Biodisel From Castor Oil

Embed Size (px)

DESCRIPTION

biorefinery castor plant

Citation preview

DocumentPlease cite this article in press as: Bateni, H., Karimi, K., Biodiesel production from castor plant integrating ethanol production via a bioreneryapproach. Chem. Eng. Res. Des. (2015), http://dx.doi.org/10.1016/j.cherd.2015.08.014 ARTICLE IN PRESSCHERD-1994;No.of Pages9chemicalengineeringresearchanddesignxxx(2015)xxxxxx ContentslistsavailableatScienceDirectChemical Engineering Research and Design journal homepage:www.elsevier.com/locate/cherd BiodieselproductionfromcastorplantintegratingethanolproductionviaabioreneryapproachHamedBatenia,b,,KeikhosroKarimia,c,aDepartmentofChemicalEngineering,IsfahanUniversityofTechnology,Isfahan84156-83111,IranbChemicalandBiomolecularEngineeringDepartment,OhioUniversity,Athens,OH45701,UnitedStatescInstituteofBiotechnologyandBioengineering,IsfahanUniversityofTechnology,Isfahan84156-83111,Irana r t i c l ei n f o Articlehistory:Received 21 April 2015Received in revised form 5 August2015Accepted 17 August 2015Available online xxx Keywords:Castor plantBiodieselEthanolAlkali pretreatmentIntegrated processa b s t r a c t Biodiesel, a promising alternative fuel, is not a completely renewable fuel, as it currently usesoil-based methanol for its industrial production. Integrated biodiesel and bioethanol produc-tion in a biorenery unit can overcome this challenge together with an improved economy.In this study, castor plant was applied to an integrated biodiesel and ethanol production.The extracted oil was transesteried with ethanol produced through simultaneous saccha-rication and fermentation of the castor plant residue. An alkaline pretreatment using 8%w/v sodium hydroxide at 100 C for 60min was applied to improve the ethanol produc-tion yield from 27.2 to 71.0%. An experimental design using response surface methodology(RSM) was used to optimize the biodiesel production yield. The optimum biodiesel yield was85.01.0%, obtained at 62.5 C using an ethanol to oil mass ratio of 0.29:1 for 3.46h, whichwas in agreement with the predicted yield (84.4%). Accordingly, 1kg of castor plant resultedin production of 149.6 g biodiesel and at least 30.1 g ethanol as the nal products with noextra alcohol feedstock requirement. 2015 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved. 1.IntroductionBiodiesel has attracted a great deal of attention over thelast few decades due to its environmental compatibility andbiodegradability (Aarthy et al., 2014; Reyero et al., 2014). Trans-esterication of triglycerides u sing methanol in the presenceof an alkaline catalyst is the most common process forbiodiesel production (Haigh et al., 2014; Perdomo et al., 2014).Ethanol can be used as a harmless substitute for methanol intransesterication reaction besides its individual applicationas a renewable alternative fuel. Furthermore, ethanol is pro-duced through the fermentation of renewable biomass whilemethanol is industrially produced from fossil fuel sources.However, the production of methanol from biomass with anintegrated biorenery prospective has been widely studied Correspondingauthorat :ChemicalandBiomolecularEngineeringDepartment,OhioUniversity,Athens,OH45701,UnitedStates.Tel.: +1 7405925650; fax: +1 7405930873.Corresponding author at : Ins titute of Biotechnology an d Bioengineering, Isf ahan Univer sity of Technology , Isfahan 84156-83111, Iran.Tel.: +98 3113915623; fax: +98 3113912677.E-mail addresses: [email protected], [email protected] (H. Bateni), [email protected] (K. Karimi).(Ng and Sadhukhan, 2011; Martine z -Hern ande z et al., 2014).Therefore, biodiesel production using bioethanol results ina completely renewable biofuel (Lam and Lee, 2011). Globalbiodiesel and ethanol production were about 25 and 87 bil-lion liter s in 2 010, respectively, predicted to correspondinglyincr ease to over 58 and 161 billion liters by 2020 (Pinto,2011).According to the U.S. Department of Energy, ethanol,biodiesel, and regular diesel prices were around $0.42, $1.06,and $0.82 pe r liter at the beginning of 2015, demonstratingthat biodiesel still cannot compete with diesel in the mar-ket (U.S. Department of Energy, 2015). Over 70% of biodieselcost is c orrelated to oil f eedstock which can be vegetableoil (edible or non-edible), waste cooking oil, or animal fat(Dias et al., 2013). The use of inexpensive non-edible oil, asa second generation feedstock, not only is economically morehttp://dx.doi.org/10.1016/j.cherd.2015.08.0140263-8762/ 2015 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved. Please cite this article in press as: Bateni, H., Karimi, K., Biodiesel production from castor plant integrating ethanol production via a bioreneryapproach. Chem. Eng. Res. Des. (2015), http://dx.doi.org/10.1016/j.cherd.2015.08.014 ARTICLE IN PRESSCHERD-1994;No.of Pages92chemicalengineeringresearchanddesignxxx(2015)xxxxxx NomenclatureANOVAanalysis of varianceCCDcentral composite designGCgas chromatographHPLChigh performance liquid chromatographRSMresponse surface methodologySEMscanning electron microscopySSFsimultaneous saccharication and fermenta-tion favorable but also can reduce food versus fuel conict (Bateniet al., 2014). However, precise planning and management areneeded to identify and minimize further challenges towardsustainable development of biodiesel production (Sukkasiet al., 2010). In fact, it should be considered that cultivation ofnon-edible oil cr op may be associated with using arable lan dand consequently increasing load on soil, water and biodi-versity (Antizar-Ladislao and Turrion-Gomez, 2008). However,some of the non-edible plants, e.g., castor oil plant, can becultivate d in the wastelands irrigated by w astewater whichcan suppress the aforementioned challenges (Chakrabarti andAhmad, 2008; Tsoutsos et al., 2013).Ethanol is industrially produced from sugar and starchbased materials; however, it can be produced from inex-pensive and abundant l ignocellulosic substrates such asagriculturalresidue,forestresidue,and municipal solid wastes(Du et al., 2009; Karimi et al., 2013; Misailidis et al., 2009).Lignocell uloses are dominantly composed of cellulose , alongwith he micell ulose and lignin in a highly compact structure(Karimi and Chisti, 2015; Karimi and Pandey,2014; Karimi et al.,2013).Therefore, a pretreatment process is required to removelignin and hemicell ulose and increase the accessibility o f theenzyme to cellulose prior to the hydrolysis and fermentationprocesses (Asgher et al., 2014; Bateni et al., 2014). Alkaline pre-treatment is one of the most promising chemical processeswhich can effectively facilitate further processes. Alkaline pre-treatment using sodium hydroxide was frequently ap plied forsoftwoo ds, hardwoods, and agricultu r al residues to improveethanol yield and showed promising results (Goshadrou et al.,2011; Mirahmadi et al., 2010; Mohsenzadeh et al., 2011; Nieveset al., 2011; Salehi et al., 2012; Salehian and Karimi, 2013).Sodium hydroxide pretreatment resulted in lignin removal,biomass swelling, more surface reacti v e functional groups likehydroxyl grou ps, and surface roughness improvement in thepretreated substrate (Islam et al., 2012; Teghammar et al.,2010).Although alkaline pretreatment, e.g., NaOH pretreat-ment, is associated with high alkali loading an d a large volumeof waste water for washing, it received great interest since itdoes not need complicated reactors (Wan et al., 2011). Sodiumhydroxide pretreatment are classied into dilute and concen-trated treatments. High concentration of sodium hydroxideis applied at moderate conditions, i.e., ambient pressure andrelatively low temperatures. Dissolution of cellulose at theseconditions is the main phenomenon, resulting in changingcellulose I (c rystalline cellulose ) to amorphous cellulose withlow degree of polymerization amenable to enzymatic hydroly-sis (Karimi et al., 2013). Moreover, high concentration sodiumhydroxide provides an opportunity to recover and reuse thepretreat ment solution, resulting in lower chemical waste dis -posal and consequently less environmental concerns (Karimiet al., 2013; Mirahmadi et al., 2010; Wan et al., 2011).The d evelopment of biof uels production is still limited dueto lack of economic prociency (Zhu and Zhuang, 2012; Karimiand Pandey, 2014). Biorenery, which is a facility with inte-grated, efcient and exible biomass processing to variousvalue-ad ded products and energy , can suppress the eco-nomic barriers and lead to the procient biofuel production(Sadhukhan et al., 2014; Luo et al., 2010; Zhu et al., 2014). Theconcept of biorenery was deve loped analogous to the cur-rent complex crude oil reneries leading to multiple fuels andvalue-added chemicals (Sadhukhan et al., 2014).Castor plant is an important non-edible oilseed crop whichcan toler ate various climate condi tions (humid tropic todry subtropic) and has a wide range industrial a pplications(Ramanj aneyu lu et al., 2013). Currently, the average global cas-tor se ed production yield is about 1.1t/ha (Scholz and da Silva,2008)with a productivity of 470 kg oil/ha considering seed oilcontents in a range of 4555% (Santana et al., 2010). Castorplant cul tivation under f avorable conditions can boost theseed yield to 45t/ha, leading to promising oil productivity(Scholz and da Silva, 2008).Castor oil consists mainly of ricinoleic acid (12-hydroxy-cis-o ctadec-9-enoic acid), an hydroxylated fatty acid with onedouble bond. Th e presence of ricin oleic acid in castor oilresults in some unique prop erties, e.g., very high solubility inalcohol (Perdomo et al., 2013). Therefore castor oi l can be con-verted to biodiesel (Klc et al., 2013; Perdomo et al., 2013) evenat low te mperature (Bateni et al., 2014). Castor biodiesel hasa hi gh caloric value and high cetane number (Scholz and daSilva, 2008). Howeve r , some properties of the nal biodiesel,e.g., high kinematic viscosity, may complicate its engine per-form ance (Berman et al., 2011; Scholz and da Silva, 2008).Castor b iodiesel can be dilu ted or blended with conventionaldiesel to produce an appropriate fuel for diesel motors (Scholzand da Silva, 2008).The presence of ricinoleic acid in the castor seed cakeresults in a more limited application compared to the otherseed cakes. Castor seed cake can be use d for production oforganic nitrogenous fertilizer due to their nitrogen, phos-phorous and potassium content (Ramachandr an et al., 2007).How ever, it cannot be directly used for animal feed du e tothe p oisenous and allergic properties (Ogunniyi, 2006; Adedejiet a l., 2006). Previous study (Bateni et al., 2014) showed thegreat potential of castor seed cake for biogas production.Previous studies have investigated the inuence of reac-tion parameters incl uding temperature , reaction time , catalys tconc entr ation, and alcohol content on transesterication ofcastor o il (Ca v alcante et al., 2010; Da Silva et al., 2006; DeOliveira et al., 2005; Jeong and Park, 2009; Ramezani et al.,2010).De Oliveira et al. (2005) used the Taguchi e xperimen-tal desig n to evaluate the effect of reaction conditions oncommercial rened castor oil. They (De Oliveira et al.,2005)deduced that increasing reaction temperature (3070 C)and time (13 h) positively affected the reaction conversion,whereas increasing catalyst concentration (0.51.5 w/w%)yielded a negative effect. In addition, it was conclud ed thatincreasi ng the a lcohol concentration had no signicant effecton the conversion (De Oliveira et al., 2005). Da Silva et al.(2006) used response surface methodology to optimize thetransestrication parameters of a commercial castor oil.Optimum results were achieved with the lowest tem pera-ture and the h i ghest alcohol concentra tion in the presence of0.81.3 w/w% sodium methoxide as a catalyst (Da Silva et al.,2006).Ramezani et al. (2010) evaluated the potential of dif-ferent basic catalysts (NaOCH , KOCH , NaOH, KOH) together33 Please cite this article in press as: Bateni, H., Karimi, K., Biodiesel production from castor plant integrating ethanol production via a bioreneryapproach. Chem. Eng. Res. Des. (2015), http://dx.doi.org/10.1016/j.cherd.2015.08.014 ARTICLE IN PRESSCHERD-1994;No.of Pages9chemicalengineeringresearchanddesignxxx(2015)xxxxxx3with different reaction conditions to optimize the methyl estersynthesis from rened castor oil. The optimum b iodieselyield wa s obt ained at 65 C using 8:1 alcohol to o il molarratio after 2h in the presence of 0.5% KOCH3(Ramezani et al.,2010).On the other hand, Dias et al. (2013) used raw castoroil for biodiesel production and evaluated the effects of timeand temperature on the conversion using a central compositeexperimental design in the presence of a 6:1 methanol to oilmolar ratio and 1% w/w KOH. The highest yield of 73.62 w/w%and purity of 83.41 w/w% appeared at 65 C for 8h.Although many research studies have been carried out onthe biodiesel production from castor oil using ethanol, to thebest of our knowledge, none of them used the ethanol pro-duced from the castor plant residues to covert the raw castoroil to biodiesel using a biorening approach. Within this con-text, this study seeks to evaluate the potential of castor plantfor biodiesel production with no external alcohol require-ment via an integrated process leading to ethanol productionas a byproduct. Alkaline pretreatment using 8% w/v sodiumhydroxide was used to improve the ethanol production yield.Further optimization was performed using response surfacemethodology (RSM) to get the best biodiesel yield due to thelack of agreement in the literature for the optimum conditionsof castor oil transesterication.2.Experimental2.1.FeedpreparationThe castor plant was collected from land in Isfahan, Iran. Allparts of the plant were dried using an oven at 35 C for 96h,and the castor seeds were then separated and dried in an ovenat 70 C for an extra 24h . Casto r oil was extracted by Soxhletextractor using n -hexane through the method presented byPerdomo et al. (2013) for 8h. The solvent was then recov-ered by a vacuum rotary evaporator. Thereafter, the castorresidues, i.e., the ste m, seed cake , and le aves, were milled andscreened to achieve a uniform substrate size within a range of0.10.8mm (Bateni et al., 2014).2.1.1.AlkalinepretreatmentAlkaline pretreatment using 8% w/v NaOH solution at 100 Cfor 60min was used as an effective pretreatment method p riorto ethan ol pro duction. The pretreatment involves the follow-ing steps but the details can be found elsewhere (Bateni et al.,2014):1.The residues were mixed with 8% (w/v) sodium hydroxidesolution.2.The slurry was placed in a bath at 100 C and periodicallymixed.3.After 60min, the suspension was ltered and rinsed withdistilled water until neutral pH.4.The washed substrate was dried using a freeze dryer(Christ, Alpha 1-2 LDplus Model, Germany) and kept inresealable bags until use.2.1.2.FeedcharacterizationA gas chromatograph (GC) equipped with a ame ionizationdetector (Beijing Beifen-Ruili, Series Sp-3420A, China) anda capillary column (60m0.25mm0.25m, SolGel-WAX,SGE, UK) was used to determine the fatty acid composition ofthe castor oil. Pure nitrogen gas with a ow rate of 30mL/minwas applied as a carrier gas. The GC temperature programconsisted of holding at an initial temperature of 150 C for3min , increasing the temperature to 210 C with a 5 C/minramp, holding for 6min, increasing to 240 C at a rate of5C/min, and holding for 10min. The temperatures of injec-tor and detector were set at 230 and 250 C, respectively.The fatty acid methyl esters (FAMEs) were prepared throughthemethylationreactionwith0.5Nsodiummethoxide.The FAMEs content was calculated by a calibration curveconsidering the peak area.The extractive content and the composition (carbohydratesand lign in content) of the untreated and p retreated castorresidues were measured using a NREL/TP-510-42618 (Sluiteret al., 2008) and NREL/TP-510-42619 (Sluiter et al., 2005) meth-ods, respectively.Ahigh-performanceliquidchromatograph(HPLC)equipped with UV/Vis (Jasco International Co., Tokyo, Japan)and an ion-exchange Aminex HPX-87P column (Bio-Rad,USA) at 85 C was employed to measure the sugar content forcarbohydrate analysis. Deionized water with a ow rate of0.6mL/min was used as a mobile phase.Scanning electron microscopy (SEM, ZEISS, Germany) wasperformed at 15kV to investigate the effect of pretreatmenton the surface. The substrates were coated with gold (BAL-TECSCD 005, Leica Microsystems, USA) before SEM analysis.2.2.EthanolproductionSimultaneous saccharication and fermentation (SSF) pro-cesses were conducted at 37 C and 130rpm under anaerobicconditions for 96h in 2.5L glass bottles. A occulating strainof Saccharomyces cerevisiae (CCUG 53310, Culture Collectionof Gothenburg University, Sweden) was used as a ferment-ing microorganism. Fermentation media were prepared using5g/L yeast extract, 3.5 g/L (NH ) SO , 0.75 g/L MgSO42447H O,21g/L CaCl22H O, and 50g/L substrates in 50mM buffer citrate.2The mixture pH was adjusted to 5, poured into the fermenta-tion bottle, and autoclaved at 121 C for 20min. After coolingthe suspension to roomtemperature,1g/L S. cerevisiae togetherwith 15 F PU cellulase and 30 IU-glucosidase per gram of sub-strates (based on dry weight) were added to the fermenter(Bateni et al., 2014). The ethanol production yield was calcu-lated using the following equation (Salehian and Karimi, 2013):Ethanol yield(%)=Produced ethanol(g/L) 1.1110.5150(g/L)glucan(%)100(1)Ethanol production was conducted using untreated and pre-treated castor plant residue to conrm the ethanol yieldimprovement. The produced ethanol was puried using threesequence distillation steps followed by a dehydration step bymolecular sieves (4A, SigmaAldrich). An HPLC equipped withUV/Vis (Jasco International Co., Tokyo, Japan) and an AminexHPX-87H column (Bio-Rad, USA) at 60 C was used to analyzethe SSF samples for determination of ethanol content. Sulfu-ric acid (5mM) with a ow rate of 0.6mL/min was used as aneluent.2.3.BiodieselproductionPrior to t he transest erication reaction, castor oil was rst l-tered by an 80 mesh sieve to remove the solid particles (Bateniet al., 2014) and then preheated at 80 C for 60min to eliminate Please cite this article in press as: Bateni, H., Karimi, K., Biodiesel production from castor plant integrating ethanol production via a bioreneryapproach. Chem. Eng. Res. Des. (2015), http://dx.doi.org/10.1016/j.cherd.2015.08.014 ARTICLE IN PRESSCHERD-1994;No.of Pages94chemicalengineeringresearchanddesignxxx(2015)xxxxxx Table1Experimentalconditionsforthesurfaceresponseanalysis.LevelsFactors Time (h)Temperature ( C)Ethanol to oilmass ratioX1X2X3 11500.203600.315700.4 the extra moisture (Berman et al., 2011). A 500 mL jacketedbatch reactor equipped with a condenser and mechanicalmixer was used for biodiesel production. Ethanol produced inthe previous step was applied to the transesterication reac-tion inthe presence of 1% w/w KOH. Theeffects of temperature(5070 C), time (15 h), and ethanol to oil mass ratio (0.20.4)on the biodiesel yield were evaluated. The range s of reactionparameters were selected according to the previous studiesconsidering an economic point of view (Dias et al., 2013). Th eproduct was le ft in a separation funnel overnight to separatebiodiesel (upper phase) from glycerol (lower phase) (Klc et al.,2013).Next, the biodiesel was was hed with hot distilled water .The yield was determined using the following equation basedon dehydrated biodiesel (Bateni et al., 2014; Klc et al., 2013):Biodiesel yield(%)=Produced biodiesel(g) Raw oil(g)100(2)Further characterizations were performed on the samplewith the hi ghest yield. Free fatty acid content of the sampleoil and biodiesel were measured using the method presentedby Sadasivam and Manickam (2005). Flash point, kinematicviscosity, free glyc erin, and water and sediment content weremeasured, and the results were compared to the standardvalues in A STM D6751 (2003). The heating value (heat of com-bustion) was measured using the standard method presentedin ASTM D240 (2003).2.4.ExperimentaldesignA response surface methodology (RSM) using Minitab 16 wasemployed to evaluate the inuence of reaction parameters onthe biodiesel yield. The experimental design for the reactionwas carried out using a central composite design (CCD) withthree factors, three levels, and six central point repetition. Thevalue of(alpha) was limited at level 1; therefore, the low estand highest levels are assigned to be1and +1, respectively.The experimental design factors are presented in Table 1.A general second order polynomial equation was used tot a correlation between the response (biodiesel yield) and theindependent variables:Y=a+ki=1b Xii+ki=1c Xi2i+ki=1kj>1d X Xijij(3)where Y is the response, a , b , c , and diiijare intercept, linear,quadratic, and interaction constant coefcient, respectively;k is the number of factors studied and optimized in theexperiment; X and X are the independent variables. Reac-ijtion temperature, time, and ethanol to oil mass ratio were theindependent variables in this study. Table2Characteristicsofextractedcastoroil.Properties/componentsContent PropertiesViscosity236.71.4mm /s2Free fatty acid1.00.3%ComponentsRicinoleic acid89.21.2%Linoleic acid4.70.7%Linolenic acid3.80.5%Oleic acid0.40.2% 2.5.CostestimationFinal et hanol and b iod iesel costs were rou ghly estimatedaccording to the economic evaluations performed by Shaeiet al. (2011) and Santana et al. (2010), respectively. The NMMOwas replaced with 8% NaOH solution in Shaei et al. (2011)study as the ethanol production yield was very close tothe present study. It was assumed that castor residue andspruc e wood has approximately the same price. In the caseof biodiesel production, ethano l content was consid ered as adominant factor on the biodiesel price. Therefore, the amountand price of ethanol in Santana et al. (2010) study was replacedby the optimum ethanol content and estimated ethanol priceof the current study, respectively. The effects of other param-eters were neglected in the biodiesel price estimation.3.Resultsanddiscussion3.1.RawmaterialcharacterizationThe extracted oil, whi ch repre sented 17.6% w/w of the wholedried castor plant, dominantly consisted of ricinoleic acid(89.21.2%). Table 2 exhibits other castor oil characteristics.Extractive, carbohydrate, and lignin con t ent of c astor plantresidue were measured before and after the pretreatment,and the results are presented in Fig. 1. The pretreatmentreduced the ethanol extractive of the residue from 7.40.3to 4.50.1%; however, its effect on the water extractive con-tent was not signicant. A relatively high ethanol extractivecontent of castor residue indicated the presence of remainingoil in the seed cake. Alkaline solution (the liquid fraction ofthe pretreatment process) was an appropriate medium for asaponication reaction betweenthe remaining oil inthe castor Fig.1 Composition ofcastorplantresiduebefore(darkgraybars)andafter(lightgraybars)alkalinepretreatmentusing8%(w/v)sodiumhydroxideat 100 Cfor60min. Please cite this article in press as: Bateni, H., Karimi, K., Biodiesel production from castor plant integrating ethanol production via a bioreneryapproach. Chem. Eng. Res. Des. (2015), http://dx.doi.org/10.1016/j.cherd.2015.08.014 ARTICLE IN PRESSCHERD-1994;No.of Pages9chemicalengineeringresearchanddesignxxx(2015)xxxxxx5 Fig.2 SEMimagesofcastorplantresiduebeforeandafterpretreatmentusing8%(w/v)sodiumhydroxideat 100 Cfor60min.residues and sodium hydroxide. A reduction in th e ethanolextractive content of the pretreated castor residue reectedthe progress of the saponication reaction (Bateni et al., 2014).Cellulose was the dominant carbohydrate in the plantresidue, and the pretreatment boosted its content from31.20.3to 44.20.2%. Moreover, the li gnin c ontent of theresidue decreased from 39.20.6to 32.70.4% after thesodium hydroxide pretreatment. Fig. 2 shows the SEM imagesof the castor residue before and after the pretreatment. A com-pact structure with a negligible porosity was observed for theuntreated castor residue, while the alkaline pretreatment sig-nicantly opened up the structure and created a sponge-likeshape su bstrate . Higher por osity of the castor residues pro-vided a more accessible area for further microbial process(Karimi et al., 2013).3.2.EthanolcharacterizationEthanol was produced using a simultaneous saccharicationand fermentation of the untreated and pretreated castorplantresidue.Thepretreatmentincreasedtheethanolproduction yield from 27.2 to 71.0%. The purity of ethanolproduced from the pretreated residue was increased from8.91.1to75.50.8%afterthree sequenceddistillationsteps. Further dehydration by molecular sieves resulted in81.11.6g ethanol (82.2 g with the purity of 98.7%) per kgcastor. According to the previous study (Bateni et al., 2014),around 86g ethanol per each kg castor plant was expected,this means around 7.59.7% ethanol was lost through theglassware equipment during the distillation process. Puriedethanol was used to investigate the effect of temperature,time, and alcohol content on the transesterication yield inthe presence of 1% KOH.3.3.BiodieselproductionThe experimental design matrix and results are presentedin Table 3. The rst three columns of the factors are coordi-nated to dimensionless (coded) levels of the factors and thenext three columns give the factor levels on a natural scale(uncoded).A quadratic polynomial equation was obtained from theexperimental design and data analysis, and the followingequation was obtained to predict the biodiesel yield in termsof the uncoded factors:Y= 338.146+54.167t+9.837T+147.290R5.486t20.072T2189.545R20.218tT8.937tR+0.088TR(4) Table3Experimentalmatrixofthreefactorsinthreelevelsandtheresults.Run orderCoded factorsUncoded factorsBiodiesel yield (%) X1X2X3t (h)T ( C)R11111500.229.721115500.262.731111700.247.141115700.263.951111500.434.161115500.461.271111700.452.481115700.460.891001600.348.7101005600.373.0110103500.370.7120103700.380.4130013600.282.7140013600.479.1150003600.381.9160003600.381.5170003600.382.1180003600.382.4190003600.384.0200003600.383.1 X , t :time; X , T :temperature; X , R :ethanoltooilmassratio.123 Please cite this article in press as: Bateni, H., Karimi, K., Biodiesel production from castor plant integrating ethanol production via a bioreneryapproach. Chem. Eng. Res. Des. (2015), http://dx.doi.org/10.1016/j.cherd.2015.08.014 ARTICLE IN PRESSCHERD-1994;No.of Pages96chemicalengineeringresearchanddesignxxx(2015)xxxxxx Table4Analysisofvarianceforresponsesurfacequadraticmodel.SourceDegree of freedomSum of squaresMean squareF -valueP -valueRegression95567.23618.58336.400.000Linear31414.88642.11349.20.000Square33974.471324.82720.480.000Interaction3177.8759.2932.240.000Residual error1018.391.84Lack-of-t514.252.853.440.101Pure error54.140.83Total195585.62 S =1.35603; R =99.67%; R (pred)=97.72%; R (adj)=99.37%.222 where Y , R , T , and t are the biodiesel yield (%), ethanol to oilmass ratio, temperature ( C), and time (h), respectively. Analy-sis of variance (ANOVA) was used to evaluate the adequacy ofthe empirical mod el, and the results are presented in Table 4.The model was also validated using four sets of data, pre-sented in Fig. 3. As can be seen, the model agreed well w iththe experimental results. The RSM t and parameter estima-tions for the empirical model are summarized in Table 5. TheANOVA results indicate a statistically signicant correlationbetween the biodiesel yield and the three factors studied witha condence level of 95%. The P -values and T -values were usedto checkthe signicance ofthe corresponding coefcients. Thelarger magnitude of the T -values and smaller P -value repre-sent the higher signicance of the corresponding coefcient.Accordingly, all factors (except the corresponding parameterto the temperature and ethanol content ( T R ) interaction) wereshown to ha vea sign icant effect. Moreover,the lack-of-t wasinsignicant due to the relatively high P -value (0.101).Table 5 shows that the rst-order (linear) parameters hada positive effect on the yield; however, all the interactions cannegatively affect the yield. Moreover, all of the second-order(quadratic) coefcients are negative, resulting in a reductionin the response (biodiesel yield) at very high values for thereaction parameters. Ve r y high reactio n temperatures acceler-ate both transesterication and saponication reaction (Leunget al., 201 0; Silva et al., 2011), which not only led to ester andfatty acids lost but also complicated the separation process(Snchez et al., 2015).Besides the temperature, reaction duration showed a sig-nicant effect on the biodiesel production yield. Similarly, amarked increase in the reaction duration resulted in a reduc-tion in the biodiesel yield, probably due to saponication Fig.3 Comparisonbetw ee ntheexperimentalbiodieselyieldandthepredictedyieldby theempiricalmodeldescribedinEq.(4).reaction (Lam and Lee, 2011). The same results were reportedin pr evious studies (Bateni et al., 2014; Leung et al., 2010).Higher alcohol content is favorable in the transesterica-tion process since it can facilitate triglyceride conversion tomonoglycerides. However, monoglyceride can increase glyc-erol solubility in the biodiesel, which not only make s theseparati on process more challe nging but also results in glyc-erolysis reaction leading back to triglyceride (Lam and Lee,2011; Silva et al., 2011).The con tour p l ots for the predicted values of the biodieselyield as a function of the reaction parameters are presentedin Fig. 4. Fig. 4a shows the contour plot as a function of timeand temperature at a 0.3:1 ethanol to oil mass ratio. Fig. 4brepresen ts the contour plot as a function of temperature andethanol to oil mass ratio for a 3h transesterication reaction.Fig. 4c displays the predicted response values as a function oftime and ethanol to oil mass ratio at a temperature of 60 C.According to the experimental design results, the potentialoptimum biodiesel yield was 84.4% at 62.5 C using ethanol tooil mass ratio of 0.29:1 for 3.46h. The predicted biodiesel yieldwas in agreement with the practical yield of 85.01.0%.The presence of water in the ethanol (1.3%) might beanot her barrier for higher biodiesel production yield; however,the best production yield (85.0%) is still in an acceptable rangecompared to the previous study on the transesterication ofraw castor oil using ethanol (Dias et al., 2013).Further characterizations were performed o n the optimumbiodiesel, and the results are presented in Table 6. Accordingly,all the measured properties met the ASTM D6571 standardexcept the viscosity, which was three times greater than themaximum standard amount.3.4.MassbalanceandeconomicestimationsUnder optimum conditions, 149.61.4g biodiesel/kg cas-tor plant was produced, consuming around 51.0 g ethanol. Table5TheRSMtandparameterestimations.TermCoefcientStandard errorT -valueP -valueConstant338.14628.453811.8840.000t54.1672.033826.6340.000T9.8370.99539.8840.000R147.29057.48682.5620.028t25.4860.204426.8380.000T20.0720.00828.8610.000R2189.54581.77152.3180.043tT0.2180.02409.0990.000tR8.9372.39713.7280.004TR0.0880.47940.1830.859 Please cite this article in press as: Bateni, H., Karimi, K., Biodiesel production from castor plant integrating ethanol production via a bioreneryapproach. Chem. Eng. Res. Des. (2015), http://dx.doi.org/10.1016/j.cherd.2015.08.014 ARTICLE IN PRESSCHERD-1994;No.of Pages9chemicalengineeringresearchanddesignxxx(2015)xxxxxx7 Fig.4 Contourplotsofbiodieselyieldasafunctionof:(a)temperatureandtimeforthereactionusing0.3:1ethanoltooilmassratio,(b)temperatureandethanoltooilmassratiofor3hreaction,and(c)ofethanoltooilmassratioandtimeforthereactionat 60 C(allinthepresenceof1%KOHascatalyst). Table6Characterizationofbiodieselproducedunderoptimumconditions.PropertiesBiodiesel sampleASTM D6751 Flash point ( C)1563130(min)Kinematic viscosity (mm /s)218.10.31.96.0Free glycerin (%)0.0100.010.020Water and sediment (%)0.020.020.050Heating value (MJ/kg)33.42.6 Therefore, 30.1 g ethanol/kg cas tor plant is produced as avalue-added byproduct of the process.Santana et al. (2010) estimated the castor biodiesel cost fora plant with a capacity of 1000kg/h castor oil, where 591.3kg/hethanol was used. They estimated the nal cost of biodiesel atabout $1.52 per kg biodiesel, considering the price of castor oiland ethanol for $1157/ton and $354.21/m , respectively. How-3ever, according to the current optimum conditions, the bestbiodiesel yield was obtained usi ng 290 kg/h ethanol, w hich cansignicantly affect the nal biodiesel price.On the other hand, Santana et al. (2010) used the price ofindustrial ethanol (rst generation) for the economic analysis,while the second generation ethanol (from lignocelluloses) isusually more expensive than the rst generation one (fromsugar or starch based substrates) due to pretreatment cost(Asgher et al., 2014). A rough estimation of ethanol costthrough the current proce s s was around $0.35/ L consideringa general modication in pretreatment cost in the Shaeiet al. (2011) study. Shaei et al. (2013) used NMMO pretreat-ment on spruce wood to get around the same ethanol yield asthe current study. Techno-economi c analysis exhibited thatthe NMMO cost was more than 30% of nal ethanol value($0.47/L) (Shaei etal.,201 1).Sodium hydroxideis considerablycheaper than NMMO (less than one-tenth the cost) (TecnonOrbiChem, 2013); therefore, alkaline pretreatment using 8%sodium hydroxide solution loads a signicantly lower chargeon the nal ethanol cost with respect to NMMO pretreatment.Accordingly, the biodiesel production using current technol-ogy roughly cost around $1.47/kg.4.ConclusionThis study showed that the castor plant can be comprehen-sively exploited using a biorenery approach for biodiesel andethanol production in which no extra alcohol out of the pro-cess was necessary for the biodiesel unit. However, an alkalinepretreatment at high temperature was needed to open upthe lignocellulosic structure of the castor plant residue, andsubsequently improve the ethanol production. The integratedcastor plant processing for biodiesel and ethanol productionhas both economic and environmental benets. However, aprecise investigation is needed for waste treatment and recov-ery units of the biorenery plant.The empirical model developed for biodiesel productionwas in a good agreement with the experimental results. How-ever, a wide range of experimental data can be used to evaluateand readjust the parameters to get a better model.A rough estimate based on optimum results of the currentstudy showed that integrated ethanol and biodiesel produc-tion through the proposed technology cost around $0.35/Land $1.47/kg, respectively. However, further detailed techno-economical investigation is necessary to determine a moreaccurate estimation about the feasibility of the technology andeconomic details.ReferencesAarthy, M., S aravana n, P. , G o wthaman, M . K., Rose , C., K amini ,N.R., 201 4. En z y m atic tr anse sterication for production ofbiodiesel using yeast lipases: an overview. Chem. Eng. Res.Des. 92, 15911 601.Adedeji, J.A., A p ata, D . F ., A derin o la, O.A., Ra u, T . A., Amao , S . R.,2006.Per fo r m a nc e and haematological/serum characteristicsof rabbits fed boiled castor seed cake based diet. Wo rld J. Zool .1, 9193.Antizar-Ladi slao , B. , T u r rio n -Go mez, J.L., 2008. Second-generationbiofuels and local bioenergy systems. Biofuel B ioprod.Bioren. 2, 455469.Asgher, M., Bashir , F. , Iq b al , H.M.N., 2014. A compr e hensiv eligninoly tic p re-t r eatm ent approach from lignocellulose greenbiotechnology to pr oduce bio-eth a nol. Ch e m. Eng. Re s. D es.92 , 1571 1578.ASTM, 2003a . Standard Test Me thod D240, Heat of Combustion ofLiquid Hydrocarbon Fuels by Bom b Calori m eter . W estConshoh ocken, PA .ASTM, 2003b. Sta n dard Test Method D6751, Specication forBiodiesel Fuel Blend Stock (B100) for Middle Distillate Fuels.West Con shohocken, PA .Bateni, H., K arimi, K., Zamani, A., B enakash ani, F. , 20 1 4. Cas t orplant for bi o die sel, biogas, and ethanol production with abiorenery processing perspective. Appl. Ener gy 136 ,1422.Berman, P., Nizri , S. , W iesman, Z., 2011. Castor oil biodiesel andits blends as alternative fuel. Biomass Bioenergy 35,28612866.Cavalcante, K.S., Penha, M.N ., Mendonca , K.K., L o u zeiro , H. C .,Vasconce los, A ., Maciel, A.P., de So uz a, A.G., S ilva, F . C., 2010 .Optimiza tion o f tra nsesterication of castor oil with ethanolusing a central composite rotatable design (CCRD). Fuel 89,11721176. Please cite this article in press as: Bateni, H., Karimi, K., Biodiesel production from castor plant integrating ethanol production via a bioreneryapproach. Chem. Eng. Res. Des. (2015), http://dx.doi.org/10.1016/j.cherd.2015.08.014 ARTICLE IN PRESSCHERD-1994;No.of Pages98chemicalengineeringresearchanddesignxxx(2015)xxxxxx Chakrabarti, M.H., Ahm a d, R ., 2 008 . Tr a nses t eric ation studies oncastor oil as a rst step towards its use in biodieselproductio n. Pak. J. Bot. 40 , 115311 5 7.Da Silva, N.L ., Maciel, M.R.W., Bat i stella, C .B ., F ilho, R.M., 2006.Optimization of biodiesel production from castor oil. Appl.Biochem. Biotechnol. 130, 405414.De Oliveira, D.L., Faccio , M.D ., Rosa, C ., Bender , C.D., Lip k e , J.P .,Amrogin ski, N. , D a riv a , C., De Oliveira, C.J.V ., 2005 .Optimiza tion of al kali ne transesterication of soybean oil andcastor oil for biodiesel production. Appl. Biochem. Biotechnol.122, 553 560.Dias, J., Arajo, J., Costa, J., Alvim-Ferraz, M., Almeida, M., 2013.Biodiesel production from raw castor oil. Energy 53, 5866.Du, C., Cam pbe l l, G. , Misa il id i s, N. , Mateos- S alvador , F. ,Sadhukh an, J. , Mus t a f a, M., Weightman , R., 2009. E v aluatingthe feasi b i lit y o f commercial a rabino x ylan production i n t hecontext o f a whea t biorene r y pr i ncipally p ro d ucin g etha nol,Part 1. Experimental studies of arabinoxylan extraction fromwheat br an . Chem. E ng. R e s. De s. 87, 123 2 1 2 38.Goshadrou, A., K ar i m i, K., T a he r zad e h, M.J ., 2011. Bioethanolproduction from sweet sorghum bagasse by Mucor hiemalis .Ind. Crop. Prod. 34, 121 9 1 225 .Haigh, K.F., Vladisavlje vi c, G .T ., Reynold s, J. C., Nagy , Z ., Sah a, B. ,2014.Kin eti c s o f the pre-treatment of used cooking oil usingNovozyme 435 for biodiesel production. Chem. Eng. Res. Des.92, 713719.Islam, M.S., Hamdan, S. , J u s oh, I., Rahman , M.R., Ahmed, A .S.,2012.The effect of al k ali pr e treatme nt o n me chanical andmorphological properties of tropical wood polymercomposit es . Mater . D es. 3 3 , 419 4 24 .Jeong, G.-T., Park, D.-H., 2 0 09. O p timizatio n of biodies elproduction from castor oil using response surfacemethodo logy . Appl. Biochem. Biotechnol. 156, 111.Karimi, K., Chisti, Y., 2015. Future of b ioetha m ol. B iofuel Res. J. 5,147.Karimi, K., Pandey, A., 2014. Current and future ABE p r oc e ss e s.Biofuel R es. J. 3, 77 .Karimi, K., S ha e i, M., Kumar , R., 201 3. P ro gress on P h ysical andChemical Pretreatm e nt of L ignocellulosic Biomass in Biofuelsand Bioreneries: Recent Developments. Springer Berl inHeidelber g, Berlin .Klc, M., Uz un, B . B. , Pt n , E. , Pt n, A . E., 2 013. Optimization ofbiodiesel production from castor oil u s ing factor i al desig n.Fuel Proc ess. Tec h nol. 111 , 105110 .Lam, M.K., L ee , K.T., 2011. Mixed me thanolet hanol t echnolo g y toproduce gre e ner b iod i esel f rom waste cooking oil: abreakthrough for SO42/SnO SiO22catalyst. Fuel P r ocess.Technol. 92 , 16391645.Leung, D.Y.C ., Wu , X ., Leun g, M.K.H., 2010. A review on biodieselproduction using catalyzed transesteri cation. Ap p l. Energy87, 1083 1095.Luo, L., Voet, E., Hup p es, G. , 2010. Bior e ning of lign o cellu l osicfeedstock technical, economic and environmentalconsiderations. Bioresour. Tec hnol. 1 0 1, 5023503 2 .Martinez-He rnandez, E., Marti n ez - Herrera, J. , Ca m pbell, G.M .,Sadhukh an, J. , 20 1 4. P r oce ss integra t ion, e n erg y and GHGemission analyses of Jatropha-based biorenery systems.Biomass Convers. Bioren. 4, 105 124.Mirahmadi, K., K a bi r i, M. M ., J eihaniour , A ., Kari m i, K .,Taherzad eh, M.J., 201 0 . A lka l ine pretreatment of spruce andbirch to improve bioethanol and biogas production.BioResource 5, 928938.Misailidis, N., Campbe l l, G. , Du, C. , Sadhukhan, J. , Mustafa, M.,Mateos-S alvador , F. , Weightman , R., 2 0 09. Evaluati n g thefeasibility of c ommerci al arabinox y lan p roductio n in thecontext o f a w hea t bi o ren e ry pr incipally producing ethanol:Part 2, Process simulation and economic analysis. Chem. Eng.Res. Des. 87, 12 391250 .Mohsenzade h, A . , J e i hanipour , A., Karimi, K., T aherzadeh, M.J.,2011.Alk ali pr etreatme nt of softwo o d spruc e and ha r d w oodbirch by NaOH/ t hiourea, NaOH/urea , NaOH/u r ea/th iourea,and NaOH/PEG to improve ethanol and biogas production. J.Chem. Technol. Biotechnol. 87, 12091214.Ng, K.S., Sad hukhan, J. , 2011. Pr o ces s integr a tion an d economicanalysis of bi o -oil p latform for the production of methanoland combined heat and power. Biomass Bioenerg y 35,1153116 9.Nieves, D.C., Kar i mi, K., Hor v t h , I.S ., 2011 . Improvement ofbiogas production from oil palm e mpt y frui t bunches (O PEFB).Ind. Crop. Pr o d. 34, 10971101.Ogu nniyi, D.S., 2006. Castor oil: a vital industrial raw material.Bioresour. Technol. 97, 10861091.Perdomo, F.A., Acosta-O sorio , A.A., Herr e ra, G ., Vasco-Lea l, J. F .,Mosquer a-Artam o nov , J. D., Millan- Ma lo , B. , R o dr i guez-Gar c ia,M.E., 201 3. Physico c hemi c al characterization of sevenMexican Ricinus communis L. seeds & oil contents. BiomassBioenergy 48 , 1724 .Per domo, F.A ., Pe r domo , L., M illn, B . M ., Ar a gn, J .L., 2014. Designand impr ove m en t of bi odies el fuels blends by optimization oftheir molecular structures and compositions. Chem. Eng. Res.Des. 92, 1 4821492.Pinto, M.S., 2 011. G lobal Biofuels Outlook 20102020, Available at:http://www.unece.lsu.edu/biofuels/documents/2013Mar/bf13 04.pdf (accessed A pril 2015) .Ra machandran, S., Singh, S.K., Larroche, C., Soccol, C.R., Pan dey ,A.,2007.Bioresour . T e chnol. 9 8 , 200 0 2 009.Ramanjaney ul u , A., Redd y, A . V., Madh a vi, A., 2013. T he i mpac t ofsowing d ate and irrig a tion re g ime o n c astor ( Ri c in u s communisL.)seed y ie l d, o il qu a lit y c h ar acteristics and fatty acidcomposition during post rainy season in South India. In d.Crop. Pro d. 44, 2531.Ramezani, K ., Rowshanza m ir , S. , E i ka n i, M. , 2010 . Castor oiltransesterication reaction: a kinetic study and op timizat i onof param ete r s. Energy 35, 4142 4 148.Reyero, I., Ar zamendi, G ., Ganda, L.M., 2014. Heterogeni z ation o fthe biodi esel s yn t hesi s cata lysis: CaO and novel calciumcompounds as transesterication catalys ts. Chem. Eng. Res.Des. 92, 1 519 1530.Sa dasivam, S., Manickam, A., 2005. Biochemical Methods. NewAge Inter national Pub l ish e rs, New Delhi.Sadhukhan, J. , Ng, K.S., M a rtinez-H e rnand e z, E. , 2014.Bioreneries and Chemical Processes: Design, Integration andSustaina bility Analysis . W il ey , Chic h ester .Salehi, S.M.A ., Karimi, K., Behzad, T. , Poorne j ad, N., 2 012. E fcie ntconversion of rice straw to bioetha nol us i ng sodiumcarbonat e pret re atment. E ne r gy Fuel 26, 73547361.Salehian, P., Karimi, K., 2 013. Al k ali pr etr e atm e nt forimprovement of biogas and ethanol production from differentwaste pa rt s of pine tree . Ind . Eng. Chem. Re s. 52, 972978 .S nchez, B., Benitez, B . , Quer i ni, C ., Me n d o w , G., 2015.Tra n seste ricatio n of sun owe r o il w ith ethanol using sodiumethoxide as catalyst. Effect of the reaction conditions. FuelProcess. Technol. 131, 2935.Santana, G., Martins, P., Da Silva, N.d.L. , Batistella , C., M a cielFilho, R., Ma ci el, M.W., 2010. Simulation and cost estimate forbiodiesel production using castor oi l. Chem. E ng. Res. D es . 88 ,626632.Scholz, V., da Silva, J.N., 2008. Prospects and risks of the use ofcastor oil as a f uel. Biomass B i oen e r gy 32, 9 5100.Shaei, M., K arimi, K., Taherzade h, M. J., 201 1 . Techno-eco n om i calstudy of ethanol a n d biogas f rom spru c e wo od byNMMO-pretreatment and rapid fermentation and digestion.Bioresour. Tech nol. 102, 7879 78 86.Shaei, M., Z iloue i , H., Za m ani, A ., Taherzadeh , M.J., Karimi, K.,2013.Enhancement of ethanol production from sp ruce woodchips by io nic liqu i d pretr e atment. Appl . Energy 102,1 16 3 169.Sil va, G.F., C amargo , F.L ., F e rreira, A.L., 2011. A pplication o fresponse surface met h odology for opt i miza tion of biodieselproduction by transesterication of soybean oil with ethanol.Fuel Proc ess. Technol. 9 2 , 4 07 4 13.Sl uiter, A., R uiz, R., Scarlata, C. , Sluite r, J. , Templ e ton, D. , 200 5.Determin ation of Extractives in Biomass. National RenewableEnergy Laboratory (NREL) Analytical Procedures,NREL/TP-510-42619.Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton,D., Crocker, D., 2008. Determination of Structural Please cite this article in press as: Bateni, H., Karimi, K., Biodiesel production from castor plant integrating ethanol production via a bioreneryapproach. Chem. Eng. Res. Des. (2015), http://dx.doi.org/10.1016/j.cherd.2015.08.014 ARTICLE IN PRESSCHERD-1994;No.of Pages9chemicalengineeringresearchanddesignxxx(2015)xxxxxx9Carbohyd rates and Lignin in Biomass. National RenewableEnergy Laboratory (NREL) Analytical Procedures,NREL/TP- 510-42618.Sukkasi, S., Chollaco o p , N. , Ellis, W . , Grimley , S. , Jai -In, S. , 201 0.Challeng es and c onsi d era t ions fo r plannin g tow a rdsustainab le b i odies el de v elopment in developing countries:lessons from the Greater Mekong Subregion. Renew. Sustain.Energy Rev. 14, 31003107.Tecnon Orbi Chem, 2013. Caustic Soda, Chem-Net Fact ChemicalMarket Insight and Foresight-on a Single Page, Available at:www.orbichem.com/userles/CNF%20Samples/cas 13 11.pdf (accessed April 2015).Teghammar, A., Y ngvesson, J., Lundin, M ., Taherza de h, M.J.,Horvth, I.S., 2 010. Pretreatment of paper tube residuals forimproved biogas production. Bioresour. Technol. 101,12061212.The U.S. Department of Energy, 2015. Clean Cities AlternativeFuel Price Report, Available at: http://www.afdc.energy.gov/uploads/publication/alternative fuel price report jan 2015.pdf (accessed April 2015).Tsoutsos, T., Chatzak i s, M., S a ran t o p oulos , I., N ikol o giann i s, A. ,Pasadaki s, N . , 201 3. Effect of w a st e w a ter i rrig ation onbiodiesel quality and productivity fr om castor and s u n o weroil seeds. Ren e w . Ener g y 5 7, 2112 1 5.Wan, C., Zho u, Y. , Li, Y. , 2 011. Liqui d ho t wat e r a n d alk a linepretreatment of soybean straw for improving c ell u losedigestibil ity . Biore s our . Techn o l. 10 2, 62546259.Zhu, J.Y., Zh uan g , X.S., 201 2. Con c eptual net energ y ou t pu t forbiodiesel pr od uct ion from lignocellulosic biomass throughbiorening. Progress Energy Combust. Sci. 38,583598.Zhu, Q.L., Sh ao , R., D o ng, R., Yun, Z., 2 0 14 . A n integratedapproach for obta ining biodiesel, sterols, gossypol, andrafnose from cottonseed on a biorenery concept. Energy 70,149158.