofuels (Shereena & Thangaraj, 2009;Cruz et al., 2022;Zanini et al., 2022). Other advantages of biofuels include the fact that they are non-toxic and constitute a neutral source of carbon, that is, they generate an amount of carbon dioxide similar to that absorbed by the source of biomass, plants, via photosynthesis (Zore et al., 2021).
For a better utilization of vegetal oils its necessary a refining process, that promotes the remotion of unwanted substances originated from lipidic oxidation processes, such as phospholipids, free fatty acids and proteins, making it more resistant to degradation, especially during storage (Ramalho & Suarez, 2013;Cordeiro, 2020). The phospholipids present in vegetal oils are found in the hydrated form and they can be removed by treatments involving addition of water, or they can be in the not hydrated form, generally phosphatidic acid salts resulting of enzymatic hydrolysis (Gunstone, 2011).
The degumming is the process of removal of substances as hydratable phosphatides (phospholipids), through aqueous degumming, and not hydrated through acid degumming (Morais et al., 2012;Oliveira et al., 2022). This process considerably modifies physical-chemical characteristics of oils, such as levels of acidity, peroxide, iodine, free fatty acids and of proteinaceous and metabolites remains (Luz et al., 2011a). In Brazil chemical refining is the most used, where after degumming is realized a neutralization, which the objective is reduce the free fatty acids, that when reacting with sodium hydroxide produce soap (Aued-Pimentel et al., 2009). Posteriorly the oil wash is realized, that consists in subsequent additions of water at room temperature and warmed alternately, in order to remove soap remains that it can still be present (Prado et al., 2014). The drying consists in removing water remains, that it can be realized in evaporator with constant agitation or by addition of sodium sulphate, separating posteriorly the precipitate (Encarnação, 2008).
Mauritia flexuosa L.f., (buriti) is a palm tree of the Arecaceae family found in the whole Brazilian territory and especially in Amazon and Cerrado biomes (Cândido et al., 2015). M. Flexuosa oil is similar in composition of fatty acids to others oleaginousness of great commercial value as olive oil, canola and peanut, in some cases, presents superior indices (Freire et al., 2016). Considering the presence of many oleaginousness plants in the North of Minas Gerais and the importance of M. flexuosa to many industries sectors of national scope, this present work objectived evaluate the physical-chemical characteristics of M. flexuosa oil before and after refining process.

Study area
The fruits were collected from a natural population in Environmental Protection Area of Pandeiros River -APA Pandeiros, north region of Minas Gerais. The conservation unit embrace municipalities of Januária, Bonito de Minas and Cônego Marinho, presenting different phytophysonomies, like veredas (palm swamps) which has M. flexuosa as a characteristic palm species (IEF, 2015).

Extraction and storagement
The oil used for physical-chemical analysis was extracted from the fruit's mesocarp. Samples were benefited with the cold pressuring method of extraction, with a mechanical press ERT 75 II -Scott Tech. After extraction process, the obtained raw oil was put in a centrifuge for separation of impurities. These processes were executed in the Agricultores Familiares cooperative and Agroextrativistas Grande Sertão LTDA.

Oil tenor
The oil's tenor was obtained by chemical extraction with solvent through a device of Goldfish type, according to methodology of National Institute of Science and Animal Technology. The extraction process of the oil occurred by continuous reflux of the solvent in samples for 4 hours, with condensation velocity between five to six drops per second. After extraction, the cups, returned to heating chamber at 105° C, for 30 minutes, for complete removal of reagent. The difference between the last cup weight and the empty cup corresponded to the extracted oil's weight. The calculation of yield was obtained by gravimetric relation (Detmann, Souza & Valadares, 2012).

Physical-chemical analysis
The acidity level was determined by the method that uses as titrant, sodium hydroxide 0.01 mol/L and phenolphthalein as indicator (American Oil Chemists` Society -AOCS, 2004). To refraction index determination an Abbé refractometer was used, adjusted with distilled water at 25 ºC (AOCS, 2004). To density determination a 25 mL pycnometer was used at 25°C, determining posteriorly the reason of mass of the sample in relation to the mass of water (AOCS, 2004). To determination of viscosity a Ford cup viscosimeter was used, using the number three orifice according to the characteristics of the sample. The viscosimeter was leveled, adjusting the levelling feet. The orifice was closed manually and the cup filled with the sample to the highest level, spilling the sample slowly to avoid bubble formation. The excess of sample was removed with a glass plate. The flowability time was measured in seconds for viscosity calculation (ASTM, 2016). Thcamarge peroxide indexes were realized according to the method of AOCS (2004), Where 5 g of oil were dissolved in 25 mL of acetic acid solutionchloroform (3:2 v/v) with addition of 1 mL of saturated solution of potassium iodate. After repose, was added distilled water and 2 mL of 1% amid solution. The titrimetric method was applied, having as titrant solution the sodium thiosulphate at 0.01M (AOCS, 2004).

Gas chromatography coupled to mass spectrometry (GC-MS)
The samples were analyzed through chromatography in gaseous phase coupled to mass spectrometry, using the device Perkin Elmer Clarus 560-600 MS, with fused silica capillary column DB-5 (30 m; 0,25 mm internal diameter; 0.25 μm film) and helium as carrier gas. The injector temperature was at 260 °C. For the column, the initial temperature was at 100 °C, increasing from 100° to 250 °C in the rate of 10 °C/min, staying in this temperature for 40 min. The detector temperature was at 290 °C and in the system interface GC-MS at 290 °C (Collins, Braga & Bonato, 1997). The mass detector was adjusted to operate with electron impact ionizations (70 eV) and mass scan with interval of 30 to 600 Da. The compounds identification was realized through the comparation of the samples mass spectra to those existing in the device database, with data of literature and with standards injections of samples. For quantitative analysis of the device GC-MS it was calibrated with reference compounds, representatives of the principal classes of compounds present in the samples (Munari, Cavagnino & Cadoppi, 2007).

Infrared spectrocopy (IR)
Samples were submitted to analysis by infrared spectroscopy in spectrometer with Fourier transform FTIR 640. The spectra were registered in KBr tablets.

Degumming
For the acid degumming process, a sample of 200 g of raw oil was heated to 90° C and acidified with 0.30 mL of phosphoric acid. After 10 minutes of agitation, was added 9.8 mL of sodium hydroxide at 5%. The oil was centrifuged at 3000 rpm for 20 minutes, according to adapted methodology (Penedo & Coelho, 1997). Posteriorly, the oil was poured in a separate hopper for process of washing with water at room temperature and at 90 °C alternately every 30 minutes. After testing the pH of the last washing, the sample was liberated for dehumidification, with addition of 4 g of anhydrous sodium sulfate, maintained for 15 minutes e posterior discard of precipitate. In the aqueous degumming, 200 g of raw oil were heated at 80 °C, adding 8 ml of distilled water with posterior agitation for 30 minutes. The subsequent processes of neutralization with sodium hydroxide at 5%, washings with water in alternative temperatures and dehumidification with anhydrous sodium sulfate were identical to the utilized in the acid degumming.

Statiscal analysis
For the physical-chemical analysis was utilized the One-way ANOVA test and for percentage of majoritarian fatty acids, Two-way ANOVA and Turkey test for multiple comparation between the averages with 5% of probability (P<0.05). The software used was GraphPad Prism 6.01.

Caracterization of Mauritia flexuosa oil
The acid and aqueous degumming did not lead to alterations in composition of the fatty acids of the M. flexuosa oil ( Figure 1). There was no significative difference between the treatments for majoritarian oleic, palmitic, stearic and linoleic fatty acids (Table 1).  This constancy observed in the oil's constituent fatty acids even after refining is a relevant characteristic for its utilization (Oliveira et al., 2013). The averages of relative areas obtained in chromatographic analysis are in accordance with the values cited by Freire et al. (2016) for oleic and palmitic acids, but differs for stearic and linoleic acids.
The chromatograms obtained in this study showed similar peaks for samples before and after the refining processes ( Figure 2). The infrared spectra (Figure 3-A, B, C) presented absorption bands around 2920 and 2850 cm-1 attributed to C-H sp3 stretching (groups CH, CH2 and CH3) common in many classes of aliphatic compounds such as fatty esters (Silverstein et al., 2006).

Acid and aqueous degummings
The acid level wasn´t present significative difference although a reduction of its percentage of RWO (3.14%) for ACD (2.84%) and for AQD (2.77%) ( Table 2). This small reduction can be related to the degumming process, since as, this value is related to purity conditions of the oils' constituent lipids. Elevated acidity levels indicate a higher lipids decomposition and a higher consume of sodium hydroxide for pigment neutralization (Aquino et al., 2012).
Other two parameters that weren´t presents variation during the treatments were the relative density and refraction index. According to Silva, Carvalho, Conrado, Fook and Leite (2012), the density of an oil is a key-factor, since, in smaller values present a higher oil fluidity increasing its applicability. In turn the refraction index, is specific for each oil and can suffer alterations by thermal and oxidative processes (Jung, Park & Yoon, 2016). In this study, the refining processes didn´t modify the oil constitution to the point of influencing these two indices that remained unaltered. There was significative variation (p < 0.05) of the peroxide index after the acid degumming process (Figure 4-B). Source: Authors.
The obtained values for this index (Table 2)

Effects of temperature in the values of peroxide and viscosity
The peroxide and viscosity values are physical-chemical properties susceptible to significative variations in function of temperature. In oils storage at room temperature, the viscosity tends to increase with the elevation of peroxide index (Ribeiro et al., 2010). Such peroxides, in turn, are related to the deterioration of fats and oils by hydrolysis and oxidation (Mitrea et al., 2022). However, with temperature variation, the relation of viscosity with peroxide index is different (Figure 5-A). According to Canciam (2010) the viscosity is connected to the size of the constituent carbon-chains of an oil and also to the amount of unsaturation of these molecules which have its forces reduced with temperature increase, promoting the decline of this index. Due to the process of auto-oxidation the peroxide index tends to get higher proportionally to the temperature increase ( Figure 5-B). However, there is a small decline in this index, at a given temperature, related to the oxygen consume and the transformation of peroxides into hydroperoxides. This step consists in the finishing of the auto-oxidation process, where the absence of free radicals to reacts with oxygen promotes the end of the process (Manhães, 2014).

Conclusion
The degumming processes reduced the acidity and peroxide values of raw M. flexuosa oil. There was a significative reduction for the peroxide index after acid degumming and an increase of viscosity in the same process. The degummed oil assembles inside the acceptable parameters for its commercial utilization even though after refining processes. The stable characteristic of M. flexuosa oil was observed before degumming treatments involving addition of chemical substances and temperature increase necessary for the treatments. The results after the physical-chemical analysis, chromatographic and spectrometric in infrared confirmed this stability.
Research of this nature should be extended to other palm trees native to the cerrado, such as those of the genus Acrocomia and Attalea, covering not only the mesocarp as a source of raw material for the production of oils, but also the almond. In this way, it is possible to add value to plant species and encourage the income of communities and traditional peoples who depend on the use and exploitation of palm trees as a source of income and survival.