2. Materials and Methods

2.1 Materials

The materials used for the present work are virgin soy oil (SOYA), soy oil used in frying (donated by a local supplier) and Magnesol®(DALLAS GROUP). Reagents with analytical grade - 99.8% Methanol (ANIDROL), 85% Potassium hydroxide-KOH (VETEC), 99% Tetrahydrofuran-THF (VETEC), 99.18% Hexane (NEON), 99.5% Ethyl acetate (VETEC), 99.8% Dichloromethane (MERCK), 99.8% acetonitrile (VETEC) and 97% NaOH (VETEC) - were purchased from a local supplier and used without pre-treatment.

2.2. Methods

The methodology developed in this work is illustrated in Figure 1, which is described in two sections: biodiesel production and recovered of Magnesol® .

2.2.1 Biodiesel Production

First, on a bench scale biodiesel plant, 10 L of biodiesel was produced through a batch process from soybean frying oil and virgin soybean oil. After, the oils (virgin and frying) had their acidity corrected with KOH, as they were not in compliance with the requirements of the National Agency of Petroleum, Natural Gas and Biofuels (ANP). Posteriorly, the two biodiesels were produced using the same conditions, through the alkaline transesterification process with KOH 1% (w/w). Thus, the conditions used in this process were: 6:1 methanol/oil ratio, 65 °C temperature and 1.5 h reaction time. The reactions occurred under constant agitation and constant oil pumping in the reactor for better homogenization. After that, the biodiesels sat for approximately 30 h before the separation of glycerin was carried out, and after which the excess methanol was removed by heating at 65 °C for 1 h (Kucek et al., 2007).
Purification of biodiesel was performed by the dry route with 1% w/w Magnesol®/biodiesel, heated at 50 °C under constant stirring for 30 minutes. After this time, the biodiesel was vacuum filtered and the adsorbent was taken for purification (Farag and El-Anany, 2006).

2.2.2 Recovered of Magnesol®

The magnesol extracted from the biodiesel purification process was purified according to the methodology presented in Figure 2, to which 1g of contaminated magnesol was placed in a conical flask and 11.2 ml THF was added (30:1 ratio (w/w)). The system was stirred for 30 min, subsequently being centrifuged for 3 minutes at 4000 rpm for decantation of Magnesol® and removal of the supernatant. This process was repeated three times, and at the end the product was dried at 70 °C for 24h.

2.4. Characterizations

2.4.1. Characterization of Purified and Unpurified Biodiesel

Some parameters of the biodiesels from the virgin soybean oil and frying oil were analyzed in order to compare the efficiency of the purification with virgin and recovered Magnesol®. The main parameters were established in accordance with the standards described below.

Determination of acid value

Based on standard EN14104, 2.5 g of biodiesel was placed in a 250 ml erlenmeyer flask and 50 mL of ethyl ether/ethanol 2:1 was added, stirring until complete dissolution. Two drops of phenolphthalein solution - 1% in ethanol - was added, and it was titrated with a standardized solution of 0.1 M NaOH. The analyses were performed in triplicate and the results were expressed as mg KOH/g sample (Alves et al., 2016).

Determination of the glycerol

For the determination of glycerol ester contents, about 250 mg of biodiesel was transferred into a glass balloon and 5 ml methyl heptadecanoate - internal standard solution (C17; 10 mg/ml) - was added. Approximately 1.0 μl of the prepared solution was injected into a GC-14B Shimadzu gas chromatograph, equipped with a Flame Ionization Detector detector and a capillary column (DB-5MS, 30 m × 0.25 mm ID, THK 1µm, Agilent). Nitrogen gas was used as the mobile phase. The concentration of methyl esters in biodiesel samples was calculated using the GC data by the method presented in EN14103:2003(E).

Determination of the flash point

The flash point was determined according to ASTM D 93 which limits its minimum value to 100 °C. The equipment used was a FLASHPOINT TESTER - HFP339 - Pensky Martens.

2.4.2 Analysis of the Purified and Unpurified Magnesol®

Elemental Analysis

The total carbon percentage (%TC) and total nitrogen percentage (%TN) was performed analysing the samples in the TruSpec CHN non-dispersive elemental analyser, mark LECO®, equipped with an infrared detector for carbon and hydrogen, and a thermal conductivity detector for nitrogen. For the determination, a small amount of the sample was used, which ranged from 50 mg to 100 mg (± 0.01). The samples were weighed into thin sheets of tin and taken directly to the equipment. The samples were burned under a 6.0 oxygen (99.9999 %) atmosphere with 10 lpm flow and 950 ºC LECO® according to the ASTM D5291-16. The test was performed in triplicates (Ramos et al., 2015).

Field Emission Scanning Electron Microscopy (FEG-SEM)

The surface morphology of pure and recovered Magnesol®was studied using a FEG-SEM, FEI Inspect F50 equipment, and the samples were coated with a thin layer of gold. The dimensions were calculated using the program Image J (n = 20).

Fourier Transform Infrared Spectroscopy (FTIR)

FTIR analysis was performed using a Perkin Elmer Instruments Spectrum One FTIR Spectrometer, and spectrum acquisition was performed with the Attenuated Total Reflectance (ATR) sample attachment in the 650 to 4000 cm-1 wave number range.

Thermogravimetry analysis (TGA)

The thermal stability was evaluated by thermogravimetric analyses using the SDT- Q600 model (TA Instruments) in the range 25 - 800°C with a heating rate of 20 °C.min-1 under nitrogen. The weight of the sample was approximately 7.8 mg, and the analysis was carried out in triplicate.