Modeling and simulation of waste engine oil treatment process using solvent extraction and carbonized-clay adsorption methods

 Corresponding author: Peter Muwarure. Copyright © 2025 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution License 4.0. Modeling and simulation of waste engine oil treatment process using solvent extraction and carbonized-clay adsorption methods Juliet Ijeoma Ugwuanyi 1, Lucky Uyigue 1 and Peter O. Muwarure 2, * 1 Department of Chemical Engineering, University of Port Harcourt, Rivers State, Nigeria. 2 Centre for Gas, Refining and Petrochemical Engineering, University of Port Harcourt, Rivers State, Nigeria. Global Journal of Engineering and Technology Advances, 2025, 24(02), 119-130 Publication history: Received on 19 June 2025; revised on 05 August 2025; accepted on 07 August 2025 Article DOI: https://doi.org/10.30574/gjeta.2025.24.2.0238 Abstract There is the irresistible need to recover refined engine oil from waste engine oil through recycling procedures and treatment methods. To achieve this objective, waste engine oil samples were subjected to acid precipitation, carbonizedclay adsorption and solvent extraction methods. In it, 5000 ml of acid treated engine oil was recovered from the preliminary stage of treatment. Also, carbonized-clay adsorbent was produced from the admixture of activated charcoal from coconut husk and bentonite clay in specific ratios of 50, 100 and 200 %w/w. Actual waste engine oil treatments proceeded with the separate applications of carbonized-clay adsorption and solvent extraction methods on the basis of ratios of carbon-clay, adsorbent-oil and solvent-oil, while using experimental designs proposed by design expert version 7.0. Physicochemical property tests were conducted for waste and refined engine oil samples in order to check efficacies of treatments. The results showed increase in viscosity from 135cP to 185cP for waste engine oil to refined engine oil. Water and heavy metal contents in waste engine oil reduced after treatment, respectively from (1100 ppm, Cu (88.9 mg/l), Fe (322.8 mg/l) and Pb (12.4 mg/l)) to (40 ppm, Cu (8.5 mg/l), Fe (58.6 mg/l) and Pb (8.8 mg/l)). Optimal yields of refined engine oil were respectively 77.8 % and 76.5 %, for carbonized-clay adsorption and solvent extraction methods. Empirical models determined for the treatment processes also showed good correlation coefficients (R2) of 0.87 and 0.97, between experimental and predicted yields of refined engine oil for carbonized-clay adsorption and solvent extraction methods. Keywords: Waste Engine Oil; Simulation; Modeling; Solvent extraction; Carbonized-Clay 1. Introduction Engine oil reduces friction in internal combustion engines used in cars, warships, jet engines, tractors, electric generators, and other industrial devices. Reducing frictional force between engine parts reduce wear and tear which extends engine life [1, 2]. Overexposure to physical, chemical, and mechanical processes degrades it which causes physical qualities including colour, flash point, specific gravity, viscosity, cloud, and pour point to be lost. This engine oil would contain impurities and pollutants, rendering it spent or waste and unsuitable for engine lubrication [3]. As a hazardous waste and severe pollutant to soil and water, inappropriate waste engine oil disposal might harm ecosystems and human health. Waste motor oil recycling include precipitation, filtration, bleaching, distillation, solvent extraction, adsorption, and more. Solvent extraction is a potential technology for recovering waste motor oil since it efficiently separates useful lubricant components from impurities [4]. This procedure uses an extractive solvent to isolate the required oil fraction from lowgrade oils. In precipitation, precipitants are applied to waste engine oil to react and precipitate oil contaminants. Global Journal of Engineering and Technology Advances, 2025, 24(02), 119-130 120 Distillation removes volatile contaminants and recovers stable, usable oil from the bottom. Adsorption will transfer soluble and dispersed contaminants including volatiles, organic acids, and heavy metals from waste motor oil to the adsorbent, making the recovered oil more refined and reuseable. This technique dissolves oil in a solvent to remove contaminants [5]. The aim of this study is to carry out the modelling and simulation of waste engine oil treatment process using methods of solvent extraction and carbonized-clay adsorption. It would do so with the specific objectives are stated below as follows: collect waste engine oil samples from local mechanic workshops, including other raw materials relevant to the study; carbonize and activate charcoal obtained from coconut husk; select possible solvent for extracting refined engine oil from waste engine oil; subject the waste engine oil samples to both solvent extraction and carbonized–clay adsorption experiments; optimize the experimental process for the waste engine oil recycling using design expert as the simulation tool; and develop an empirical model for predicting the yield of the refined engine oil based on recycling process conditions. This study will to a large extent reduce the indiscriminate disposal of waste engine oil, which in turn reduces environmental pollution associated with waste engine oil. It will also support effective design of an engineering process plant for the mass production of refined engine oil from waste engine oil materials at an optimal cost. This would help build competence and capability into our local technology. 2. Material and methods 2.1. Materials and Reagent Waste engine oil was collected from different local mechanic workshops around the University of Port Harcourt neighbourhood, and stored in different cans. The SAE quality grades of the engine oil from where the waste engine oil samples were obtained were high performance SAE 20-50 grades. Coconut husk was obtained from de-husked coconuts at the local market near the University of Port Harcourt. These husks were sun dried, gathered together and stored in a sack bag. Other materials include bentonite clay (obtained from Chuz chemicals). Chemical reagents and solvents used for the study are also presented as Sulphuric acid (65 % w/w, MandB), Ethanol (95+ %w/w, Aldrich), NaOH (65%w/w, Fischer), Distilled water (pure state, Uniport), CaCl2 (95 %w/w, Aldrich) and CaO (high purity, BulkCem). 2.2. Equipment Relevant equipment used in the laboratory for carrying out this study and their respective functions are presented in Table 1. However, equipment not mentioned in the list will be explained in the appropriate sections. Table 1 List of equipment and their functions Equipment Function Hydrometer Density and specific gravity measurement Viscometer (Rotational and capillary glass type) Viscosity measurement Furnace/Oven Measurement of water and ash content Tubular furnace Carbon production and test Electric heater Heating device Vacuum pump Vacuum distillation Mixer Mixing and agitation Atomic Absorption Spectrophotometer, AAS. Detecting and measuring concentrations of metals in liquids. Distillation column and Condenser For distilled vapours and cooling. Shaker Particle size measurement. Centrifuge Oil separation from clay Thermometer Measuring temperatures Global Journal of Engineering and Technology Advances, 2025, 24(02), 119-130 121 Glass beakers For holding samples and other laboratory uses. Graduated cylinder Sample volume measurement Crucibles For holding solid samples include ash and charcoal. Tong Holding hot beakers Hand gloves Personal protection Nose masks Personal protection Safety cloth and coverall Personal protection 2.3. Preparation of Materials This section outlines the methods used to carry out the preparation of materials required for the study: acidification of waste engine oil, carbonization of coconut husk and activation of carbonized bentonite clay (or carbonized-clay). 2.3.1. Acidification of waste engine oil This is a preliminary treatment for the waste engine oil sample, first level impurities were removed by chemical precipitation and de-asphalting. In it, 20 ml of H2SO4 (65%w/w) was added into a 250 ml beaker containing 150 ml waste engine oil, amidst vigorous agitation with the aid of a mechanical stirrer or mixer operating at 360 rpm for a period of 1 h. The content was then allowed to rest for another 6 h, during which precipitates were observed to have settled at beaker bottom. Thereafter, the acidified waste engine oil was washed with 10 ml of NaOH (65%w/w), then filtered to recover the filtrate and discards the sediments. This procedure was repeated several times in order to produce about 5000 ml of acid treated waste engine oil. 2.3.2. Carbonization of coconut husk The sun-dried coconut husk was chopped into pieces, weighed and inserted into a tubular furnace, which was fired with the aid of combustion in excess air. The burning was monitored for a period of 2 h until the entire fibers were completely burnt into charcoal. The charcoal was further crushed into fines and sieved with the aid of a shaker to recover charcoal of size range: 0.5 mm to 3 mm from the un-carbonized chaff. 2.3.3. Activation of carbonized-clay Mixtures of bentonite clay and charcoal from coconut husk were prepared in the ratios of clay to charcoal of 2:1, 1:1 and 1:2 to form carbonized-clay adsorbent samples A, B and C respectively. 100 ml of H2SO4 (65% w/w) and 10 mg of CaCl2 (95 %w/w) were both added to each 500 mg of the carbonized-clay sample, amidst thorough mixing and mashing. The samples were then slowly dried in an oven at 110oC for 15mins. After cooling, they were washed thoroughly with distilled water, in order to remove its acid content. The washed carbonizedclay samples were again oven-dried at 90oC to remove residual moisture, and followed by pulverization of the activated carbonized-clay into fine particulates. 2.3.4. Characterization of engine oil sample The methods of characterization presented here were used for both the waste and refined engine oil samples. Physicochemical properties considered include: Density, viscosity, pour point, cloud point, flash point, acidity or neutralization number and ash content. Others were coking test, water content and metal content. Density: A hydrometer was used for the density measurement. In it, a given volume of oil sample was measured into a graduated cylinder, wherein a hydrometer was inserted into the cylinder from the top. The device was allowed to float in the oil until the point when oil level in cylinder coincides with the hydrometer instrument. That point is read as the specific gravity or density of the oil sample. Viscosity: A rotational viscometer (equipped with digital meter and memory device) was used for the measurement. Oil sample at 40°C was measured into a cup, while the viscometer spindle at 600 rpm was inserted into the oil cup followed by shearing. After a few seconds, a stable digital value of dynamic viscosity was directly read from the instrument digital meter. This procedure was repeated thrice; while a mean value was taken from the triplicate. Global Journal of Engineering and Technology Advances, 2025, 24(02), 119-130 122 Alternatively, a capillary glass viscometer was also used for the measurement. A ball bearing of 5 mm diameter was dropped into the capillary glass tube containing engine oil, and with the aid of stop-watch, the time of fall from top to bottom of glass tube was measured. This procedure was also repeated thrice in order to obtain a mean time of fall, t. The Dynamic viscosity was estimated using equation 1. Also, the kinematic viscosity, υ is obtained by dividing the dynamic viscosity, µ by engine oil density. µ=𝐾(𝜌𝑏−𝜌)𝑡 ………………….. (1) Where; μ = viscosity (cP); ρb = density of ball bearing; ρ = density of engine oil; Sample; t = mean time of fall (s); K = instrument constant. Viscosity Index: This parameter measures the extent at which fluid viscosity is affected by temperature change. The higher the viscosity index, VI, the more will engine oil viscosity decrease as temperature increases, and vice versa. Standard procedure according to ASTM D-2270 was followed for the measurement. Thus, at standard temperature condition of 40 oC, two reference oils viscosities of between 0 and 100 were selected from table, and named L and H. The viscosity of test oil sample was measured as U. Use equation 2 to estimate the viscosity index (VI) of the test oil sample: 𝑉𝐼=(𝐿−𝑈) (𝐿−𝐻)×100 ………………….. (2) Flash Point: This is the ignition temperature of oil vapour. The flash point of oil was determined by heating it at constant pressure until it produced enough vapour to make an ignitable combination with air. Since test engine oil was in an open cup, its temperature was progressively elevated and monitored using a thermometer. A test flame was swept over the cup using matches at prescribed intervals until the test oil sample vapour ignited. The flash point is the temperature at which fire sparks. Neutralization Number: The test for Neutralization Number or Total Acid Number (TAN) of the used oil sample was conducted using ASTM D-974. It is the quantity in milligrams of potassium hydroxide (KOH) per gram of oil necessary to neutralize acidity. Two grams of used oil sample was weighed and mixed with 100 mL of the titration solvent (toluene and isopropyl alcohol containing a small amount of water) and 0.5 mL of the indicator solution (p-naphtholbenzein) and swirled until the sample was entirely dissolved by the solvent. The mixture assumes a yellow-orange colour and titrated with 0.1 M KOH solution in increments and mixed vigorously near the end point i.e. green colour; to observe the end point of dark-coloured oil, the flask is shaken vigorously to produce momentarily slight foam and the colour change occurs under a white fluorescent. The neutralization number or Total Acid Number (TAN) is calculated as; 𝑇𝐴𝑁(𝑚𝑔 𝐾𝑂𝐻 𝑔 𝑠𝑎𝑚𝑝𝑙𝑒)=[(𝐴−𝐵)×𝑀×56] 𝑊 ………………….. (3) Where; A = KOH solution required for titration of the sample, mL, B = KOH solution required for titration of the blank, mL, M = Molarity of the KOH solution, and W = mass of sample used, g. Coking Test: The test followed the ASTM D524 standard for coking test measurement. It is the amount of carbon residue obtained as oil is heated to a high temperature in the absence of air is called coking test. In it, 3 g of used oil is measured on a crucible and put into a tubular stainless-steel apparatus which was inserted into a tubular furnace. The apparatus was then flashed with N2 gas so as to make the environment inside inert. The test is conducted at 550°C for thirty minutes. Finally, the remaining residue is measured to calculate its percentage of the initial sample. % 𝐶𝑎𝑟𝑏𝑜𝑛 𝑅𝑒𝑠𝑖𝑑𝑢𝑒 = 𝑚(𝑐𝑎𝑟𝑏𝑜𝑛 𝑟𝑒𝑠𝑖𝑑𝑢𝑒) 𝑚(𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑠𝑎𝑚𝑝𝑙𝑒) ×100 ………………….. (4) Ash Content: ASTM D-482 standard was followed. The ash content is the measure of the amount of incombustible material present in a lubricant which can be determined by measuring the amount of ash after combustion of oil in a furnace. In it, measured amount of oil sample is put in a crucible and burned for 5 h in the furnace at 800°C. Mass of the remaining ash was measured and its percentage is calculated by dividing it with initial mass of the sample. % 𝐴𝑠ℎ= 𝑚𝑎𝑠𝑠 𝑜𝑓 𝑎𝑠ℎ 𝑚𝑎𝑠𝑠 𝑜𝑓 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑠𝑎𝑚𝑝𝑙𝑒 ×100 ………………….. (5) Global Journal of Engineering and Technology Advances, 2025, 24(02), 119-130 123 Water Content: This is the amount of water present in the lubricant. It is estimated by putting a measured amount of sample in an oven at 120°C for 1 h. The dehydrated sample is then reweighed to calculate the loss. % 𝑊𝑎𝑡𝑒𝑟= (𝑚𝑎𝑠𝑠 𝑜𝑓 𝑖𝑛𝑖𝑡𝑖𝑎𝑙−𝑚𝑎𝑠𝑠 𝑜𝑓 𝑓𝑖𝑛𝑎𝑙) 𝑠𝑎𝑚𝑝𝑙𝑒 𝑚𝑎𝑠𝑠 𝑜𝑓 𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑠𝑎𝑚𝑝𝑙𝑒 × 100 ………………….. (6) Heavy Metal Content: A fast sequential atomic absorption spectrophotometer (AAS) was used for the measurement according ASTM D-4628-2 standard. Before the analysis, the oil sample was heated to 60 °C and stirred to ensure homogeneity of the sample, it was then mixed with ten volumes of kerosene. Sets of standardized organo-metals from (Cu, Fe, Pb) and 4-cyclohexylbutyric acid salts were prepared, and metal concentrations were determined by introducing the test solutions of engine oil samples into the flame of the AAS, and recording the responses. Metal concentrations were determined from the calibration curve that is obtained from standard solutions. 2.4. Design of Experiment This study made use of design expert to in order accommodate the influence of the experimental variables (or factors) and design factor levels on the number of experimental runs for each method being studied: carbonized-clay adsorption and solvent extraction. The experimental variables considered were defined as follows: Carbon-clay ratio: It is the ratio of amount of carbon obtained from coconut husk to the amount of bentonite clay in the adsorbent mixture. The ratio is fixed at 1:2, 1:1 and 2:1. Mixing or blending the adsorbents will increase adsorption capacity and performance of adsorption of the waste engine oil treatment. Adsorbent ratio: It is the ratio of amount of carbonized-clay (adsorbent) to the amount of waste engine oil being treated. The ratio is fixed at 3:20, 1:5 and1:4. The adsorbent drives the removal of metals and organic impurities from the waste engine oil samples. Solvent ratio: It is the ratio of quantity of solvent applied to the quantity of waste engine oil being treated. The ratio is fixed at 2:1 to 11:1. The solvent will create the required surface area for extraction. Therefore, the proposed matrices for the experimental design for both solvent extraction and carbonized-clay methods are presented as shown in Tables 2 and 3. Table 2 Experimental design for solvent extraction treatment Solvent Ratio (v/v) Ethanol 2 Run 1 4 Run 2 5 Run 3 6 Run 4 7 Run 5 8 Run 6 9 Run 7 10 Run 8 11 Run 9 Table 3 Experimental design for carbonized-clay treatment Carbon-clay ratio (%w/w) Adsorbent-oil ratio (%w/v) 15 20 25 50 Run 1 Run 2 Run 3 100 Run 4 Run 5 Run 6 200 Run 7 Run 8 Run 9 Global Journal of Engineering and Technology Advances, 2025, 24(02), 119-130 124 2.5. Experimental Treatment of Waste Engine Oil 2.5.1. Carbonized-clay adsorption Treatment In line with specifications from design expert, actual quantities of adsorbent-oil ratios (based on 15, 20 and 25%w/v) and carbon-clay ratios in the adsorbent (based on 50, 100 and 200 %w/w) were prepared and administered for the experimental runs as shown in Table 3. For run 1, 200 ml of acid treated waste engine-oil sample was dosed with 30 mg of carbonized-clay adsorbent (ratio 50%w/w or 1:2) amidst vigorous mixing in a 250 ml beaker at 250 rpm. The content was allowed to rest for another 8 h, after which it was filtered to recover the engine oil as filtrate. The filter medium was a filter cloth which squeezes out residue engine oil. The yield was recorded, while AAS was used to check the residual metal concentrations. This procedure was repeated for other runs in line with design expert specifications. Optimal yield of engine oil from the adsorption treatment was also measured. 2.5.2. Solvent Extraction High purity ethanol was used as solvent. The solvent-oil ratio was followed as specified for in the respective runs in the design expert. A soxhlet extractor was used for the extraction. 100 ml of acid treated waste engine oil was used as treatment quantity for all the runs. For run 1, 100 ml waste engine oil was extracted with 200 ml ethanol solvent. With a heating mantle base, the soxhlet extractor operated at constant temperature of 60 oC amidst agitation for 1 h contact time. The extract was collected at extractor overhead via a condenser unit. The solvent was recovered by distillation. The quantity of oil recovered was measured as yield of extraction in millilitre volume. This procedure was repeated for all the runs. 2.6. Response Surface Modelling In the RSM modelling environment (MINITAB), the procedure is to enter into the worksheet with a single column of measurement for the response variable (i.e. yield of engine oil), and also enter a column of data for each predictor variable (i.e. carbonized-clay, adsorbent-oil and solvent-oil ratios respectively). Then follow the stat > regression > regression > fit regression model path, and click model in order to specify the interaction and polynomial terms. Then enter your response and predictor variables including other terms of the model. Thus, fit regression model can be categorical and continuous predictors. Under predictor, press control (ctrl) key and select carbonized-clay ratio, X1 and adsorbent-oil ratio, X2, or solvent-oil ratio, X3, for solvent extraction method. Using interactions through order, choose 2 for carbonized-clay adsorption method and 1 for solvent extraction method; click add, and then click ok in order to output the model expression and the related statistics. Based on a full factorial design, second order model in quadratic response format was developed for carbonized-clay adsorption method of treatment, while a first order model in linear response format was developed for solvent extraction method of treatment. These proposed equations are shown in equations 7 and 8. 𝑌=𝑎0+∑𝑎𝑖𝑋1 𝑛 𝑖=1 +∑𝑏𝑗𝑋2 𝑛 𝑗=1 +∑𝑎𝑖𝑖𝑋1 2 𝑛 𝑖=1 ………………….. 7) and 𝑌=𝑝0+∑𝑝𝑖𝑋3 𝑛 𝑖=1 ………………….. (8) Where, Y = response variable; X1, X2, X3 = Predictor variables; a0, ai, aii, bj and p0, pi = constant coefficients. 2.6.1. Goodness-of-fit The accuracy of the generated models is determined by method of goodness-of-fit. The R2 determine the degree of correlation between experimental and predicted data, the percentage deviation (p) measures the level of variance between experimental and predicted results. Standard error of estimate (SE) measures the error associated with the measurement. Analysis of variance (ANOVA) determines the degree of regression between the variables. If R2 = 1, it means all plotted data aligns to form a smooth curve, if R2< 1, it means some of the plotted points are out of the curve, but if R2 = 0, means there is no correlation or a case of highly scattered data is evident. For this study, a percentage deviation of 5% and below shall be considered as insignificant difference. Global Journal of Engineering and Technology Advances, 2025, 24(02), 119-130 125 Coefficient of Correlation (R2) Coefficient of correlation for a regression model can be determined using the spearman product moment correlation coefficient (SPMCC) for the engine oil treatment process. The formula for SPMCC equation is given as shown in equation 9. 𝑅2=𝑃𝑌𝑒𝑌𝑝=𝑁(∑𝑌𝑒𝑌𝑝)−(∑𝑌𝑒∑𝑌𝑝) √(𝑁(∑𝑌𝑒2)−(∑𝑌𝑒)2)−((𝑁∑𝑌𝑝 2)−(∑𝑌𝑝 2)) ………………….. (9) Where, Ye = experimental yield of refined-engine-oil (%); Yp = predicted yield of refined-engine-oil (%) and N = number of runs. Percentage Deviation (D) Percentage deviation, D measures the accuracy associated with the prediction from the empirical model. In this case, the accuracy of the predicted yield of refined-engine-oil compared to its experimental yield is being determined. Percentage deviation is estimated using equation 10. % 𝐷=(𝑌𝑃−𝑌𝑒 𝑌𝑒) × 100 1 ………………….. (10) Where, Yp = predicted yield of refined-engine-oil; Ye = experimental yield of refined-engine oil. 3. Results 3.1. Experimental results The experimental results obtained from this study are presented in Tables and Figures as follows. Table 4 Physicochemical properties of waste and refined engine oil samples Properties Waste Engine Oil Refined Engine Oil (with carbonized-clay adsorption) Refine Engine Oil (with solvent extraction) Flash point 210 225 220 Dynamic viscosity @ 600 rpm at 40o C (cP) 135 195 185 Kinematic viscosity @ 40 oC (cSt) 160.34 222.7 211.2 Cloud point >15 >15 >15 Pour point, oC >15 >15 >15 Specific gravity @ 40 oC 0.90 0.876 0.876 Water content. ppm 1100 84.4 40 Ash content, % 1.8 0.4 0.45 Heavy metals: Cu (mg/l) Fe (mg/l) Pb (mg/l) 88.9 322.8 12.4 3.5 53.6 7.2 8.5 58.6 8.8 Global Journal of Engineering and Technology Advances, 2025, 24(02), 119-130 126 Table 5 Results from using carbonized-clay adsorption method Experimental run Carbon-clay ratio (%) Adsorbent-oil ratio (%) Yield of engine oil (%) 1 50 15 77.8 2 50 20 74.1 3 50 25 70.8 4 100 15 69 5 100 20 65.2 6 100 25 65.1 7 200 15 61.4 8 200 20 60.0 9 200 25 55.8 Table 6 Results of solvent extraction on waste engine oil Run Solvent to Used-engine-oil ratio Yield of engine oil (%) 1 2 52.2 2 4 56.5 3 5 59.7 4 6 61.9 5 7 63.3 6 8 64.8 7 9 72.5 8 10 74.3 9 11 76.5 Global Journal of Engineering and Technology Advances, 2025, 24(02), 119-130 127 Figure 1 Graph of yield of refined engine oil versus carbonized-clay and adsorbent-oil ratios Figure 2 Graph of yield of refined engine oil versus solvent-oil ratio 3.2. Regression models for waste engine oil treatment Results of the regression models from RSM method for adsorption and solvent extraction treatments of waste engine oil are presented as follows: Adsorption treatment model: 𝑌=60.9+0.45𝑋1−0.08𝑋2−0.004𝑋1 2 (𝑅2=0.87) ………………….. (11) Solvent Extraction model: 𝑌=49.7+2.985𝑋3 (𝑅2=0.97) ………………….. (12)