Microsoft Word - 164.docx CHEMICAL ENGINEERING TRANSACTIONS VOL. 56, 2017 A publication of The Italian Association of Chemical Engineering Online at www.aidic.it/cet Guest Editors: Jiří Jaromír Klemeš, Peng Yen Liew, Wai Shin Ho, Jeng Shiun Lim Copyright © 2017, AIDIC Servizi S.r.l., ISBN978-88-95608-47-1; ISSN 2283-9216 Optimisation of Lipid Extraction from Primary Sludge by Soxhlet Extraction Nor Aiman Hafizi Mustaphaa, Wang Sing Hiia, Roshanida Abdul Rahman*,a, Norzita Ngadib, Ismail Mahmoodb aDepartment of Bioprocess and Polymer Engineering, Faculty of Chemical & Energy Engineering, Universiti Teknologi Malaysia, 81310, Johor Bahru, Johor, Malaysia. bDepartment of Chemical Engineering, Faculty of Chemical & Energy Engineering, Universiti Teknologi Malaysia, 81310, Johor Bahru, Johor, Malaysia. roshanida@cheme.utm.my Domestic sewage sludge is a potential raw material to produce energy in the form of renewable biofuels (biodiesel) due to the high amount of lipid content in the sludge. Soxhlet extraction is commonly used in extracting lipid from primary sewage sludge. Different solvents and extraction parameters in Soxhlet extraction may affect the lipid yield. Hexane, methanol and toluene were evaluated in extracting lipids from primary sewage sludge. Methanol was identified as the most suitable solvent to be used in lipid extraction due to its high polarity. Extraction parameters such as temperature, extraction time, and sludge-to-solvent ratio were investigated in order to optimise the lipid extraction process from primary sewage sludge. The extracted lipid yield was observed to be optimum at 40.21 wt% when the temperature and extraction time were at 86 °C and 7.45 h. A sludge-to-solvent ratio of 2:1 was determined to be the optimum value in extracting lipid yield. The optimisation process was conducted using response surface methodology (RSM) and was analysed using ANOVA with R2 of 0.8146 for lipid concentration. The analysis of free fatty acid content in the lipid extracted from primary sewage sludge revealed that the composition of free fatty acid was dominated by monounsaturated oleic acid. 1. Introduction Sewage sludge is a solid, semisolid, or liquid muddy looking residue that results after plain sewage is treated at a wastewater treatment plant. It is mainly produced from domestic and industrial activities. Malaysia has found to produce approximately 3 × 106 m³/y of sewage sludge in wastewater treatment plants in these recent years and the amount is keep increasing (Abu Hasan et al., 2012). As a result, many new sludge treatment and disposal facilities are needed to manage the large volume. Abu Hasan et al. (2012) estimated that the sewage sludge produced in Malaysia will increase to 7 Mt in year 2020. The estimation for the total cost in managing sewage sludge is about 1 B MYR (Mohd Safuan et al., 2014). Sewage sludge is a good source of micro/macronutrients for plants besides its richness in organic matter and beneficial to the soil, crops, and livestock productivity. However, the use of sewage as a fertiliser is restricted in many countries due to bad odour and the presence of heavy metals and toxic substances. Recently, domestic sewage sludge has been used and proven to be a good raw material to produce energy in the form of renewable biofuels such as biodiesel and is considered to be economically and eco-friendly process; although the process itself is quite complicated (Girisha et al., 2014). Sewage sludge contains significant concentrations of lipids derived from the direct adsorption of lipids onto the sludge (Kargbo, 2010). The amount of lipid-rich wastewater increases every year due to urbanisation (Chipasa and Mędrzycka, 2006). Lipids are one of the most important components of natural foods that constitute one of the major types of organic matter found in municipal wastewater which may find their way into surface waters. Therefore, an alternative to sewage management and disposal challenge is to utilise it as a source of lipid feedstock for biodiesel production (Jardé et al., 2005). However, the compositions of lipids in sewage sludge are different according to the stage of wastewater treatment that been passed through. The presence of microorganisms DOI: 10.3303/CET1756221 Please cite this article as: Mustapha N.A.H., Hii W.S., Rahman R.A., Ngadi N., Mahmood I., 2017, Optimization of lipid extraction from primary sludge by soxhlet extraction, Chemical Engineering Transactions, 56, 1321-1326 DOI:10.3303/CET1756221 1321 during wastewater treatment will also affect the composition of lipids due to the phospholipids component of the cell membrane of the microorganisms. As a raw material for lipid extraction, sludge can be collected at primary, secondary, or tertiary treatment system. Most of the studies done by other researchers have extracted lipid from sewage sludge by using primary and secondary sludge (Pastore et al., 2013). Due to the great chemical complexity and diversity of the sludge produced, the lipid extraction becomes difficult (Bharathiraja et al., 2014). Many investigations have been done by other researchers to solve this problem using different methods to extract the lipid from sewage sludge. Olkiewicza et al. (2014) studied that the solvent extraction using methanol can produce 14.16 % of lipid from primary sewage sludge. However, study on the optimisation of operating parameters for lipid extraction from primary sludge by Soxhlet extraction has not been done yet. Therefore this study was performed to determine the optimum parameters of temperature, time, and sludge-to-solvent ratio in Soxhlet extraction using response surface methodology (RSM). 2. Materials and Method 2.1 Wastewater sample and samples drying Sludge samples were collected from one of the primary waste water treatment plants in Johor, Malaysia. Sludge samples (15 L) were taken manually and kept in a closed sealed plastic container. The moisture content of the sludge sample was 96 %. The samples were immediately transported back to the laboratory for immediate use. Sewage sludge was dried first before being used for the extraction process. Twenty grams of wet sample was mixed with 0.3 mL fuming hydrochloric acid. After the acidification process, 25 g of magnesium sulphate monohydrated was added to the mixture. The mixture was stirred in order to obtain a smooth paste and it was left until solidified (15 - 30 min). Then, the dried sample was ground using a pestle and mortar until it was homogenous before filled into cellulose thimble for extraction process. 2.2 Extraction of lipids An empty thimble and an empty round bottom flask were weighed. After adding up the dried sludge (20 g) into the thimble, it was weighed again and the reading was recorded. Then, 200 mL of hexane was added into 250 mL round bottom flask. The heating process was monitored to allow 80 cycles of the extraction in approximately 6 h. The hexane was removed from the flask by rotary evaporator after the lipid had been extracted. Then, the lipids were stored in a desiccator overnight and weighed the next day. The yield of extracted material was determined gravimetrically and expressed as g of extractable lipids per g of dry sludge. The step was repeated with different solvents, which were methanol and toluene. 2.3 Optimisation of lipid extraction process by response surface methodology The optimisation experiments were conducted for three factors with two numerical factors and one categorical factor with the levels of each factor are shown in Table 1. The factors were temperature, time, and sludge-to- solvent ratio. The response was extracted lipid yield with 22 set of experiments as shown in Table 2. Analysis of variance (ANOVA) was used for the analysis of the experimental data to determine the interactions between the process variables and response obtained as well as model validation. The quality of fitness in the polynomial model was expressed by the correlation coefficient (R2) and its statistical significance was verified by an F-test. The model terms in this study were evaluated based on the P-value (probability) corresponding to a 95 % confidence level. The optimisation process was done with the aid of Design Expert Software for mathematical model and statistical analysis (Design Expert, 2005). Table 1: Level of factors for the optimisation process Factor Level of value (Numerical Factor) -1 0 +1 Temperature 70 80 90 Time (h) 4 6 8 Level of value (Categorical factor) Level Level 1 Level 2 Sludge : solvent ratio 2 1:2 2:1 1322 Table 2: Experimental data for the optimisation process Run Temperature (°C) Time (h) Solvent : solvent ratio Run Temperature (°C) Time (h) Solvent : solvent ratio 1 80 6 1:2 12 70 4 2:1 2 90 4 2:1 13 90 8 2:1 3 70 8 1:2 14 80 6 1:2 4 80 4 1:2 15 80 4 2:1 5 80 6 2:1 16 70 4 1:2 6 70 8 2:1 17 80 6 1:2 7 80 8 1:2 18 70 6 2:1 8 80 8 2:1 19 70 6 1:2 9 90 6 2:1 20 90 4 1:2 10 90 6 1:2 21 80 6 2:1 11 90 8 1:2 22 80 6 2:1 2.4 Total primary sludge lipid content The lipid content was calculated as the total crude lipid yield for each experimental set. The extracted samples was calculated as percentage based on Eq(1). Extraction (%) = (w2 - w1) / (m2 - m1) × 100 (1) Where m1 represents the mass (g) of thimble, m2 is the total mass (g) of thimble and sample (g), w1 is the mass (g) of round bottom flask, and w2 is the total mass of round bottom and lipid after extraction (g) 2.5 Free fatty acid analysis Gas chromatography with flame ionisation detector (GC-FID) was used to determine the free fatty acid contents in primary sewage sludge oil. Primary sewage sludge oil (5 mL) was allowed to dry and reconstituted in 200 µL of ethanol. In this chromatographic separation, helium gas was used as the carrier with 0.9 mL/min flow rate. The sample volume used for injection was 1 µL. 3. Results and discussion 3.1 Lipid yield of different solvent extraction Sewage sludge contains various solute molecules with more than one functional group so it was difficult to predict the solubility of the solutes in particular solvents. The polarity concept of the solvent was used to solve the problem of solubility. In this study, the lipid extracted from primary sewage sludge was successfully performed only using methanol. Both n-hexane and toluene were not able to extract the lipids from the primary sewage sludge at described conditions. Meanwhile, 10.5 % of lipid was successfully extracted by using methanol. The non-polar characteristic of hexane may have come out this observation. However, high polarity of methanol breaks the weak phospholipid bond of the sludge sample by the reaction of methanol on the cell membrane of the microorganisms in the sewage sludge. It results in the separation of lipid from the phospholipid bond. Normally, high polarity solvent that has high value of polarity index is able to extract a high amount of lipid oil from primary sewage sludge. Therefore, methanol was able to extract a higher fraction of the more polar free fatty acids, whereas n-hexane appeared to be a better extraction agent for the non-polar glycerides (Melero et al., 2015). 3.2 Optimisation of lipid extraction using RSM The optimisation process was done using response surface methodology (RSM). The results for 22 runs are shown in Table 3. Based on the sequential model sum of square, the model for lipid yield was selected based on the highest order linear equation. The models were coded X1 for lipid yield percentage and the independent variables in the models were temperatures (A), reaction time (B), and sludge-to-solvent ratio (C). The final empirical models used to generate coded factors for each variable is shown in Eq(2). In addition to the ANOVA statistical analysis, the quality of the model was also evaluated based on the coefficient of determination. The ANOVA results for the quadratic model for lipid yield percentage are shown in Table 5. 1323 Table 3: Response value for different experimental conditions Run A = Temperature (°C) B = Time (h) C = Solvent : solvent ratio Lipid yield (%) 1 80 6 1:2 33.50 2 90 4 2:1 31.00 3 70 8 1:2 32.01 4 80 4 1:2 26.90 5 80 6 2:1 37.80 6 70 8 2:1 34.50 7 80 8 1:2 36.50 8 80 8 2:1 38.80 9 90 6 2:1 37.60 10 90 6 1:2 34.60 11 90 8 1:2 38.00 12 70 4 2:1 27.30 13 90 8 2:1 40.00 14 80 6 1:2 34.90 15 80 4 2:1 28.50 16 70 4 1:2 27.90 17 80 6 1:2 31.90 18 70 6 2:1 33.50 19 70 6 1:2 30.10 20 90 4 1:2 28.10 21 80 6 2:1 37.80 22 80 6 2:1 38.10 Lipid Yield (%) R1 = 33.57 + 2.0A + 4.18B + 1.39C + 0.95AB + 0.22AC + 0.24BC - 0.98A2 – 2.26B2 (2) In this study the value of the determinations coefficients (R2= 0.9428) indicates that only 5.72 % of the total variations are not explained by the model. The models obtained from the design fitted well with the experimental data. The value of the adjusted determinations coefficients (Adj R2 of 0.9076) was also high, which indicates a high significance of the model. Based on the ANOVA result in Table 4, the parameter gives a significant effect on the lipid yield percentage. The model F-value of 26.78 implies the model is significant due to Prob > F, which is less than 0.01 % chance of the “Model F-value” to occur due to noise. Table 4: Analysis of variance for lipid yield percentage Source Sum of square Mean square F value P-value Prob>F Model 347.49 43.44 26.78 < 0.0001 Significant A – Temperature 47.96 47.96 29.57 0.0001 B – Time 209.25 209.25 129.03 < 0.0001 C – Ratio 42.26 42.26 26.06 0.0002 AB 7.20 7.20 4.44 0.0551 AC 0.57 0.57 0.35 0.5642 BC 0.70 0.70 0.43 0.5238 A2 4.90 4.90 3.02 0.1057 B2 25.85 25.85 15.94 0.0015 Residual 21.08 1.62 Lack of fit 16.52 1.84 1.61 0.3417 Not significant Pure error 4.57 1.14 Cor Total 368.58 Std Dev 1.27 R2 0.9428 Adj R2 0.9076 1324 3.3 Extraction efficiencies To assess the interactive relationships between independent variables and the responses in the model, a 3D surface response and contour plots were utilised using Design Expert software. As shown in Figure 1, the optimum value observed for lipid yield is 40.21 % using a process condition of sludge-to-solvent ratio of 2:1 at 86 °C for 7.45 h. Figure 1: Response surface for lipid yield percentage. Effect of reaction time and temperature at fixed ratio (2:1) 3.4 Effects of temperature, extraction time and ratio sludge to solvent on extracted lipid yield By referring to the data show in Table 5, the experimental data was in the range with previous research. Thus primary sewage sludge was proved as suitable raw material in lipid extraction. It is expected that a higher lipid yield could be obtained due to the significant effect of temperature and reaction time that had been determined in the optimisation process. Besides, the longer the extraction period, the higher degree of adsorption of lipid from the sewage sludge could be because highly non-polar lipid molecule in the sewage sludge need a longer time to be separated (Olkiewicza et al., 2015). However, the disadvantage of longer extraction period was the contamination with non-saponifiable matter which was not convertible into biodiesel (Olkiewicza et al., 2015). Table 5: Lipid yield percentage by other researchers Type of sludge Method Lipid yield (wt %) References Dried primary sludge Soxhlet extraction 37.30 Bozhagian (2014) Dried primary sludge Soxhlet extraction 24.80 Pokoo-Aikins et al. (2010) Dried primary sludge Liquid-liquid extraction 26.60 Olkiewicza et al. (2014) Dried primary sludge Soxhlet extraction 27.20 Olkiewicza et al. (2015) Raw primary sludge Ionic liquid extraction 26.90 Olkiewicza et al. (2015) Primary sludge Solvent extraction 13.60 Melero et al. (2015) Primary sludge Soxhlet extraction 40.21 This study 3.5 Analysis of the free fatty acid content Siddiquee and Rohani (2011) stated that the most dominant free fatty acid contents in the lipid oil from primary sewage sludge was palmitic acid (C16:0), stearic acid (C18:0) and oleic acid (C18:1). In this study, the dominant fatty acid was oleic acid with the highest concentration obtained at 77.1 ppm. The concentration of palmitic acid was too small which was not significant to be identified. The monounsaturated oleic acid gives strong oxidation stability of biodiesel. The saturated or monounsaturated fatty acids, such as palmitic acid, stearic acid, and oleic acid, promote oxidation stability. The polyunsaturated fatty acids, such as linoleic acid and linolenic acid, tend to give a poor oxidation stability. Most of the biodiesel from the polyunsaturated fatty acid are difficult to achieve the minimum limit of six hours for oxidation stability (Olkiewicza et al., 2012). However, the melting points of biodiesel product depend on the chain length and the degree of unsaturation, 1325 therefore the polyunsaturated fatty acids will have a favourable cold flow properties (Melero et al., 2015,). Most of the researchers used oleic acid as a raw material to produce biodiesel (Kusmiyati et al., 2010). According to Tran et al. (2013), oleic acid is known to be a main component of biodiesel, accounting for more than 50 % of the weight of total fatty acids. The nature of monounsaturated oleic acid gives an advantage in the oxidation stability and cold flow properties (Patel and Narkhede, 2012). 4. Conclusion Methanol was preferable as solvent in extracting lipid from primary sewage sludge. The optimisation process for lipid extraction process using methanol was affected by temperature, extraction time, and sludge-to-solvent ratio. The extraction process was successfully performed and yielded an increased amount of lipid up to 40 % from 10.5 % before the optimisation process. The analysis of free fatty acid content in the lipid extracted from primary sewage sludge revealed that the composition of free fatty acid was dominated by monounsaturated oleic acid. Reference Abu Hasan H., Sheikh Abdullah S.R., Tan Kofli, N., Kamarudin S.K., 2012, Isotherm equilibria of Mn2+ biosorption in drinking water treatment by locally isolated Bacillus species and sewage activated sludge, Journal of Environmental Management 111, 34-43. Bharathiraja B., Chakravarthy M., Ranjith Kumar R., Yuvaraj D., Jayamuthunagai J., Praveen Kumar R., Palani S., 2014, Biodiesel production using chemical and biological methods – A review of process, catalyst, acyl acceptor, source and process variables, Renewable and Sustainable Energy Reviews 38, 368-382. Bozaghian M., 2014, Characterization and synthesis of biodiesel from sludge available in the Umeå region, Degree Project, Faculty of Science and Technology, Department of Chemistry, Umeå University. Chipasa K.B., Mędrzycka K., 2006, Behavior of lipids in biological wastewater treatment processes, Journal of Industrial Microbiology and Biotechnology 33 (8), 635-645. Design Expert V6.0, 2015, Stat-Ease Inc., Minneapolis, United States. Girisha S.T., Ravikumar K., Mrunalini Girish V., 2014, Lipid extraction for biodiesel production from municipal sewage water sludge, European Journal of Experimental Biology 4 (1), 242-249. Jardé E., Mansuy L., Faure P., 2005, Organic markers in the lipidic fraction of sewage sludges, Water Research 39 (7), 1215-1232. Kargbo D.M., 2010, Biodiesel production from municipal sewage sludges, Energy Fuels 24, 2791-2794. Kusmiyati K., Sugiharto A., 2010, Production of biodiesel from oleic acid and methanol by reactive distillation, Bulletin of Chemical Reaction Engineering & Catalysis 5 (1), 1-6. Melero J.A., Sánchez-Vázquez R., Vasiliadou I.A., Castillejo F.M., Bautista L.F., Iglesias J., Morales G., Molina R., 2015, Municipal sewage sludge to biodiesel by simultaneous extraction and conversion of lipids, Energy Conversion and Management 103, 111-118. Mohd Safuan Z., Hassan S., Faizairi M., 2014, Thermal drying of Malaysian sewage sludge, Journal of Advanced Research in Fluid Mechanics and Thermal Sciences 3 (1), 1-5. Olkiewicza M., Fortuny A., Stüber F., Fabregat A., Font J., Bengoa C., 2012, Evaluation of different sludges from WWTP as a potential source for biodiesel production, Procedia Engineering 42, 634-643. Olkiewicza M., Caporgno M.P., Fortuny A., Stüber F., Fabregat A., Font J., Bengoa C., 2014, Direct liquid- liquid extraction of lipid from mucnicipal sewage sludge for biodiesel production, Fuel Processing Technology 128, 331-338. Olkiewicza M., Plechkova N.V., Fabregat A., Stüber F., Fortuny A., Font J., Bengoa C., 2015, Efficient extraction of lipids from primary sewage sludge using ionic liquids for biodiesel production, Separation and Purification Technology 153, 118-125. Pastore C., Lopez A., Lotito V., Mascolo G., 2013, Biodiesel from dewatered wastewater sludge: A two-step process for a more advantageous production, Chemospere 92 (6), 667-673. Patel A., Narkhede N., 2012, 12-Tungstophosphoric acid anchored to zeolite Hβ: Synthesis, characterization, and biodiesel production by esterification of oleic acid with methanol, Energy Fuels 26 (9), 6025-6032. Pokoo-Aikins G., Heath A., Mentzer R.A., Mannan M.S., Rogers W.J., El-Halwagi M.M., 2010, A multi-criteria approach to screening alternatives for converting sewage sludge to biodiesel, Journal of Loss Prevention in the Process Industries 23 (3), 412-420. Siddiquee M.N., Rohani S., 2011, Lipid extraction and biodiesel production form municipal sewage sludges: A review, Renewable and Sustainable Energy Reviews 15, 1067-1072. Tran H.L., Ryu Y.J., Seong D.H., Lim S.M., Lee C.G., An effective acid catalyst for biodiesel production from impure raw feedstocks, Biotechnology and Bioprocesses Engineering 18 (2), 242-247. 1326