Production of biodiesel from waste vegetable oil using impregnated diatomite as heterogeneous catalyst

Edward Modiba ,Christopher Enweremadu ,Hilary Rutto ,*

1 Department of Chemical Engineering,Vanderbijlpark Campus,Vaal University of Technology,Private Bag X021,Vanderbijlpark 1900,South Africa

2 Department of Mechanical and Industrial Engineering,University of South Africa,UNISA Science Campus,Private Bag X6,Florida 1710,South Africa

Keywords:Central composite design Transesterification Impregnation Diatomite

ABSTRACT In this study,biodiesel was produced from waste vegetable oil using a heterogeneous base catalyst synthesized by impregnating potassium hydroxide(KOH)onto diatomite.Response surface methodology based on a central composite design was used to optimize four transesterification variables:temperature(30-120°C),reaction time(2-6 h),methanol to oil mass ratio(10%-50%)and catalyst to oil mass ratio(2.1%-7.9%).A quadratic polynomial equation was obtained to correlate biodieselyield to the transesterification variables.The diatomite-KOH catalyst was characterized using X-ray diffraction(XRD),Fourier transform infra-red spectroscopy(FTIR)and a scanning electron microscope(SEM)equipped with an energy dispersive X-ray detector(EDS).A maximum biodiesel yield of 90%(by mass)was obtained.The reaction conditions were as follows:methanol to oil mass ratio 30%,catalyst to oil mass ratio 5%,reaction time 4 h,and reaction temperature 75°C.The XRD,FTIR and SEM(EDS)results confirm that the addition of KOH modifies the structure of diatomite.During impregnation and calcination of the diatomite catalyst the K2O phase forms in the diatomite structural matrix and the active basicity of this compound facilitates the transesterification process.It is possible to recycle the diatomite-KOH catalyst up to three times.The crucial biodiesel properties from waste vegetable oil are within the American Standard Test Method specifications.

1.Introduction

Biodiesel,a clean renewable fuel,has recently been considered as the best candidate for a diesel fuel substitution because it can be used in any compression ignition engine without the need for modification[1].Most of biodiesel production feed stocks come from edible oils and non-edibles.Non-edibles that have been used to produce biodiesel include animal fats,rice bran,and yellow grease[2-4].Edible vegetable oils such as canola,soybean and corn have been used for biodiesel production and found to be good as a diesel substitute[5].The production of biodiesel is performed through transesterification of triglycerides using alcohol in the presence of a catalyst[6].

Different catalysts have been explored and used for biodiesel production from seed oils using homogenous bases such as potassium hydroxide and sodium hydroxide and homogenous acids such as sulfuric acid[7].However,homogeneous base catalysts such as sodium hydroxide for the production of biodiesel cannot be recovered or regenerated after the reaction and produce toxic wastewater.Heterogeneous solid catalysts from egg shell,tungstated zirconia,activated lime,calcium oxide and zeolites have been used[8-13].Heterogeneous catalysts,for example,can be prepared by impregnating a homogeneous catalyst on bentonite surface structure[14].Heterogeneous solid catalysts have been studied as substitutes for homogeneous catalysts,with the advantage of being easy to recover and reuse,and they also offer the major advantages of easy separation and purification of the final product.

Diatomite is a siliceous,porous and low density[15,16]material,which has been used as a filter aid[17],insulating material[18],insecticide,catalyst support,etc.[19].Diatomite is normally calcined at temperature about 900°C for its industrial application.Diatomite can also be chemically treated or activated to modify its porous surface structure for several applications in water treatment.Diatomite contains traces of K2O and NaO so that it can be used as a catalyst in biodiesel production.Diatomite as a siliceous material can undergo a pozzolanic reaction to form hydrated silicates[20].

Optimizations using surface response methodology have been applied in the production of biodiesel from manketti oil,rapeseed oil and palm oil[21-23].The main objective of this work is to produce biodiesel from waste vegetable oil using impregnated diatomite with KOH as a heterogeneous catalyst.Surface response methodology technique is applied using a central composite design(CCD)to optimize the process.The reaction process variables under investigation are the methanol to oilratio and catalystto oilratio,reaction time and reaction temperature.A mathematical model is developed to correlate the yield of methyl esters(biodiesel)to the transesterification process variables.

2.Materials and Methods

2.1.Materials

The waste vegetable oil was obtained from the Vaal University of Technology cafeteria and calcined diatomite was supplied by In figro Pty South Africa.Potassium hydroxide,isopropanol,methanol and phenolphthalein indicator were supplied by Rochelle Chemical,a local chemical supplier.

Reference standards of fatty acid methyl esters(FAMEs),including palmitic,stearic,palmitoleic,linoleic,oleic,alpha linolenic,icosnoic,iconsenoic and docosenoic methyl ester,all with purity greater than 99%,were purchased from the Sigma Chemical Co.Ltd.The X-ray fluorescence(XRF)analysis of the diatomite shows that the chemical composition of diatomite is as follows in mass percentage:SiO288.02%,Fe2O31.5%,CaO 0.3%,Na2O 2.22%,AI2O33.68%,TiO20.01%,MgO 0.28%,K2O 2.79%and loss on ignition 1.2%.

2.2.Experimental design

The experimental design selected for this study was CCD,which helps in investigating linear,quadratic,cubic and cross-product effects of the four transesterification process variables independently on the yield of biodiesel(response)[24].The four transesterification process variables studied were methanol to oil ratio(X1),catalyst to oil ratio(X2),reaction time(X3)and reaction temperature(X4),with each variable being considered as two levels:low(?2)and high(+2).Table 1 shows the range and the levels of the four transesterification variables.For each categorical variable,a 24fullfactorial CCD for the four variables,consisting of16 non-center points and 5 replicates at the centerpoints is used.A full experimental design matrix is shown in Table 2.

Table 1 Levels of transesterification process variables employed

Table 2 Experimental design and results

A mathematical model is developed to correlate the yield of methyl esters(biodiesel)to the transesterification process variables through first order,second order and interaction terms according to the following third order polynomial equation:

where Y is the predicted biodiesel yield,Xiand Xjrepresent the parameters,βois the offset term,βiis the linear effect,βijis the first-order interaction effect,βjjis the squared effect,and βjjjis the second-order interaction effect.

2.3.Model fitting and statistical analysis

Design expert(6.0.6)software is used as a regression analytical tool to fit experimental data to the third-order polynomial regression model.The statistical significance of the model is evaluated.

2.4.Catalyst preparation

KOH/diatomite blends(catalyst)were dry-mixed according to following ratios:1:3,1:4,1:5 and 1:6.An Erlenmeyer flask(500 ml)was fitted with a reflux condenser.A magnetic heating stirrer was used to heat the blend of KOH/diatomite,at 60°C for 24 h with a stirring speed of 400 r·min?1.After the impregnation process,the slurry was dried in an oven at 500°C for 4 h.

2.5.Catalyst characterization

X-ray diffraction(XRD)analysis was conducted using a PANalytical Empyrean diffractometer,and PIXcel detector fixed slits with Fe filtered Co-Ka radiation.Qualitative analysis of diatomite-KOH catalyst was conducted using the KBr method with the Fourier transform infra-red spectroscopy(FTIR)technique.The FTIR analysis was carried out at a wavenumber range of 4000-500 cm?1using a Shimadzu 8400s FTIR apparatus.The scanning electron microscope(SEM model JSM-7001F)equipped with an energy dispersive X-ray detector(EDS)was used to study the structural morphology and elemental composition of the catalysts.

2.6.Biodiesel production

The waste vegetable oil was heated at 100°C to remove water.The free fatty acid content of waste vegetable oil was determined using titration,which was below 2%using the method described by Van Gerpen et al.[25].Therefore,a one-step alkali transesterification was only required.Transesterification of waste vegetable oil was carried out in a temperature-controlled hot plate equipped with a reflux condenser and magnetic stirrer.The impregnated diatomite was used as a catalyst to produce biodieselunder the following conditions:temperature 60°C,reaction time 3 h,methanol to oil mass ratio 30%,catalyst to oil mass ratio 5%,and stirring speed 400 r·min?1.The catalyst with the highest biodiesel yield was chosen for the study.

The mixture of biodiesel,methanol and glycerol was placed in a decanter for settling.The catalyst and glycerol layers were separated and the biodiesel phase was washed thoroughly with water and then heated at 100°C to remove water and excess methanol.The experiment was conducted according the design of experiment as shown in Table 2.At the end of the reaction,the catalyst was filtered,washed with distilled water and dried in an oven at 120°C.To test the recyclability the catalyst was reused under the conditions which gave the maximum yield of biodiesel.The fatty acid methyl ester(FAME)analysis was conducted on the biodiesel using a gas chromatograph(GC).

2.7.Determination of fatty acid methyl esters

The biodiesel products were analyzed by a Varian GC(model number CP3400)equipped with an autosampler(model CP3800).A polysiloxane coated column of length 30 m,i.d.0.3 mm and film thickness 0.53 μm was used.The oven temperature was kept 90 °C for 1 min,increased at 15 °C·min?1to 230 °C and held for 2 min and then ramped at 5 °C·min?1to 380 °C and held for 2 min.The injector temperature was started at 90 °C and ramped to 380 °C at a rate of 5 °C·min?1while the detector temperature was maintained at 380°C throughout the reaction.The samples were prepared using hexane as solvent and 0.5 μl was injected into the GC.Calibration standards were prepared from a known concentration of FAME solution,and then the standards were used to plota calibration curve for determination of methyl esters.The yield of biodiesel was determined using the following equation:

2.8.Characterization of waste vegetable oil biodiesel fuel properties

The biodiesel produced under the conditions which gave the maximum yield was measured using the biodiesel standard of the American Standard Test Method(ASTM).The viscosity,density and flash point were determined using the ASTM D445,ASTM 1298 and ASTM D93,respectively.

3.Results and Discussion

3.1.Chemical characterization

3.1.1.XRD analysis

Fig.1.The XRD pattern of pure diatomite and KOH/diatomite catalyst blend at different ratios.

Identification of crystallinity,bulk phase and internal structure on impregnated diatomite catalyst was performed using XRD analysis.Fig.1 shows typical X-ray diffraction peaks of pure diatomite,KOH:diatomite 1:4 and KOH-diatomite 1:6.Pure diatomite catalyst diffraction peaks at Bragg angle(2θ)are at 23.36°,26.78°,35.97°,and 44.36°.Cristobalite(SiO2)is found to be the major crystalline phase at Bragg angle(2θ)of 21.93°,35.85°,and 44.51°and these relate fairly to the chemical composition of diatomite.Other minor crystalline phases found are microcline[K(AlSi3O8)],kalicinite[K(CHO3)],and pectolite[NaCa2Si3O8(OH)].The intensity of the catalyst decreases gradually as KOH increases,showing that the crystalline structure of the catalyst reduces and a new amorphous structure becomes prominent.During calcination of the diatomite catalyst,it is possible to form CaO,Na2O,K2O,and MgO in the diatomite structural matrix and the active basicity of these compounds facilitates the transesterification process.

3.1.2.FTIR analysis

The study on impregnated diatomite with KOH solids by FTIR spectroscopy confirms the presence of silica on them.The spectra ofpure diatomite,KOH:diatomite 1:4 and KOH:diatomite 1:6 are shown in Fig.2.The wide band centered at ca.1090 cm?1is caused by Si-O-Si asymmetric stretching.Another observation can be seen at ca.790 cm?1.FTIR results show that the addition of KOH affects the structure of diatomite network,reducing the crystallinity.This is in excellent agreement with XRD results.The diagram shows the presence of three new functional groups(active site)at ca.705,1650 and 1440 cm?1.This could be due to the stretching vibration of the Si-O-K group.This active site could support the transesterification process.The formation of new functional groups could be attributed to the hydration and pozzolanic reaction where a pozzolanic material(diatomite)reacts with potassium hydroxide to form potassium hydrate silicate[26,27].This is shown by the following equations.

Fig.2.FTIR spectra of pure diatomite and KOH/diatomite catalyst blend at different ratios.

KOH dissolution:

Reaction to form calcium silicates:

Fig.3.SEM and EDS of pure diatomite(a),KOH:diatomite 1:6(b),and KOH:diatomite 1:4(c).

3.1.3.SEM and EDS analysis

Fig.3 shows SEM and EDS of pure diatomite,KOH:diatomite 1:4 and KOH-diatomite 1:6.The structure of pure diatomite is more porous than that of impregnated diatomite catalyst.As the amount of KOH increases the structural porosity of the catalyst reduces due to agglomeration of particles,making the morphology appearance look irregular.The intensity of morphology irregularly increases with the increase of KOH.The EDS analysis results show that K2O active sites increase from 0.58%to 11.59%as KOH loading to diatomite structure increases.This result is in excellent agreement with SEM morphological results.

3.2.Reaction mechanism of transesterification using impregnated KOH into diatomite

Fig.4.Proposed reaction mechanism of transesterification using KOH/diatomite catalyst.

Fig.4 shows the transesterification mechanism of biodiesel using KOH/diatomite catalyst.The mechanism is outlined as follows:in the presence of various active sites such as Na2O,K2O,and CaO on the diatomite catalyst,it is highly possible that all these active sites play a role in the transesterification process though K2O composition is highly present compared to other active sites.Initially,1 mol of potassium oxide react with 2 mol of methanol to form potassium methoxide(KCH3O),hydrogen ion(H+)and hydroxyl ions(OH?).As the reaction progresses,potassium methoxide dissociates in to potassium ion(K+)and methoxide ion(CH3O?).Because methoxide is a strong organic base catalyst,it has high catalytic active sites that facilitate the transesterification reaction[11].Secondly,the methoxide has strong nucleophilic affinities,which facilitate the attack of methoxide to a trigonal planar of the carbonyl group(triglyceride)to form tetrahedral intermediate,from which ion diacyl-glycerol and methyl ester form[28].Thirdly,diacyl-glycerol molecule forms in the reaction of hydrogen ion(H+)with diacyl-glycerol ion,while alternative diacyl-glycerol ion reacts with the second methanol molecule to form a regenerated and active methoxide ion,starting another catalytic cycle.Finally,the methoxide ion reacts with diacyl-glycerol molecule to form methyl ester molecule and monoglyceride molecule.Similarly,the present Na2O,and CaO could follow the same transesterification mechanism as that of K2O.

3.3.Effect of KOH/diatomite ratio on the yield of biodiesel

Fig.5 shows that as the amount of diatomite increases,from 75%to 85.7%,the biodiesel yield increases and vice versa.This could be because the K2Ophase forms with more diatomite and this increases the basicity and catalytic activity as compared to the formation of the Al-O-K phase with more KOH.Similar results were obtained when KOH was impregnated onto bentonite and zeolite[14,29].

Fig.5.Effect of KOH/diatomite ratio on the yield of biodiesel from waste vegetable oil.

Fig.6.Experimental data versus predicted biodiesel yield values.

3.4.Development of regression equation and analysis of variance

The final equation in terms of actual values after excluding insignificant terms is identified using Fisher's Test.

The negative sign in front of the terms indicates antagonistic effect,while the positive sign indicates synergistic effect.The value of R2for Eq.(3)is 0.9842,implying that 98.4%of the total variation in the transesterification responses is explained by the model.Fig.6 shows the predicted values versus the experimental data of biodie-sel yield plotted against a unit slope.The results show that the regression equation provides an accurate description of the experimental data.

Based on the 95%confidence level,the model is significant as the computed F value(78.801)is much higher than the theoretical F0.05(14,6)value of 3.96,indicating that the regression model is reliable in predicting the yield of biodiesel.Table 3 shows that for the four individual variables studied the reaction temperature(X4)and methanol to oil ratio(X1)have a huge influence on the yield of biodiesel.The catalyst to oil ratio(x2)and reaction time(X3)has little effect on the yield.The cubic terms do not affect the yield of biodiesel.Generally interaction terms(X1X2),(X2X4)and(X3X4)have very large influence on the yield in the descending order,while the interaction terms(X1X3)and(X2X3)have very low impact.The interaction term(X1X4)has the least effect on the biodiesel yield.

3.5.Effects of transesterification variables

3.5.1.Effect of catalyst to oil mass ratio and methanol to oil mass ratio on the yield of biodiesel

Fig.7 shows the effect of catalyst to oil mass ratio and methanol to oil mass ratio on the transesterification of waste vegetable oil,with the reaction temperature and time held constant at 75°C and 4 h respectively.At low methanol to oil ratio,the methyl ester yield is higher at higher catalyst to oil ratio.However,at high quantity of catalyst,as the methanol to oil ratio increases the biodiesel yield decreases significantly.Higher methanol to oil mass ratio increment leads to the conversion of glycerol and biodiesel back to triglycerides(glycerolysis)and thus the biodiesel yield decreases[30].

3.5.2.Effect of temperature and time on the yield of biodiesel

Fig.8 shows the effect of temperature and time on the transesterification of waste vegetable oil,with methanol to oil mass ratio and catalyst to oil mass ratio held constant at 30%and 5%,respectively.At lower temperature,the amount of methyl ester from waste vegetable is higher,as the transesterification period is prolonged and the biodiesel yield increases significantly.However,at higher temperature as the transesterification time increases the biodiesel yield reduces to a greater extent.High temperature diminishes the residence molecular interaction time between methanol,oil and catalyst,causing thermal degradation and reducing the biodiesel yield.Loss of methanol at higher temperature above 60°C(boiling point)due to evaporation could also reduce the biodiesel yield[31].

3.5.3.Effect of methanol to oil mass ratio and reaction time on the yield of biodiesel

Fig.9 shows the effect of reaction time and methanol to oil ratio on the yield of biodiesel from waste vegetable oil,with the temperature held constant at 75°C and catalyst to oil mass ratio at 5%,respectively.With longer reaction period the biodiesel yield is higher.Short reaction period leads to incomplete reaction as there is less molecularinteraction between triglyceride and methanol and thus the biodiesel yield is reduced.As the methanol to oil ratio increases the biodiesel yield reduces for both high and low reaction periods.Excess methanol causes glycerolysis reaction to occur,forming monoglyceride instead of biodiesel.

3.5.4.Optimum condition

Since no maximum was examined within the experimental domain,the modeldeveloped cannotgive the optimumcondition for the highest biodiesel yield.

3.5.5.Individual influence of reaction variables on the yield of biodiesel

The perturbation plot shows the effect of all variables on transesterification of waste vegetable oil.The effect of all reaction variables at a point in the design space can be discussed from the perturbation plot as shown in Fig.10.Influence of one factor is evaluated and plotted alongside the yield while the other parameters are kept constant.Reaction temperature and methanol to oil mass ratio present greater effect on biodiesel yield.Reaction time and catalyst to oil mass ratio show the least influence on the yield of biodiesel.The result is in excellent agreement with the result obtained from the analysis of variance.It can generally be observed that the biodiesel yield decreases as reaction temperature and methanol to oil mass ratio increase,while there is a very slight increase in the biodiesel yield as the reaction time and catalyst to mass oil ratio increase.

Fig.7.Effect of catalyst to oil mass ratio and methanol to oil mass ratio on the transesterification of waste vegetable oil.

Fig.8.Effect of temperature and time on the transesterification of waste vegetable oil.

Fig.9.Effect of time and methanol to oil mass ratio on transesterification of waste vegetable oil.

Fig.10.Effects of individual variables on transesterification of waste vegetable oil to biodiesel.A—methanol to oil mass ratio;B—catalyst to oil mass ratio;C—reaction time;D—reaction temperature.

Fig.11.Effect of catalyst reusability.

Table 4 Fuel properties of waste vegetable biodiesel

3.6.Catalyst recyclability

After re-use three times of the catalyst as depicted in Fig.11,biodiesel yield obtained in the first,second and third runs were 88.6%,85.6%and 84.2%,respectively.The biodiesel was produced under the conditions which produced the maximum yield.This suggests that it is possible to re-use and recycle the heterogeneous catalysts[14].

3.7.Fuel properties of waste vegetable oil methyl ester compared to other oil methyl ester

Important fuel properties of biodiesel from waste vegetable oil(WVO)were determined and compared to those of jatropha[32]and palm[33]oils as shown in Table 4.The fuel properties of biodiesel produced from waste vegetable oil are within the ASTM standards of biodiesel.The density of waste vegetable oil methyl ester is 861 kg·m?3,which is lower than that of jatropha(880 kg·m?3)and palm methyl ester(864.4 kg·m?3).All are within the specified limit of 860-900 kg·m?3.The kinematic viscosity of waste vegetable oil methyl ester(4.22 mm2·s?1)at 40 °C is slightly lower than that of jatropha(4.4 mm2·s?1)and palm oil(4.5 mm2·s?1),all meeting the ASTM standard of biodiesel viscosity.The flash point of biodiesel from waste vegetable oil is also within the ASTM standard.

4.Conclusions

This work shows that it is feasible to synthesize a heterogeneous catalyst from diatomite through an impregnation process and produce biodiesel from waste vegetable oil using the catalyst.Using a surface response methodology a maximum biodiesel mass yield of 90%was obtained.The reaction conditions were as follows:methanol to oil mass ratio 30%,catalyst to oil mass ratio 5.0%,reaction time 4 h,and reaction temperature 75°C.XRD,FTIR and SEM(EDS)results confirm that the addition of KOH modifies the structure of diatomite.During impregnation and calcination of the diatomite catalyst the K2O phase forms in the diatomite structural matrix and the active basicity of this compound facilitates the transesterification process.It is possible to recycle diatomite-KOH catalyst up to three times without any significant change in its catalytic activity.The common biodiesel properties from waste vegetable oil are within the American Standard Test Method.

Acknowledgments

The support by the centre of research excellence(Vaal University of Technology)grant no 2188-2892 to fund this project is gratefully acknowledged.