Optimization of Biodiesel Production from Spent Cooking Oil by Fungal Lipase Using Response Surface Methodology

This study evaluated the potential of optimizing Spent Cooking Oil (SCO) transesterification for biodiesel production using Response Surface Methodology (RSM). Potential biodiesel yield from transesterification was optimized using a three-level four factor Response Surface Central Composite Design (RSCCD) with methanol oil ratio (1:1 to 3:1), temperature range 35-45 °C, agitation speed range 100-200 rpm and enzyme load 10-20%. Biodiesel properties including fatty acid methyl ester (FAME), Flash Point (FP), Pour Point (PP) and kinematic viscosity were compared with American (ASTM D6751) and European Union (EN 14214) standards. Biodiesel production was optimum at 3:1 methanol to oil ratio, temperature of 35 °C, agitation speed of 150 rpm and 20% enzyme load. 9-octadecanoic acid-hydroxyl methyl ester (33.83%) was the prominent 2 3 FAME produced, while the viscosity (6 mm /s), density (893 kg/m ), FP (260 °C) and PP (0.5 °C) all met both American and European standards. This study showed that RSM is a viable methodology which could be used for optimization of biodiesel production from biological sources.


Introduction
corn among others (Dermibas, 2003;Akoh, 2007). Oils from various Vast depletion of non-renewable energy feed stock are used for the enzymatic production sources has led to the search for alternative of biodiesel, with vegetable oil currently being energy sources (Owusu and Asumadu-Sarkodie, used worldwide as a sustainable commercial 2016). Price hike in petroleum-based products as feedstock (Dermibas, 2006). The use of well as greenhouse gas emissions contribute vegetable oil and other edible oils sourced from significantly the need to search for alternative feedstock such as sunflower, soyabeans, and renewable energy sources (Abbasi et al., rapeseed, in biodiesel 2011;Ribeiro et al., 2011).
production is of great concern to food security Biodiesel is a non-toxic, non-sulphur due to competition. Since the prices of edible containing renewable energy fuel consisting of vegetable oils, e.g. soybean oil, are higher than long chain fatty acid derived from vegetable oils that of diesel fuel, waste vegetable like palm, or animal fats (Vincente et al., 2007). Enzymatic sunflower, and algal oils (Encinar, 1999; Hossain conversion of oils to biodiesel by lipases as 2009b) and non-edible crude vegetable oils have biocatalysts is receiving much interest in been intensively investigated as potential low biodiesel production due to its high efficiency priced biodiesel sources. Biodiesel made from and production of a highly purified product this feedstock could be more economical than the biodiesel produced from refined vegetable oil. The waste oil product can be of advantage as they have a higher proportion of saturated fatty acids (Hossain, 2009a). Waste cooking oil is a good feedstock to produce biodiesel for waste management and recycling process (Sarin, 2007). Moreover, waste cooking oil can be used as an alternative source of fuel because it is cheap and readily available. Response surface methodology (RSM), on the other hand, is a statistical tool which is used by scientists to optimize processes; example includes fermentation processes (Zhang et al., 2000). The RSM is equipped with statistical tools to determine the significance of a factor over a response. The evaluation of factors using the RSM uses experimental design in order to distribute the selected variables within the boundaries of the design.
Hence, the aim of this study was to optimize fatty acid methyl esters (FAME) production from spent cooking oils by lipase of Aspergillus niger using Response Surface Methodology.

Materials and Methods Materials
activity was confirmed according to Akpan (2004). SDA medium was modified with bromocresol green (0.1%) and Tween 80 (1%). Strain F7-02 was inoculated and medium incubated for 72h at 30°C. Lipase production was confirmed by colour change around the colonies.

Lipase production by Solid State Fermentation
For lipase production, Aspergillus niger F7-02 was grown via solid-state fermentation on a modified compounded medium described by Osho et al. (2014). The compounded medium substrates included rice bran, palm kernel cake waste, groundnut cake waste, and starch flour in the ratio 5:5:3:1 (% w/w), and moistened with 55 % water. Inoculated medium was incubated at 30 °C for 72 h. Moldy medium was dissolved in 50 mM sodium phosphate buffer pH 8 (1:10 w/v) and the mixture incubated at 4°C for 3 h with intermittent shaking. The filtered extract served as the crude enzyme source.

Assay of Lipase Activity
Lipase activity was determined according to combined methods of Praphan and Kirk (2001) and Janaina et al. (2006). Olive oil substrate emulsion was prepared by mixing 25 mL of olive oil with 7 % arabic gum solution (75 mL) in a conical flask and incubate at 37°C for 15 minutes Spent cooking oil (SCO) was collected in a water bath (Nickel Electro Ltd, England). locally from homes and restaurants. Raw Reaction mixture was made up of 50 ml olive oil materials for the composition of culture media emulsion and 10 ml crude enzyme incubated at such as rice bran, Palm Kernel Cake (PKC), 50°C for 30 minutes with intermittent shaking in Groundnut Cake (GNC), starch and Olive oil were water bath. At 5-minute intervals, 5 mL of purchased locally in Abeokuta, Ogun State, and reaction mixture was removed and mixed with 5 sterilized accordingly prior to use. All chemicals mL ethanol (95%) and thymolphtalein indicator including methanol, ethanol, Tween 80, (2-3 drops) in a conical flask to stop the reaction. bromcresol green, lactophenol blue, sodium The released fatty acid was titrated with sodium diodoethyl sulphate, chloroform, gum arabic, hydroxide (0.05 N) in a burette until a light blue Thymolphtalein, Bovine serum Albumin (BSA), color appears. sodium hydroxide, sodium dihydrogen The quantity of fatty acid liberated is phosphate, monosodium hydrogen phosphate, equivalent to the volume of NaOH used and it Sodium potassium tartarate, Copper sulphate was calculated using equation (1), where N is the pentahydrate, and Potassium iodide were of normality of NaOH used. analytical grade.

Microorganism
Aspergillus niger F7-02, is a lipase-producing One unit (U) of lipase activity is defined as the fungus used for transesterification from a amount of enzyme that releases from the previous study (Adio et al., 2015). It was stored emulsion substrate 1 µmole of fatty acid per ml at 4°C and sub-cultured bimonthly on a SDA.
per minute under specific assay condition. Prior to its use in this study, retention of lipolytic Enzymatic production of biodiesel by lipase of Statistical analysis of the biodiesel yield from Aspergillus niger lipase treated SCO Biodiesel production was carried out as Responses obtained from three-factorial described by Taufiq-Yap et al. (2011). SCO (30 g experimental runs were analyzed by ANOVA for w/v) was heated to 40°C in 250 ml Erlenmeyer response surface linear model and the effect of flasks, crude lipase enzyme (10 %v/w of oil) was factors considered singly and in combination was added and mixture was incubated in a shaker determined. The prediction model was also used incubator for 24 hours at 40°C and 150 rpm.
to predict possible optimum yield. Thereafter, methanol was added to the mixture and reaction proceeded for a further 24h.
Characterization of biodiesel Reaction products were separated into fractions Properties of biodiesel produced (FAME) by sedimentation and biodiesel was separated was analyzed via Gas Chromatography (Agilent from glycerol and other impurities in separating Technologies 7890A  determined, and all properties compared with methanol-to-SCO molar ratio (1-3, with the SCO both American Standard (ASTM 6751-3) and fixed at 1) and enzyme load (10-20%) and European Union Standard (EN 14214) in agitation (100-200 rpm). Transesterification of properties and quality of biodiesel. Biodiesel SCO proceeded using the three-level-four-factor properties were analyzed in triplicate.

Response Surface Central Composite Design (RSCCD) of Design-Expert software version
Results and Discussion 7b1.1. Thirty (30) experimental runs were The optimization of biodiesel production generated using the combination of factors at from spent cooking oil using a three-level four different levels (low, mid and high; -1, 0, +1) and factor Response Surface Central Composite the randomized experiments were carried out Design (RSCCD) showed that maximum simultaneously as previously described.
biodiesel yield of 97% was obtainable under Tra n s e s t e r i f i c a t i o n p r o d u c t f o r e a c h three experimental runs. Experimental runs experimental run was allowed to settle for 24h in which produced highest biodiesel yield had a separating funnel, and the fatty acid methyl factors temperature (35°C) and molar ratio (3:1) ester (FAME) called the biodiesel, was separated at similar levels, while agitation and enzyme load from the glycerol layer. Biodiesel yield (% wt.), were different. Similarly, minimum biodiesel yield which is relative to the amount of SCO was of 92% was obtained at transesterification calculated according to the method of Fan et al. experiment which proceeded at temperature (2011) as described in equation 2: 45°C and of 1:1 molar ratio respectively (Table  1). The 3D Response surface plot of interactions at enzyme load and temperature of 35°C, while the different levels of factors is described in minimum yield of 93.20% was predicted to occur Figure 1. A curved response surface was at molar ratio (1:1) and enzyme load 10%. obtained, with highest response at the upper left Optimization parameters are in agreement with hand corner of the plot, corresponding to 35 °C the report of Taufiq-Yap et al. (2011) who temperature and 3:1 molar ratio. Furthermore, reported that stoichiometric ratio for effect of temperature and enzyme load on SCO transesterification requires 3 moles of methanol biodiesel yield was described in response surface and 1 mole of oil to yield 3 moles of biodiesel and plots (Figure 2). It was observed that a mole of glycerol, which indicates that excess transesterification proceeding at 20% enzyme methanol, is required to drive the reaction load and temperature of 35°C were conditions towards the product, with enzyme load also a required for optimum biodiesel yields of 96.34 % vital factor. Similar observation was reported by from SCO. The effects of molar ratio and enzyme Nadir et al., (2009), and was attributed to the load on biodiesel yield from SCO are described in fact that excess enzyme can make oils viscous, Figure 3. Figure 4 shows a cube-plot which causing problem of mixing and demanding described the prediction model for SCO biodiesel higher power consumption for adequate stirring optimization. Maximum biodiesel yield of (Kim et al., 2004;Xie and Li, 2006). 97.42% was predicted at molar ratio (3:1), 20%  (Table 3) showed that oleic acid is the pour point, and cloud point of crude SCO and the major fatty acid in the SCO, with 51.87% produced biodiesel are described in Table 2. The composition. Lipase transesterification of SCO viscosity of SCO biodiesel met the ASTM however converted it to more Fatty acid methyl 2 standard of 6 mm /s maximum, while its density esters (FAME) of 9-octadecanoic acid-hydroxyl 3 of 893 kg/m at 15°C also met the EN standard of methyl ester-the compound desired in 3 900 kg/m maximum. The flash point and cloud biodiesels-with composition of about 33% point were also within the limits of both displayed between retention times of 20.24standards.

Conclusion
This study showed that biodiesel yield from spent cooking oil can be successfully improved using the Response Surface Methodology (RSM) optimization tool. Fuel properties of produced biodiesel, which is comparable to both ASTM and EU standards, further give credence to this method of optimization. However, further research is needed to improve the yield quality of biodiesel produced.