Development of greener sample preparation methods for extraction and spectrometric determination of metals in selected fuel samples

Abstract

Crude oil is an unrefined petroleum which forms because of dead organisms buried under mud over a long period of time (million years). The mud is then converted to sedimentary rocks, which create intense heat and pressure, resulting in the formation of crude oil reservoir. However, crude oil contains trace elements which cannot be controlled as they occur naturally during crude oil formation. Some of these elements are unfavourable. For example, Cr, Fe and Ni can cause severe corrosion on refinery equipment. Additionally, Cd, Hg, Pb and As are associated with air pollution, while Ni, V, Pb, Pt and As are known to be catalyst poisoners during refinery process. The crude oil is then refined to form crude oil derivatives like gasoline, diesel, kerosene, just to name the few. Therefore, the challenges associated with metal ions in crude oil and crude oil derivatives have ignited an interest for many researchers to conduct investigations on the development of analytical methods for quantitative determination of metal ions in various fuel oils. However, most of the literature reported methods showed several limitations which include the use of toxic and costly reagents, long extraction time, high temperatures, etc. Therefore, this study aimed at developing greener and cost-effective sample preparation methods, followed by inductively coupled plasma-optical emission spectroscopy (ICP-OES) for the determination of metals and metalloids in crude oil and crude oil derivatives (gasoline, diesel and kerosene). The developed sample preparation methods were; a) microwave assisted-hydrogen peroxide digestion (MA-HPD), b) ionic liquid assisted-extraction induced by emulsion breaking (ILA-EIEB) and magnetic-solid phase extraction (m-SPE). For all the above mentioned sample preparation methods, multivariate optimization was used for the determination of the most influential parameters. During multivariate optimisation of MA-HPD it was observed that 245 ℃ microwave temperature, 25 minutes digestion time, 0.1 g sample mass and 5 M H2O2 were the optimum digestion conditions with accepted accuracy (104.8-117.7%) and precision (≤ 4.1%). The proposed MA-HPD method resulted in MDL of 0.046, 0.030, 0.408 and 0.057 μg/g for Ba, Na, Ni and V, respectively. The concentration levels of the selected metals (Al, Ba, Cd, Co, Cr, Cu, Mg, Na, Ni, Pb, Sb, Ti and V) ranged between 1.21-58.86 μg/g, 0.55-36.37 μg/g, 0.56-47.0 μg/g and 0.6-35.1 μg/g for crude-oil, diesel, kerosene and gasoline, respectively. The sensitivity, accuracy and precision of the MA-HPD method made it qualify to be an alternative digestion method for mineralization of fuel oils. Additionally, ILA-EIEB and the m-SPE were evaluated for the preconcentration of the selected elements (As, Ba, Cd, Cr, Cu, Mn, Mo, Ti, V, Pb, Sb, Sn, Tb, Te and Zn) that were in trace levels. The optimum conditions for ILA-EIEB were found to be 0.035 % for 1-Ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl), 18% for nitric acid, 15% for Triton X-100 and 0.1 g for sample mass. The emulsions were broken by heating at a controlled water bath at 80 ±2 °C for 30 ±4 minutes and a further centrifugation step was performed for 15 minutes at 3 500 rpm. The optimum conditions were able to give good accuracy (80.1-101%) and precision (1.9-4.7 %). This method was also able to report very low MDL for Ba, Na, Ni and V which were 0.107, 0.013, 3.494 and 0.560 μg/g, respectively. The concentrations of As (0.084- 0.46 μg/g) reported in this study ware in line with other literature reports. Alternatively, m-SPE was also used for the preconcentration of selected metals in fuel oils. The Fe3O4@Al2O3 nanoparticles were used as adsorbents and their formation was confirmed by various characterization techniques (FT-IR, SEM-EDX, TEM, XRD and UV-Vis). The two level fractional factorial design (FrFD) and the central composite design (CCD) resulted in optimum conditions of 40 mg adsorbent mass, 35 minutes sonication time, 6.5 pH, 20 μg/L spike concentration and 1M of HNO3 eluent concentration. The optimised m-SPE was able to give good accuracy (74-96%), precision (0.9-4.8%) and MDL (0.114-0.62 μg/g). The optimised and validated m-SPE method was then applied in real fuel oil samples. All the investigated metal ions below 10 μg/g. The ILA-EIEB was compare with m-SPE, in terms of their sensitivity, accuracy and precision. The ILA-EIEB was more sensitive (0.013-3.494 μg/g), accurate (80.1-101.1%) and precise (1.9-4.7%) than m-SPE (0.114-0.62 μg/g, 74-96% and 0.9-4.8%, respectively). Therefore it can be concluded that the three sample preparation methods (MA-HPD, ILA-EIEB and m-SPE) were greener. This is because, dilute H2O2 coverts to water during MA-HPD, ionic liquids are environmentally friendly as compared to organic solvent when performing EIEB and Fe3O4@Al2O3 used as adsorbent in m-SPE is also environmental friendly and accelerates the separation process by the use of an external magnet.