Process Safety and Environmental Protection 136 (2020) 334–342 Contents lists available at ScienceDirect Process Safety and Environmental Protection journal homepage: www.elsevier.com/locate/psep Employing sodium hydroxide in desulfurization of the actual heavy crude oil: Theoretical optimization and experimental evaluation Yusra A. Abd Al-Khodor, Talib M. Albayati ∗ Department of Chemical Engineering, University of Technology, 52 Alsinaa St., PO Box 35010, Baghdad, Iraq a r t i c l e i n f o Article history: Received 25 August 2019 Received in revised form 18 January 2020 Accepted 27 January 2020 Available online 8 February 2020 Keywords: Desulfurization Response surface methodology Actual heavy crude oil Environment Optimization Sodium hydroxide a b s t r a c t The desulfurization of the actual heavy crude oil is one of the most important processes in petroleum industries due to the low quality of those types of oil and containing large amounts of sulfur compounds, high viscosity and density. In the present work, the desulfurization of the actual heavy crude oil with a sulfur content 5.8 wt. % from Al-Halfaya Oil Field in southern Iraq was studied using a sodium hydroxideassisted process. Effects of the operating conditions such as: reaction time (30–60 min), temperature (30–50 ◦ C), the amount of NaOH in its solution (10–30 gm), and mixing speed (300–500 rpm) were investigated. The desulfurization process was achieved in a batch reactor by implementation of the experimental design technique. The objective function (response) was the sulfur content wt. % while a response surface method (RSM) was applied to define the significant factors that affect the desulfurization process. It was found that effects of the four variables take the following sequence: mixing speed > weight of NaOH > time > temperature. The optimum conditions of the proposed model were obtained using optimization techniques and found as follows: time = 60 min., temperature = 40 ◦ C, NaOH solution = 18gm and mixing speed = 500 rpm. The optimum conditions of the sulfur content were applied experimentally and theoretically was equal to 2.5 and 2.3 wt. %, respectively. It is concluded that the efficiency of the sulfur removal content for actual heavy crude oil by this process was 56.89 %. © 2020 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved. 1. Introduction Environmental pollution problems caused by exhaust emissions have received increasing attention worldwide with the development of society. Sulfur-containing compounds can be converted to sulfur oxides during the combustion process which cause serious damages to the environment. Many countries have adopted more stringent environmental regulations to restrict the sulfur level of fuels limiting the sulfur level to less than 10 ppm (Eßer et al., 2004; Stanislaus et al., 2010; Li et al., 2018). The quality of crude oil depends mainly on the sulfur content and API gravity. The characteristics of crude oil vary according to the geographical location of the crude oil reservoir. The quality of crude oil produced worldwide suffers from high sulfur content which has encouraged many large studies by the industrial or academic world to try to reduce sulfur content (Bridge, 2008; Albayati and Kalash, 2020; Chong et al., 2015). Processing of crude oil with high sulfur concentration has become the research focus around the world because the equipment in oil refineries cannot handle high sulfur ∗ Corresponding author. E-mail address: 80046@uotechnology.edu.iq (T.M. Albayati). concentration crude oil through oil refining operations. Increasing sulfur compounds in the crude oil also leads to increase the sulfur compounds in its products. Therefore, reducing sulfur compounds from crude oil has become an urgent task to meet clean fuel production needs. The study of the new sulfur removal technology and improvement of sulfur removal processes are key factors for greater profits for oil refineries (Babich and Moulijn, 2003; Lin et al., 2010; Srikanth et al., 2018; Ugal et al., 2018). Many components must be removed from the crude oil and its products before shipping to the market. Sulfur compounds that form in oil and its products such as hydrogen sulfide are very corrosive and extremely toxic. So, many refineries worldwide are using a variety of methods to reduce the concentration of sulfur in crude oil (Breysse et al., 2003; Song, 2003; Javadli and De Klerk, 2012; Speight, 2015; Lee and Valla, 2017). It is known that several methods are applied to reduce the sulfur in crude oil, such as Oxidative desulfurization, Adsorptive desulfurization, Desulfurization by extraction, Hydro desulfurization, Alkylation desulfurization, Bio desulfurization, Chlorinolysis-based desulfurization, Radiation desulfurization, Supercritical water desulfurization and caustic desulfurization (Sodium hydroxide process) (Javadli and De Klerk, 2012). Sodium hydroxide (NaOH solution) is one of the desulfurization processes used and able to remove most H2 S, light mercaptans https://doi.org/10.1016/j.psep.2020.01.036 0957-5820/© 2020 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved. 335 Y.A. Abd Al-Khodor, T.M. Albayati / Process Safety and Environmental Protection 136 (2020) 334–342 and thiophenols from the sour crude oil (Manihe and Ghorayeb, 1981). This method continues to attract much attention due to low cost and easy to operate, but it is not able to remove heavy mercaptans and polycyclic sulfur compounds (Lin et al., 2010; Moaseri et al., 2013). Shakirullah et al. investigated desulfurization of kerosene, heavy residue, commercial furnace oil and diesel oil with sodium hydroxide solution. The desulfurization efficiency of the kerosene, diesel oil, heavy residue and commercial furnace oil reached to 60, 68, 70 and 71 % respectively (Shakirullah et al., 2010; Sharma et al., 2012). Jeyajothi used NaOH solution to study the effect of desulfurization of the sour crude oil in the petrochemical industry. The sulfur content of crude oil was reduced from 700 ppm to less than 130 ppm (Jeyajothi, 2015). Response Surface Method (RSM) is a highly significant application in Chemometrics (usage of mathematics or statistical methods with chemical data), since additional data were required about chemicals’ behavior through the processes or system (Alkafajy and Albayati, 2020). Central Composite Design (CCD) was used for the design of the experiment to identify the factors that affect the desulfurization process and find the optimum conditions that give the best sulfur removal from the actual heavy crude oil. The sodium hydroxide process was used to treat petroleum products but very few studies have assessed the performance of this method with the actual heavy crude oil. In the fact that most of the previous research refer to petroleum products rather than crude oil. Thus, the objective of this work is the treatment of actual heavy crude oil with high sulfur content from Al-Halfaya Oil Field in southern Iraq using a sodium hydroxide-assisted process. Effects of the various operating conditions were investigated such as: reaction time (30–60 min), reaction temperature (30–50 ◦ C), weight of NaOH (10–30 gm), and mixing speed (300–500 rpm). The desulfurization process was performed in a batch auto cleave reactor by the application of the experimental technique while the optimization of the applied factors (variables) was adopted during the desulfurization process using CCD. 2. Materials and methods 2.1. Chemicals In the experiments, the samples of crude oil were obtained from Al-Halfaya Oil Field in southern Iraq. The chemicals included sodium hydroxide (NaOH, 95 % purity) and Toluene (C6 H5 CH3 , 99.5 % purity) which was used for the purpose of washing the glasses. All chemicals were purchased from Sigma Aldrich and used as received without any treatment. 2.2. Analysis method The sulfur content of the treated and untreated heavy crude oil samples was measured by X-ray fluorescence analysis according to ASTM d-4294 made by Horiba Company, USA. This device is equipped with a spectra membrane that typically supports the thin film of a frame-mounted material that acts as a carrier. 2.3. Experimental design The experiments were designed using Design-Expert 11 software where four factors were selected as investigated variables based on findings from the literatures. The desulfurization process was included the implementing sodium hydroxide solution with studied variables such as: time, temperature, weight of NaOH and mixing speed. The batch desulfurization experiments were performed to remove sulfur content from actual heavy crude oil obtained from Al-Halfaya Oil Field in the south of Iraq. Then, the experiments were performed to investigate the performance of Table 1 Factors used in the CCD and their levels. Range and levels Factors Time (min) o Temperature ( C) Concentration of NaOH (gm) Mixing speed (rpm) Low Level Central Level High Level 30 30 10 300 45 40 20 400 60 50 30 500 Table 2 Experiments of four variables central composite design (CCD). Run No. Operating parameter o Time (min) Temperature ( C) Weight (gm) Speed (rpm) 45 60 45 30 30 45 30 60 60 30 60 45 60 45 60 45 30 30 45 45 30 60 45 30 60 45 45 60 45 30 40 50 40 30 30 40 40 30 50 50 50 40 50 40 30 40 50 50 40 50 50 30 40 30 30 40 30 40 40 30 20 30 20 10 30 20 20 30 30 30 10 10 10 20 10 20 30 10 20 20 10 10 30 10 30 20 20 20 20 30 300 300 500 500 500 400 400 500 500 300 300 400 500 400 300 400 500 300 400 400 500 500 400 300 300 400 400 400 400 300 sodium hydroxide solution with the optimization of the experimental factors using CCD. The optimization study using response surface methodology (RSM) was used to determine the significant factors affecting the experiment and reducing the number of experiments. The illustration of the CCD is given in Table 1, which shows a low, central, and high level. Three levels factorial CCD (face centered) were used for the four factors with 25 non-center points, 5 center points and a total of 30 runs, as shown in Table 2. 2.4. Desulfurization procedures The experimental procedure as shown in Fig. 1 started by taking 50 ml from actual heavy crude oil in a first flask. The solution of NaOH was prepared in different molarities by taking 10, 20 and 30 gm from NaOH in the second flask, in order to make different concentration by adding a constant volume of water for each weight of NaOH at 10, 20 and 30 gm. The two-flask solution was added into the third batch, the auto cleaves reactor flask, in a 1:1 mass ratio, and the reactor contents were stirred for 30, 45 and 60 min to get a good contact between the phases. Temperatures of 30, 40 and 50 ◦ C and mixing speeds of 300, 400 and 500 rpm were set up using the hot plate stirrer. After mixing, a funnel was used to separate the solvent from the bulk of actual heavy crude oil. The treated samples were measured by X-ray fluorescence according to ASTM d-4294 made by Horiba Company, USA. 336 Y.A. Abd Al-Khodor, T.M. Albayati / Process Safety and Environmental Protection 136 (2020) 334–342 Fig. 1. Schematic diagram of the experimental setup. Table 3 Results test of actual crude oil. No. Test Unit Results Test method Density @ 20/20 Specific gravity API Salt Water cut Sediment Water cut & Sediment Sulfur Sulfur Pour point Freezing point DI Hydrogen sulfide H2S (vapor) Flash point Fire point Dynamic viscosity @ 28 ◦ C Dynamic viscosity @ 50 ◦ C Asphaltene gm./cm3 --------ppm ppm ppm ppm ppm % o C o C ppm o C o C CP CP ppm 0.91566 0.91657 22.88 17.20 0.1 0.00 0.1 58202 5.8015 −37 −42 400 120 140 94.8 24.65 5.7681 ASTM D 5002 ASTM D 5002 ASTM D 5002 ASTM D ASTM D 4006 ASTM D 4006 ASTM D 4006 ASTM D 4294 ASTM D 4294 ASTM D 97 ASTM D 97 ----ASTM D 93 ASTM D 93 ASTM D 7042 ASTM D 7042 ASTM D 3279 ASTM D 6560 IP 143 3. Results and discussion 3.1. Heavy crude oil analysis The sample of actual heavy crude oil used in this experiment was tested in the laboratory of Al-Halfaya Oil Field. As shown in Table 3., the crude oil used was heavy crude oil because of its API = 22.8 and specific gravity (sp.gr.) = 0.91657 with high sulfur content =5.8015 wt.% or 58,202 ppm. 3.2. Interpretation of regression analysis The experiments carried out using Design-Expert are shown in Table 4. These indicated a decrease in the sulfur content with increases in the contact time, temperature, NaOH concentration and mixing speed. The final equation in terms of coded factors was as follows: R = 4.58 − (0.36*A) − (0.068*B) − (0.427*C) − (0.64*D) + (0.14*A*B) − (0.19*A*C) − (0.086*A*D) + (0.010*B*C) + (0.089*B*D) + (0.24*C*D) − (1.15*A2 ) + (0.33*B2 ) + (0.43*C 2 ) − (0.16*D2 ) Where: o R = sulfur content (wt. %); A = Time (min.); B = Temperature ( C); C = Weight of NaOH (gm); D = Speed (rpm). Two variables indicate an interaction effect, while a 2nd order term of a variable denotes the square effect. 3.3. Anova An ANOVA test was used to define the importance of each variable in the designed experimental research. The ANOVA is checked from the user’s point of the view with a focus on assessing the model’s construction and suppositions underlying the method. It is recommended to use graphic means to envisage the ANOVA model as well as to analyze data. The main models of ANOVA have been expanding in some detail, involving one-factor ANOVA, cross-design, interaction designs, repeated ANOVA measures, and estimation of variance components (Zarei et al., 2010; Khataee et al., 2010). The comparison is made by the value of F, which is the ratio of the average form the box and the remaining error. The value of F should be greater than the one specified for distribution F if the model is to be considered as an ideal indicator of the results of the pilot program. As shown in Table 5, the value of model F of 23.17 indicates that the model is important. The P value is used in order to assess whether the F distribution is statistically significant, when P is less than 0.05. The L̈ack of Fit F-valueöf 1.92 implies the Lack of Fit is significant. The R-square and R-square (adj.) provides information on whether the program is accurate in 337 Y.A. Abd Al-Khodor, T.M. Albayati / Process Safety and Environmental Protection 136 (2020) 334–342 Table 4 The experiments of four variables central composite design (CCD) with response (Total sulfur content). Run No. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. Operating parameter o Time (min) Temperature ( C) Weight (gm.) Speed (rpm) 45 60 45 30 30 45 30 60 60 30 60 45 60 45 60 45 30 30 45 45 30 60 45 30 60 45 45 60 45 30 40 50 40 30 30 40 40 30 50 50 50 40 50 40 30 40 50 50 40 50 50 30 40 30 30 40 30 40 40 30 20 30 20 10 30 20 20 30 30 30 10 10 10 20 10 20 30 10 20 20 10 10 30 10 30 20 20 20 20 30 300 300 500 500 500 400 400 500 500 300 300 400 500 400 300 400 500 300 400 400 500 500 400 300 300 400 400 400 400 300 Total sulfur content (actual) Total sulfur content (predicted) 5.1136 4.1122 3.8662 3.6886 4.5360 4.8460 3.7724 2.6542 3.0611 4.5350 5.0143 4.9927 2.7589 4.4430 4.4642 4.3618 4.1480 5.1580 4.4655 4.7600 3.0007 3.2022 5.1560 5.0750 3.9113 4.3811 5.1939 3.2305 4.5520 5.0687 5.060 4.125 3.781 3.727 4.551 4.578 3.796 2.987 2.970 4.788 4.983 5.013 2.875 4.578 4.684 4.578 3.979 4.876 4.578 4.839 3.113 2.933 4.996 5.132 3.784 4.578 4.976 3.068 4.578 5.003 Table 5 ANOVA for the sulfur content of desulfurization by Sodium hydroxide Process. Source Sum of Squares *DF Mean Square F-value p-value Prob > F Model A-A B-B C-C D-D AB AC AD BC BD CD A2 B2 C2 D2 Residual Lack of Fit Pure Error Cor Total 16.71 2.39 0.084 1.324E-003 7.37 0.31 0.59 0.12 1.720E-003 0.13 0.91 3.40 0.28 0.47 0.064 0.77 0.61 0.16 17.48 14 1 1 1 1 1 1 1 1 1 1 1 1 1 1 15 10 5 29 1.19 2.39 0.084 1.324E-003 7.37 0.31 0.59 0.12 1.720E-003 0.13 0.91 3.40 0.28 0.47 0.064 0.052 0.061 0.032 23.17 46.33 1.63 0.026 143.06 6.00 11.54 2.31 0.033 2.48 17.66 66.01 5.47 9.18 1.24 >0.0001 >0.0001 0.2215 0.8748 >0.0001 0.0271 0.0040 0.1491 0.8575 0.1361 0.0008 >0.0001 0.0336 0.0084 0.2824 Significant 1.92 0.2445 not significant Table 6 Coefficient of Determination (R2 ) for the sulfur content of desulfurization by Sodium hydroxide Process. Parameters R2 Adjusted-R2 Predicted-R2 Std. Dev. C.V.% 0.9558 0.9145 0.7323 0.23 5.34 the resulting predictions. Here, 0.9558 and 0.9145 were obtained for Square R and Square R (adj.) respectively (Table 6), indicating an acceptable level of suitability (goodness-of-fit). The coefficient of determination (R2 ) presented in Table 6 was used to check the fit of the models. The value of R2 was high (close to 1), qualified and reasonable with the quadratic model of empirical data. It is proposed that the predicted R2 must be less than 0.80 for the model to be considered to have a good fit (Ölmez, 2009). In this case, the value of the selection coefficient (R2 ) for sulfur content was 0.9558. The higher value of the adjusted-R2 , as explained by the model, adjusts the contrast ratio around the center, while the predictedR2 is a measure of the model’s ability to predict the value of the response, which was close to R2 . The difference between R2 -pred. and R2 -adjust. should be lower than 0.20 to ensure that the data or model are significant (Mook et al., 2016). In our model, the fact that the R2 is close to R2 -adjust. Indicates that the model is significant. The low standard deviation showed that the quadratic (square) model is the best option.The coefficient of variance (CV) can be 338 Y.A. Abd Al-Khodor, T.M. Albayati / Process Safety and Environmental Protection 136 (2020) 334–342 Fig. 2. Residual plots for the desulfurization by NaOH. defined as the standard error ratio with respect to the mean value of the intercepted response, known as the reproducibility of the model. A relatively low value of the CV signals good accuracy and reliability of the model (Kuehl and Kuehl, 2020). Checking the normal distribution of data is generally used as the normal probability (Montgomery, 2005). The summary of the ANOVA is displayed in Tables 7 and 8. The results showed how well the model was satisfied with the assumptions of ANOVA. 3.4. Interpreting the graphs of the process One way to test the normality of distribution is to measure the proximity of the points in the normal prospect scheme in a linear plot. The experimental design gives the normal prospect plot (as observed in Fig. 2). It is clear from this that the data are distributed normally. In analysis of data, the difference between empirical value and expected one is a significant component in the model interpretation. For a perfect model, the residuals must be distributed randomly and naturally. In Fig. 2, internally residuals the predicted values provided vs. residuals do not offer any particular pattern and are approximately normally distributed. The final graph of the residual run is balancing and centering near to zero with no clear boundaries in the run number. 3.5. Effect of variables Graphical interpretation of the model was carried out by 3dimensional response surface and 2-dimention contour plots. The 3-dimensionsal response surface plots sulfur content, as the dependent variables were plotted against two independent variables, while keeping the other variables constant. In addition, the 2-dimensional contour plots are similar to the 3-dimensionsal response surface plots, and can explain the effect of independent variables on the response. The 2D contour plots show the type of the Table 7 Actual total sulfur content wt.% and predicted total sulfur content wt.%. Run No. Total sulfur content (actual) wt.% Total sulfur content (predicted) wt.% 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 5.1136 4.1122 3.8662 3.6886 4.5360 4.8460 3.7724 2.6542 3.0611 4.5350 5.0143 4.9927 2.7589 4.4430 4.4642 4.3618 4.1480 5.1580 4.4655 4.7600 3.0007 3.2022 5.1560 5.0750 3.9113 4.3811 5.1939 3.2305 4.5520 5.0687 5.060 4.125 3.781 3.727 4.551 4.578 3.796 2.987 2.970 4.788 4.983 5.013 2.875 4.578 4.684 4.578 3.979 4.876 4.578 4.839 3.113 2.933 4.996 5.132 3.784 4.578 4.976 3.068 4.578 5.003 interaction between independent variables. If the shape is circular or a parallel plane, it suggests the absence of interaction between the variables, and an elliptical or curved shape indicates the pres- 339 Y.A. Abd Al-Khodor, T.M. Albayati / Process Safety and Environmental Protection 136 (2020) 334–342 Table 8 Comparison between this study and other studies. No. Method Feed with sulfur % Sulfur reduction Reference 1 Oxidation desulfurization with ultrasound and ionic liquid Oxidation extraction desulphurization Heavy Oil and diesel, 3.85 wt.% 95 % and 65 % Houda et al., 2018 Heavy crude oil, sample A 0.87 wt% and sample B 0.32 wt% Heavy oil and bitumen Sample A (29 %) and sample B (73 %) Haruna et al., 2018 60 % Demirbas, 2016 Heavy crude oil and diesel, 4.53 wt % Actual heavy crude oil, 5.8 76.1 % Adlakha et al., 2016 56.89 % This study 2 3 4 5 Supercritical water Desulfurization Bio-desulfurization Caustic desulfurization process Fig. 3. Effect of the reaction time and temperature on sulfur wt.%. Fig. 4. Effect of the reaction time and weight of NaOH on sulfur wt.%. ence of an interaction between the variables (Anupam et al., 2011; Noorimotlagh et al., 2014). 3.5.1. Effect of reaction time and temperature Fig. 3 shows the interactive effect of the reaction time and temperature at a constant weight of NaOH and mixing speed on sulfur content wt.% which is the percentage of sulfur in the product oil. As can be seen, the elliptical nature of the contour plots indicates an interaction between reaction time and temperature. It can be observed that the sulfur content decreased as the reaction time was changed from 30 to 60 min. This behavior occurs because increasing the mixing time increases the contact time between the unreacted sulfur and NaOH solution. Similar findings were reported by (Moaseri et al., 2013). We observed that the sulfur content decreased when subjected to treatment at 40 ◦ C. This behavior is due to the fact that the composition of the compounds is sensitive to temperature, as they degrade at relatively high degrees. This shows that there is a preferred temperature, depending on the nature of the material (Alaya, 2012). 3.5.2. Effect of reaction time and weight of sodium hydroxide Fig. 4 illustrates the interaction between the reaction time and weight of sodium hydroxide, with temperature and mixing speed held constant. An increase in the reaction time from 30 to 60 min and an increase in the weight of NaOH from 10 to 20 gm together cause a decrease in the sulfur content of the oil. A further increase in the weight of sodium used, from 20 to 30 gm, led to increase the sulfur content, because the reaction is exothermic, so heat will be emitted. It is known that NaOH is a strong base, therefore, it completely and fully disassociates in aqueous solution. The heat emitted as a result of mixing solid NaOH with H2 O is responsible for the increasing temperature, due to the decomposition of NaOH to Na+ and OH– ions. This phenomenon is very important, as NaOH Fig. 5. Effect of the reaction time and the mixing speed on sulfur wt.%. crystals act as a strong drying agent because they absorb moisture from the air easily (Mason et al., 2015). As mentioned previously, the increase in heat causes the degradation (dissociation) of the material. 3.5.3. Effect of reaction time and mixing speed Fig. 5 shows that the sulfur content decreased as reaction time and mixing speed were increased within the experimental range, while temperature and weight of NaOH were held constant. It can be observed that increasing the reaction time is known to increase the mixing speed from 300 to 500 rpm. This is mainly due to increased disturbance that improves prevalence which leads to more chance to diffuse the sodium hydroxide inside the actual 340 Y.A. Abd Al-Khodor, T.M. Albayati / Process Safety and Environmental Protection 136 (2020) 334–342 Fig. 6. Effect of the reaction temperature and weight of NaOH on sulfur wt.%. Fig. 8. Effect of weight of NaOH and the mixing speed on sulfur wt.%. 300 to 500 rpm. The elliptical nature of the contour plots indicates an interaction between temperatures and mixing speed. Increasing mixing speed to 500 led to the improvement of the removal process and similar results have been reported by (Jeyajothi, 2015). Fig. 7. Effect of the reaction temperature and the mixing speed on sulfur wt.%. heavy crude oil to reach sulfur compounds found in actual heavy crude oil (Jabbar, 2015). 3.5.4. Effect of temperature and weight of NaOH Fig. 6 illustrates the combined effect of adjustments to the temperature and weight of NaOH on the removal of sulfur content, with reaction time and mixing speed held constant. The figure illustrates that temperature and weight of sodium hydroxide show an inverse relationship in their combined effects. In practical experiments, the best removal results were due to low temperatures and increased weight of sodium hydroxide: temperature at 30 ◦ C and weight of sodium hydroxide at 30 gm resulted in a decrease in sulfur content to 2.6 wt. %. This was the lowest sulfur content achieved during this process. Similarly, increasing temperatures to 50 ◦ C and decreasing the weight of sodium hydroxide to 10 gm resulted in a reduction of the sulfur content to 2.7 wt.%. Therefore, low temperatures with much sodium hydroxide, or high temperatures with little sodium hydroxide, provide a good distribution of heat in the reaction because increasing the weight of sodium hydroxide increases the heat of reaction. NaOH is, a strong base, reacted at a very high speed in an exothermic reaction leading to an increase reaction rate which leads to change decomposition of sodium hydroxide and then cannot remove the sulfur compounds; therefore, cannot reduce sulfur content when increasing the reaction temperatures with an increase in weight of sodium hydroxide (Mason et al., 2015). 3.5.5. Effect of temperature and mixing speed Fig. 7 shows that the sulfur content decreased with increases in the temperature and mixing speed at the reaction time and weight of NaOH constant. The sulfur content decreased with an increase of temperature to 40 ◦ C and an increase in the mixing speed from 3.5.6. Effect weight of NaOH and mixing speed Fig. 8 indicates that the sulfur content (wt.%) is the percentage of sulfur in the product oil which decreased with increases in the weight of NaOH and mixing speed, while reaction time and temperature were held constant. The sulfur content decreased with an increase of the weight of NaOH to 20 gm and an increase in mixing speed from 300 to 500 rpm. The speed increase was the most influential factor in this process, according to the ANOVA analysis of the results as shown in the Table 5 and the value of the final equation constants in terms of coded factors. 3.6. Optimum condition In order to obtain the optimum operating condition, the aggregate results of the total sulfur content for each variable are shown in Fig. 9. The decrease in the sulfur content with time is shown in Fig. 9a. Total de-sulfurization is relatively slow at low contact times of 30–45 min., and then the removal is clearly increased when the time is increased to 60 min. A closer examination of Fig. 9b shows that a temperature of 40 ◦ C seems appropriate from an economical and technical perspective. Furthermore, 40 ◦ C is quite suitable in terms of safety. Fig. 9c clearly shows that the total efficacy of caustic process desulfurization does not vary significantly with increases in the weight of NaOH from 10 to 30 gm. In terms of the scientific considerations, 20 gm of NaOH solution is more convenient than higher weights. The total sulfur content decreases with an increase in mixing speed from 300 to 500 rpm, as shown in Fig. 9d Similar results have been reported by (Moaseri et al., 2013). The experimental results show that the mixing speed is the key variable in the caustic process desulfurization. On the other hand, time, weight of NaOH and temperature had negligible effects on the overall desulfurization performance. Optimization of conditions that help to determine the optimal factors will lead to an optimal response. In this study, the four factors were evaluated in order to obtain values that give minimum sulfur content. The essential objective is to discover the best operating conditions (in terms of time, temperature, weight of NaOH and mixing speed) that give the lowest value of sulfur content as a result of the sodium hydroxide desulfurization process. Obtained optimum operating conditions as shown in Fig. 10, in which 60 min, 40 ◦ C, 18 gm weight of NaOH and 500 rpm were the best conditions and are considered economically and practically beneficial, when Y.A. Abd Al-Khodor, T.M. Albayati / Process Safety and Environmental Protection 136 (2020) 334–342 341 Fig. 9. Variation of total sulfur content of caustic process a) time b) temperature c) weight d) speed. Fig. 10. Optimization condition for the desulfurization by sodium hydroxide. compared with other studies. The total sulfur content before and after treatment is shown in Fig. 11. weak acid while it is known that the reaction of a weak acid with a strong base is rapid, so the NaOH is a strong base and is a good solvent to remove sulfur dioxide. 3.7. Comparative study This study sought to evaluate and verify a process developed to decrease the sulfur content in the actual heavy crude oil from an initial value of 5.8 wt. % to less than 2.6 wt. %. This implies substantial reduction in the sulfur content, of about 58,202–26,083 ppm. In addition, the process has the advantages of simplicity, ease of operation and low cost effectiveness. This provides clear benefits to the process considered in this study compared to those of other ones (as shown in Table 8), which use expensive chemicals and equipment, difficult, and time-consuming processes. Sulfur dioxide is a 4. Conclusion The efficiency of desulfurization increases with increasing reaction time and mixing speed, and with the application of moderate conditions for reaction temperature and concentration of alkaline compounds. In addition, it was found that the sulfur content resulting from the process was dependent on time, temperature, weight of NaOH and mixing speed in the following sequence: mixing speed > weight of NaOH > time > temperature. Optimum conditions of the 342 Y.A. Abd Al-Khodor, T.M. Albayati / Process Safety and Environmental Protection 136 (2020) 334–342 Fig. 11. Effect NaOH on the final sulfur content for actual heavy crude oil. studied variables were obtained as follows: time = 60 min., temperature = 40 ◦ C, NaOH solution = 18 gm and mixing speed = 500 rpm while ones of the sulfur content were applied experimentally and theoretically, where sulfur content was equal to 2.5 and 2.3 wt. %, respectively. The efficiency of the sulfur removal content for actual heavy crude oil by this process was 56.89 %. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements The authors wish to thank the Department of Chemical Engineering, University of Technology, Baghdad, Iraq and Al-Halfaya Oil Field in southern Iraq for the partial support of this study. References Alaya, M.N., 2012. Kinetics of Chemical Reactions. Faculty of Science, University of Aleppo, Aleppo, Syria. Albayati, T.M., Kalash, K.R., 2020. Polycyclic aromatic hydrocarbons adsorption from wastewater using different types of prepared mesoporous materials MCM-41in batch and fixed bed column. Process. Saf. Environ. Prot. 133, 124–136. Alkafajy, A.M., Albayati, T.M., 2020. High performance of magnetic mesoporous modification for loading and release of meloxicam in drug delivery implementation. Mater. Today Commun. 23, 100890. Anupam, K., Dutta, S., Bhattacharjee, C., Datta, S., 2011. Adsorptive removal of chromium (VI) from aqueous solution over powdered activated carbon: optimisation through response surface methodology. Chem. Eng. J. 173 (1), 135–143. Babich, I.V., Moulijn, J.A., 2003. Science and technology of novel processes for deep desulfurization of oil refinery streams: a review⋆. Fuel 82 (6), 607–631. Breysse, M., Djega-Mariadassou, G., Pessayre, S., Geantet, C., Vrinat, M., Pérot, G., Lemaire, M., 2003. Deep desulfurization: reactions, catalysts and technological challenges. Catal. Today 84 (3–4), 129–138. Bridge, G., 2008. Global production networks and the extractive sector: governing resource-based development. J. Econ. Geogr. 8 (3), 389–419. Chong, C., Ni, W., Ma, L., Liu, P., Li, Z., 2015. The use of energy in Malaysia: tracing energy flows from primary source to end use. Energies 8 (4), 2828–2866. Eßer, J., Wasserscheid, P., Jess, A., 2004. Deep desulfurization of oil refinery streams by extraction with ionic liquids. Green Chem. 6 (7), 316–322. Jabbar, S.M., 2015. Desulfurization of AL-Ahdab crude oil using oxidative processes. J. Eng. 21 (7), 102–112. Javadli, R., De Klerk, A., 2012. Desulfurization of heavy oil. Appl. Petrochem. Res. 1 (1–4), 3–19. Jeyajothi, K., 2015. Study on Effect of Desulfurization of Sour Crude Oil in Petro Chemical Industry. Department of Chemical Engineering, Periyar Maniammai University, Thanjavur, Tamil Nadu, India. Khataee, A.R., Fathinia, M., Aber, S., Zarei, M., 2010. Optimization of photocatalytic treatment of dye solution on supported TiO2 nanoparticles by central composite design: intermediates identification. J. Hazard. Mater. 181 (1–3), 886–897. Kuehl, R.O., Kuehl, R.O., 2000. Design of Experiments: Statistical Principles of Research Design and Analysis. Lee, K.X., Valla, J.A., 2017. Investigation of metal-exchanged mesoporous Y zeolites for the adsorptive desulfurization of liquid fuels. Appl. Catal. B 201, 359–369. Li, Y., Zhang, Y., Wu, P., Feng, C., Xue, G., 2018. Catalytic Oxidative/Extractive desulfurization of model oil using transition metal substituted phosphomolybdatesbased ionic liquids. Catalysts 8 (12), 639. Lin, L., Hong, L., Jianhua, Q., Jinjuan, X., 2010. Progress in the technology for desulfurization of crude oil. China Petrol. Process. Petrochem. Technol 12, 1–6. Manihe, A.A., Ghorayeb, N., 1981. September. How to Design a Caustic Wash. Hyrocarbon Process., pp. 143–144. Mason, P.E., Uhlig, F., Vaněk, V., Buttersack, T., Bauerecker, S., Jungwirth, P., 2015. Coulomb explosion during the early stages of the reaction of alkali metals with water. Nat. Chem. 7 (3), 250. Moaseri, E., Mostaghisi, O., Shahsavand, A., Bazubandi, B., Karimi, M., Ahmadi, J., 2013. Experimental study and techno-economical evaluation of Khangiran sour natural gas condensate desulfurization process. J. Nat. Gas Sci. Eng. 12, 34–42. Montgomery, D.C., 2005. Design and Analysis of Experiments, 6th ed. John Wiley and Sons, Inc, New York. Mook, W.T., Aroua, M.K., Chakrabarti, M.H., Low, C.T.J., Aravind, P.V., Brandon, N.P., 2016. The application of nano-crystalline PbO2 as an anode for the simultaneous bio-electrochemical denitrification and organic matter removal in an up-flow undivided reactor. Electrochim. Acta 94, 327–335. Noorimotlagh, Z., Soltani, R.D.C., Khataee, A.R., Shahriyar, S., Nourmoradi, H., 2014. Adsorption of a textile dye in aqueous phase using mesoporous activated carbon prepared from Iranian milk vetch. J. Taiwan Inst. Chem. Eng. 45 (4), 1783–1791. Ölmez, T., 2009. The optimization of Cr (VI) reduction and removal by electrocoagulation using response surface methodology. J. Hazard. Mater. 162 (2–3), 1371–1378. Shakirullah, M., Ahmad, I., Ahmad, W., Ishaq, M., 2010. Desulphurization study of petroleum products through extraction with aqueous ionic liquids. J. Chil. Chem. Soc. 55 (2), 179–183. Sharma, A.K., Prasad, D.S.N., Acharya, S., Sharma, R., 2012. Utility and application of FGD system (Flue gas desulphurization) in chemical and environmental engineering. Int. J. Chem. Eng. Appl. 3 (2), 129. Song, C., 2003. An overview of new approaches to deep desulfurization for ultraclean gasoline, diesel fuel and jet fuel. Catal. Today 86 (1–4), 211–263. Speight, J.G., 2015. Handbook of Petroleum Product Analysis, Vol. 182. John Wiley & Sons. Srikanth, S., Kumar, M., Puri, S.K., 2018. Bio-electrochemical system (BES) as an innovative approach for sustainable waste management in petroleum industry. Bioresour. Technol. 265, 506–518. Stanislaus, A., Marafi, A., Rana, M.S., 2010. Recent advances in the science and technology of ultralow sulfur diesel (ULSD) production. Catal. Today 153 (1–2), 1–68. Ugal, J.R., Jima’a, R.B., Al-Jubori, W.M.K., Abbas, B.F., Al-Jubori, N.M., 2018. Oxidative desulfurization of hydrotreated gas oil using Fe2O3 and Pd loaded over activated carbon as catalysts. Orient. J. Chem. 34 (2), 1091–1097. Zarei, M., Niaei, A., Salari, D., Khataee, A., 2010. Application of response surface methodology for optimization of peroxi-coagulation of textile dye solution using carbon nanotube–PTFE cathode. J. Hazard. Mater. 173 (1–3), 544–551.