Top Banner
Biochemical Engineering Journal 81 (2013) 15–23 Contents lists available at ScienceDirect Biochemical Engineering Journal jou rnal h om epage: www.elsevier.com/locate/bej Regular article Biodiesel production from soybean soapstock acid oil by hydrolysis in subcritical water followed by lipase-catalyzed esterification using a fermented solid in a packed-bed reactor Diniara Soares a , Andrei Ferreira Pinto b , Alan Guilherme Gonc ¸ alves c , David Alexander Mitchell a , Nadia Krieger b,a Departamento de Bioquímica e Biologia Molecular, Universidade Federal do Paraná, Cx. P. 19046 Centro Politécnico, Curitiba 81531-980, Paraná, Brazil b Departamento de Química, Universidade Federal do Paraná, Cx. P. 19081 Centro Politécnico, Curitiba 81531-980, Paraná, Brazil c Departamento de Farmácia, Universidade Federal do Paraná, Av. Lothario Meissner, 3400, Jardim Botânico, Curitiba, Paraná, Brazil a r t i c l e i n f o Article history: Received 19 February 2013 Received in revised form 30 June 2013 Accepted 26 September 2013 Available online 5 October 2013 Keywords: Biodiesel Hydroesterification Lipases Burkholderia cepacia Solid-state fermentation Packed-bed reactor a b s t r a c t We investigated a new hydroesterification strategy for the production of biodiesel from low-value oil feedstocks: complete hydrolysis of the feedstock to fatty acids in subcritical water, followed by the use of a packed-bed reactor, containing a fermented solid with lipase activity, to convert the fatty acids to their ethyl esters. The fermented solids were produced by cultivating Burkholderia cepacia LTEB11 for 72 h on a 1:1 mixture, by mass, of sugarcane bagasse and sunflower seed meal. The esterification of fatty acids obtained from soybean soapstock acid oil was studied in the packed-bed bioreactor, in a solvent-free sys- tem, with the best results being a 92% conversion in 31 h, obtained at 50 C. When the packed-bed reactor was reused in successive 48-h esterification reactions, conversions of over 84% of the fatty acids to esters were maintained for five cycles at 50 C and for six cycles at 45 C. Unlike previous hydroesterification processes that have used lipase-catalyzed hydrolysis followed by chemically-catalyzed esterification, our process does not expose the lipases to contaminants present in low quality feedstocks such as soapstocks. This advantage opens the possibility of operating the packed-bed esterification reactor in continuous mode. © 2013 Published by Elsevier B.V. 1. Introduction Biodiesel is currently being produced as a substitute for petrodiesel, however, it is not economically competitive, with its use requiring either subsidies or government policies, such as in Brazil, where all petrodiesel is currently required to contain 5% biodiesel [1]. It is composed of esters of short-chain alcohols (methyl or ethyl alcohols) and long-chain fatty acids. The feed- stocks used in most commercial biodiesel production processes are derived from triacylglycerols of edible vegetable oils [2]. How- ever, these oils are relatively expensive, and since the feedstock in biodiesel synthesis corresponds to 50–85% of total production costs, it is desirable to use low-cost starting materials, in order to increase the commercial competitiveness of biodiesel [3–5]. Potential low-value feedstocks include animal fat from sewage and residual oil of domestic, industrial or commercial origin. Current industrial processes for the production of biodiesel from vegetable oils use alkaline transesterification, which gives high Corresponding author. Tel.: +55 41 33613470; fax: +55 41 33613006. E-mail address: [email protected] (N. Krieger). yields (98%) in a short reaction time of about 1 h. However, alka- line transesterification requires starting materials with low levels of moisture and free fatty acids and therefore is not appropriate for the production of biodiesel from low-value feedstocks, which con- tain significant amounts of free fatty acids or water. Free fatty acids react with the alkaline catalyst, producing soaps. This decreases the reaction yield and makes the separation of products difficult. Too much water favors hydrolysis over transesterification. Addition- ally, the alkaline catalyst contaminates both the glycerol and the biodiesel that are produced. Its removal from the biodiesel requires several washings with water. This not only generates large amounts of wastewater, but also gives rise to residual water in final product [6,7]. Several strategies can be used to overcome problems caused by the presence of fatty acids in low-value feedstocks. It is possible to carry out the process using catalysis with an inorganic acid [8–10], a heterogeneous catalyst [11,12] or lipases [4,13–16], since all can simultaneously catalyze esterification and transesterification. It is also possible to use a two-step process. In the first step, the free fatty acids are converted to esters using an acid catalyst (such as sulfuric acid), with the remaining acylglycerols in the feedstock then being converted to biodiesel esters by alkaline transesterification [17,18]. 1369-703X/$ see front matter © 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.bej.2013.09.017
9

Biodiesel Production From Soybean

Sep 05, 2015

Download

Documents

Hoshi

Biodiesel Production From Soybean
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
  • Biochemical Engineering Journal 81 (2013) 15 23

    Contents lists available at ScienceDirect

    Biochemical Engineering Journal

    jou rna l h om epage: www.elsev ier .c

    Regular article

    Biodies acisubcrit esfermen

    Diniara S nc aDavid Ala Departament tro Pob Departament 8153c Departament tnico

    a r t i c l

    Article history:Received 19 FeReceived in reAccepted 26 September 2013Available online 5 October 2013

    Keywords:BiodieselHydroestericationLipasesBurkholderia cSolid-state ferPacked-bed re

    cation feedment

    ethyl esters. The fermented solids were produced by cultivating Burkholderia cepacia LTEB11 for 72 h ona 1:1 mixture, by mass, of sugarcane bagasse and sunower seed meal. The esterication of fatty acidsobtained from soybean soapstock acid oil was studied in the packed-bed bioreactor, in a solvent-free sys-tem, with the best results being a 92% conversion in 31 h, obtained at 50 C. When the packed-bed reactorwas reused in successive 48-h esterication reactions, conversions of over 84% of the fatty acids to esters

    1. Introdu

    Biodiesepetrodieseluse requiriin Brazil, w5% biodiese(methyl or stocks usedare derivedever, these in biodieselcosts, it is to increasePotential loresidual oil

    Current vegetable o

    CorresponE-mail add

    1369-703X/$ http://dx.doi.oepaciamentationactor

    were maintained for ve cycles at 50 C and for six cycles at 45 C. Unlike previous hydroestericationprocesses that have used lipase-catalyzed hydrolysis followed by chemically-catalyzed esterication, ourprocess does not expose the lipases to contaminants present in low quality feedstocks such as soapstocks.This advantage opens the possibility of operating the packed-bed esterication reactor in continuousmode.

    2013 Published by Elsevier B.V.

    ction

    l is currently being produced as a substitute for, however, it is not economically competitive, with itsng either subsidies or government policies, such ashere all petrodiesel is currently required to containl [1]. It is composed of esters of short-chain alcoholsethyl alcohols) and long-chain fatty acids. The feed-

    in most commercial biodiesel production processes from triacylglycerols of edible vegetable oils [2]. How-oils are relatively expensive, and since the feedstock

    synthesis corresponds to 5085% of total productiondesirable to use low-cost starting materials, in order

    the commercial competitiveness of biodiesel [35].w-value feedstocks include animal fat from sewage and

    of domestic, industrial or commercial origin.industrial processes for the production of biodiesel fromils use alkaline transesterication, which gives high

    ding author. Tel.: +55 41 33613470; fax: +55 41 33613006.ress: [email protected] (N. Krieger).

    yields (98%) in a short reaction time of about 1 h. However, alka-line transesterication requires starting materials with low levelsof moisture and free fatty acids and therefore is not appropriate forthe production of biodiesel from low-value feedstocks, which con-tain signicant amounts of free fatty acids or water. Free fatty acidsreact with the alkaline catalyst, producing soaps. This decreases thereaction yield and makes the separation of products difcult. Toomuch water favors hydrolysis over transesterication. Addition-ally, the alkaline catalyst contaminates both the glycerol and thebiodiesel that are produced. Its removal from the biodiesel requiresseveral washings with water. This not only generates large amountsof wastewater, but also gives rise to residual water in nal product[6,7].

    Several strategies can be used to overcome problems caused bythe presence of fatty acids in low-value feedstocks. It is possible tocarry out the process using catalysis with an inorganic acid [810],a heterogeneous catalyst [11,12] or lipases [4,1316], since all cansimultaneously catalyze esterication and transesterication. It isalso possible to use a two-step process. In the rst step, the free fattyacids are converted to esters using an acid catalyst (such as sulfuricacid), with the remaining acylglycerols in the feedstock then beingconverted to biodiesel esters by alkaline transesterication [17,18].

    see front matter 2013 Published by Elsevier B.V.rg/10.1016/j.bej.2013.09.017el production from soybean soapstock ical water followed by lipase-catalyzedted solid in a packed-bed reactor

    oaresa, Andrei Ferreira Pintob, Alan Guilherme Goexander Mitchell a, Nadia Kriegerb,

    o de Bioqumica e Biologia Molecular, Universidade Federal do Paran, Cx. P. 19046 Ceno de Qumica, Universidade Federal do Paran, Cx. P. 19081 Centro Politcnico, Curitibao de Farmcia, Universidade Federal do Paran, Av. Lothario Meissner, 3400, Jardim Bo

    e i n f o

    bruary 2013vised form 30 June 2013

    a b s t r a c t

    We investigated a new hydroesterifeedstocks: complete hydrolysis of thea packed-bed reactor, containing a ferom/ locate /be j

    d oil by hydrolysis interication using a

    lvesc,

    litcnico, Curitiba 81531-980, Paran, Brazil1-980, Paran, Brazil, Curitiba, Paran, Brazil

    strategy for the production of biodiesel from low-value oilstock to fatty acids in subcritical water, followed by the use ofed solid with lipase activity, to convert the fatty acids to their

  • 16 D. Soares et al. / Biochemical Engineering Journal 81 (2013) 15 23

    All of these processes transesterify the acylglycerols and esterifythe free fatty acid, while avoiding the formation of soaps. However,acid catalysts are highly corrosive to equipment, while both hetero-geneous catalysts and lipases are expensive and transestericationrates are rewater in thepromoting

    Other stried out anwhich convglycerols into produceacid-catalyzied the soand an acierols. This aestericatioprocesses inbe separatewastewater

    Recentlyoils has beeIn the rst sacids and glthe rst stepester and wduction whis a hydrolytent of the fthe aqueouthan that ocess [21,22]

    Althoughydroesteriauthors haplant lipasework takes sis in subcriuse lipases,estericatiobecause it iin Brazil. Inin the formfermentatio

    2. Materia

    2.1. Raw m

    Soybeancooking oilSoares e Ciawere of ana

    2.2. Hydrol

    Hydrolyin the pilot Ltda. (Pontaous hydrolyin a pressurmaterial anof catalyst. top of the tothe bottom

    were distilled and their compositions (Table 1) were determinedby gas chromatography [25]. All free fatty acid preparations con-tained less than 0.5% (w/w) water. Saponication and acid valueswere determined according to the methods of the American Oil

    sts S].

    icroo

    epacim w

    agarransfnd th0 h, as u

    lid-s

    fermf a m/w oool Mas per s

    to os don

    dry sded t wao Pved

    ated mented sted

    lessmenns.

    rying

    ee d ferm

    in ash feted

    ter). and imatn. In tced lC. Duined

    y of tsh fete aneactinterf

    fermeriv

    s witashi

    m anwereventlatively slow [7]. Also, if there is a signicant amount of reaction medium then it will compete with the alcohol,the hydrolysis of the acylglycerols.rategies have also been tried. Haas et al. [8] rst car-

    alkali-catalyzed saponication of soybean soapstock,erts both free fatty acids and the fatty acids of acyl-to soaps. The soaps were recovered and then acidied

    free fatty acids, which were then esteried in aned process. On the other hand, Wang et al. [19] acid-apstock, causing it to separate into an aqueous phased oil phase containing free fatty acids and acylglyc-cid oil phase was then subjected to an acid-catalyzedn/transesterication process. However, both of thesevolve homogeneous catalysis with acids, which mustd from the product by successive washes, generating

    contaminated with catalyst and reaction products., the production of biodiesel by hydroesterication ofn proposed [2023]. The process involves two steps.tep, triacylglycerols are hydrolyzed completely to fattyycerol. In the second step, the fatty acids recovered from

    are esteried with an alcohol to give the correspondingater. This process is advantageous for biodiesel pro-en low-value feedstocks are used. Since the rst stepsis step, the water content and the free fatty acid con-eedstock do not interfere with nal yields. Additionally,s glycerol produced in the hydrolysis step is more purebtained in the alkali-catalyzed transesterication pro-.h both the hydrolysis and esterication steps in acation process can be carried out chemically, severalve investigated the potential of using commercial ors to catalyze the hydrolysis step [2123]. The presenta different approach. We produce fatty acids by hydroly-tical water of several low-value feedstocks and we then

    produced by Burkholderia cepacia LTEB11, to catalyzen with ethanol. Ethanol was selected as the alcohols less toxic than methanol and is abundantly available

    order to minimize the costs of the lipases, we use them of dried fermented solids, obtained by solid-staten.

    ls and methods

    aterials

    oil, soybean soapstock acid oil, beef tallow and waste were donated by the company Ubaldino Rodrigues. Ltda. (Ponta Grossa, Paran, Brazil). All other reagentslytical grade.

    ysis of fat feedstocks

    sis of the different feedstocks (Table 1) was carried outplant of the company Ubaldino Rodrigues Soares e Cia.

    Grossa, Paran, Brazil). The process involves continu-sis of the feedstocks in the presence of subcritical water,e tower at 60 atm and 250 C [24]. In this process, fattyd water react in a countercurrent ow in the absenceThe free fatty acids produced are discharged from thewer and the water/glycerol mixture is discharged from

    of the tower. The free fatty acids from each feedstock

    Chemi[26,27

    2.3. M

    B. cmediuto a LBwere task afor 81broth w

    2.4. So

    The(SSF) o(1:1, wde lcseed wsunowsievingSSF wamilledwas adconten2000, Sautoclainoculthe ferfermenfermentent ofdry ferreactio

    2.5. D

    Thrfrozen40 Cthe freintegradiamethe airapproxcolumfan-forat 30

    determactivitthe fretriplica

    In ravoid iity, thelipids dwasheeach w200 rpsolids the solociety (AOCS Ofcial Method Cd 3-25 and Ca 5a-40)

    rganism

    a LTEB11 was maintained at 18 C in Luria Bertani (LB)ith 50% (w/v) glycerol. A stock culture was transferred

    plate and incubated for 48 h at 29 C. Isolated colonieserred to 30 mL of LB medium in a 250-mL Erlenmeyeren incubated on a rotary shaker at 29 C and 200 rpmwhich represents mid-exponential phase. This culturesed as inoculum for the solid-state fermentation.

    tate fermentation

    ented solid was obtained by solid-state fermentationixture of sugarcane bagasse and sunower seed mealn a dry basis). Sugarcane bagasse was donated by Usinaelhoramentos (Jussara, Paran, Brazil) and sunowerurchased in the local market. Sugarcane bagasse andeed were milled, separately, in a knife mill, followed bybtain particles ranging between 0.85 and 2.36 mm. Thee in 2000-mL Erlenmeyer asks, each containing 80 g ofubstrate. Phosphate buffer solution (0.1 mol L1, pH 7.0)to obtain 75% moisture (w/w, wet basis). The moistures determined in an infrared moisture balance (Gehaka IVaulo, Brazil). Flasks were plugged with cotton wool and

    at 121 C for 20 min. After cooling, the substrates werewith 8 mL of inoculum and incubated at 29 C. Duringtation, the hydrolytic and esterication activities of theolids were determined every 24 h. After incubation, thesolids were dried (see Section 2.5) to a moisture con-

    than 10% (w/w on a wet basis) and stored at 4 C. Theted solids were then used directly in all esterication

    and preparation of the fermented solid

    ifferent drying processes were evaluated. In the rst,ented solids were lyophilized for 24 h at 101 mbar and

    lyophilizer (Jouan LP3, Virginia, USA). In the second,rmented solids were dried in a column made with twopolyvinyl chloride tubes (each 50 cm height and 4.3 cmThe lower tube was lled with activated silica to drythe top contained 200 g of fresh fermented solids. Air ately 25 C was blown at 20 L min1 into the bottom of thehe third, 200 g of fresh fermented solids was placed in aaboratory oven (Nova tica 400-3ND, So Paulo, Brazil)ring the drying, the moisture content of the solids was

    with the infrared moisture balance, and the hydrolytiche dried fermented solids was compared with that ofrmented solids. Values are expressed as the means ofalyses the standard error of the mean.

    ons where n-hexane was used as the solvent, in order toerences in the determination of the esterication activ-ented solids were delipidated after drying to remove

    ing from the fermentation. Delipidation involved threeh n-hexane (10 mL per gram of fermented solids). Inng, the mixture was agitated vigorously for 10 min atd 25 C. The solution was then ltered and the retained

    dried in a vacuum desiccator at room temperature. In-free reaction system, the delipidation was unnecessary

  • D. Soares et al. / Biochemical Engineering Journal 81 (2013) 15 23 17

    Table 1Characterization of free fatty acids from hydrolysis in subcritical water.

    Composition (% of free fatty acids)

    FA-SO FA-SSAO FA-WCO FA-BT

    Myristic Palmitic 16.5Palmitoleic Stearic 4.2Oleic 33.4Linoleic 44.2Others 1.7Average mol 277.3Acid value (m 195.5Saponicatio 199

    Fatty acids fro -WCO

    because thecompared t

    2.6. Activity

    The hydassessed by(67 mmol L

    (2.5 mmol Lwas emulsithen for an a20 mL of emglass vesselthe reactionpotentiomepH being mNaOH solutthe releaseconditions.

    For estescrew-capp70 mmol L

    alcohol mofermented sat 200 rpm ture were rthe Lowrylated by conOne unit (Uproduced p

    Both hyunits of acti

    2.7. Stabilit

    The stabmeasuring different sythe followinoleic acid aacid and ethture). For thfermented srotary shaksolids wereto remove tthen ltereture. The remethod usithe initial h

    elimi

    liminin thce an

    oute (olamot xdualoleatcids f

    lvent

    pacl diad walumn

    with presoun

    e bull diarom r mal (coMHP

    of tixturshedl. Throus nto e

    by tiis.

    the s-bedC14:0 C16:0 7.9C16:1 0.6 C18:0 4.8 C18:1 30.1 C18:2 52.6

    4.0 ecular weight (g mol1) 278.6g KOH g1) 191.3 n value (mg KOH g1) 196

    m soybean oil (FA-SO), soybean soapstock acid oil (FA-SSAO), waste cooking oil (FA

    lipid content in the fermented solids was negligibleo the lipid content of the reaction mixture.

    measurement with the fermented solid

    rolytic activity of the dry fermented solids was the titrimetric method. The solution contained triolein1), gum arabic 3% (w/v), CaCl2 (2 mmol L1), TrisHCl

    1) and NaCl (150 mmol L1) in distilled water [28]. Ited with a blender at high speed, initially for 5 min anddditional 1 min immediately before use. For each assay,ulsion and 100 mg of fermented solids were placed in a

    maintained at 40 C. The free fatty acids released during were titrated for 5 min in a Metrohm 718 STAT Titrinotric titrator (Metrohm, Herisau, Switzerland) with theaintained at 7.0 through the addition of a 0.05 mmol L1

    ion. One unit (U) of hydrolytic activity was dened as of 1 mol of fatty acid per minute, under the assay

    rication activity, reactions were carried out in 12 mLed bottles containing 10 mL of a mixture containing1 of fatty acid, 210 mmol L1 of ethanol (i.e. an acid tolar ratio of 1:3) in n-hexane and 800 mg of delipidatedolids. These bottles were incubated on a rotary shakerand 40 C. At xed intervals, 50 L samples of the mix-emoved and analyzed for residual free fatty acids byTinsley method [29]. Ethyl ester production was calcu-sumption of free fatty acids from the reaction mixture.) of esterication activity equals 1 mol of ethyl esterer minute, under the assay conditions.drolytic and esterication activity were expressed invity per gram of dry fermented solid (U gdfs1).

    y of the fermented solid

    ility of the dry fermented solids was evaluated bythe hydrolytic activity before and after incubation instems. The fermented solids were incubated for 48 h in

    2.8. Pr

    Presolids presencarriedmixturstated 40 C. Afor resi(ethyl fatty a

    2.9. So

    TheinternareceiveThe cosolids,lightlythis amthat thinternaacids fa molaethanoner 45bottomtion mand waethanoanhydacids iminedanalys

    Forpackedg media: (a) ethanol, (b) n-hexane, (c) 70 mmol L1 ofnd 210 mmol L1 of ethanol in n-hexane and (d) oleicanol in a 1:3 molar ratio (i.e. solvent-free reaction mix-ese tests, screw-capped bottles containing 500 mg ofolids and 10 mL of reaction mixture were incubated on aer at 200 rpm and 40 C. After incubation, the fermented

    washed four times, each time with 10 mL of n-hexane,he substrates and reaction products. All samples wered and dried in a vacuum desiccator at room tempera-sidual hydrolytic activity, determined by the titrimetricng triolein as substrate, was expressed as percentage ofydrolytic activity.

    of 48 h. Thesolids and reach cycle, replaced wiexpressed abatch.

    2.10. GC an

    Ethyl estCo., Kyoto, Jtor and a S2.1 5.922.0 30.51.5 5.9

    4.4 10.0 35.6 45.3 32.6 0.8 1.8 1.6

    274.3 269.6 190.9 199.3

    198 203

    ) and beef tallow (FA-BT).

    nary studies of biodiesel synthesis

    ary tests were done to dene the amount of fermentede reaction mixture and to compare the reaction in thed absence of n-hexane as solvent. The reactions were

    in 12 mL screw-capped bottles with 10 mL of reactioneic acid and ethanol in a molar ratio of 1:3) and theunt of fermented solids, with incubation at 200 rpm anded intervals, 50 L samples were collected and analyzed

    free fatty acids by the LowryTinsley method [29]. Estere) production was calculated by consumption of freerom the reaction mixture.

    -free esterication in a packed-bed reactor

    ked-bed reactor was made of a glass column (2.7 cmmeter and 21 cm high) with an external jacket thatter from a water bath at the stated temperature (Fig. 1).

    was packed with 12 g (on a dry basis) of fermented the solids being poured in through a funnel and thensed with a glass rod. Undertaking this procedure witht of fermented solids gave a bed height of 16 cm, suchk density of the bed was 130 g L1. The reservoir (4.4 cmmeter and 9 cm high) was loaded with 100 g of fattysoybean soapstock acid oil (FA-SSAO) (on the basis ofss of 277 g, this corresponds to 361 mmol) and 50 g ofrresponding to 1089 mmol). A peristaltic pump (Sten-10, Florida, USA) pumped the reaction mixture into thehe bed at a ow rate of 5 mL min1. Samples of the reac-e were collected from the top of the packed-bed reactor

    with saturated NaCl solution to separate the excess ofe upper organic fraction was collected and dried overNa2SO4 and centrifuged. The conversion of the fattyster was monitored by measuring the acid value deter-tration [27]. The content of ester was determined by GC

    tudy of the reutilization of the fermented solids in the reactor, the reaction was repeated for eight cycles, each experimental procedure, the amount of the fermentedeaction mixtures used were as described above. Afterthe reaction mixture was collected from the reactor andth a new mixture. Reaction yields after each cycle weres fractions of the yield obtained at the end of the rst

    alysis

    er content was determined using a GC-2010 (Shimadzuapan) equipped with a hydrogen ame ionization detec-GE HT-5 capillary column (0.32 mm internal diameter,

  • 18 D. Soares et al. / Biochemical Engineering Journal 81 (2013) 15 23

    Fig. 1. Schemvoir; (2) reactpacked with feling port; (7) w

    25 m lengthwas dilutedheptadecanwith a splitand detecto120 C for 2tained for 3for 2 min. Pparison of tcontent (in of the inter

    3. Results

    3.1. Produc

    B. cepaciarcane bagaand esteriover time. Tobtained at91.6 3.3 U96 h. Since cation reacremaining s

    3.2. Stabilitfermented s

    After festorage. Prlyophilizati

    8

    10

    100

    120

    gdfs-

    1 )

    Hyd

    roly

    tic a

    ctiv

    ity (U

    gdfs-

    1 )

    ctivitmentarepres

    ce c, the ois

    y of este

    in thl mole to

    hydophith dry dec

    labod as e of 1atic representation of the packed-bed reactor system. Key: (1) reser-ion mixture fed to reactor; (3) peristaltic pump; (4) glass columnrmented solids; (5) reaction mixture removed from reactor; (6) samp-ater jacket.

    and 0.1 mm lm thickness). Prepared sample (56 mg) in 1 mL of an internal standard solution of methyl

    1

    Fig. 2. Astate ferplotted

    to redudryinghad a mactivituse inminedoriginapossib

    Theafter lying wiactivitforcedselectetenancoate (1 mg mL ) in n-heptane. Then 1 L was injected, ratio of 1:50, using N2 as the carrier gas. The injectorr were set at 250 C. The oven program was as follows:

    min, heating at 10 C min1 to 180 C, 180 C main- min, heating at 5 C min1 to 230 C, 230 C maintainedeaks in the chromatograms were identied by com-he retention times with a standard solution. The estermass percent) was determined relative to the peak areanal standard [25].

    tion of fermented solids

    a LTEB11 was cultivated on a 1:1 mixture (w/w) of sug-sse and sunower seed meal, with both the hydrolyticcation activities of the fermented solids being measuredhe highest esterication activity was 5.8 0.3 U gdfs1,

    72 h (Fig. 2). At this time the hydrolytic activity was gdfs1, although it did reach a slightly higher value atthe fermented solids were intended for use in esteri-tions, a fermentation time of 72 h was selected for thetudies.

    y of the hydrolytic activity during drying of theolid

    rmentation, the fermented solids must be dried foreviously in our laboratory they have been dried byon [14,30], but this is an expensive process. In order

    activity in festable duriobtained wdata are av20% loss of IRD43aIV [3around 33%zopus micro

    3.3. Stabilit

    The immin organic mpresent in tit in the var

    The resiabove 95% reaction min-hexane, t100% after immobilizevents has bsolvent mothereby facthe activitytion mediaincubation rm that thstability in [33]. For ex0

    2

    4

    6

    0

    20

    40

    60

    80

    120967248240

    Este

    rific

    atio

    n ac

    tivity

    (U

    Time (h)y of the fermented solids from Burkholderia cepacia LTEB11 in solid-tion. Key: () hydrolytic activity; () esterication activity. Valuesent the mean of duplicate asks the standard error of the mean.

    osts, we evaluated alternative drying processes. Beforefresh fermented solids, obtained at 72 h of fermentation,ture content of 72% (w/w, wet basis) with a hydrolytic83 5 U gdfs1. Although the solids were intended forrication reactions, the hydrolytic activity was deter-is experiment because the high water content of theist solids used as the control meant that it was notundertake the assay for esterication activity.rolytic activity of the fermented solids was maintainedlization at 4 C for 24 h (90 4 U gdfs1) and after dry-y air in a column at 25 C for 5 h (84 6 U gdfs1). Thereased to 37 5 U gdfs1 for the sample dried in a fan-ratory oven at 30 C for 8 h. Drying in the column wasthe drying method for the remaining studies. The main-00% of activity with air drying suggests that the lipolyticrmented solids obtained with B. cepacia LTEB11 is more

    ng the drying step than that in the fermented solidsith Rhizopus, the only other organism for which similarailable. Dry air and lyophilization each caused aroundlipolytic activity for fermented solids from Rhizopus sp.1], while oven drying and lyophilization each caused

    loss of lipolytic activity for fermented solids from Rhi-sporus CPQBA 312-07 DRM [32].

    y of the fermented solid in organic solvents

    obilized lipase from B. cepacia is known for its stabilityedia [3336]. However, since the stability of the lipasehe fermented solids has not been studied, we evaluatedious solvents and reaction mixtures used in this work.dual lipolytic activity of the fermented solids remainedafter 8 h of incubation in ethanol and in solvent-freexture (Fig. 3). In n-hexane and in reaction mixture withhe activity initially increased slightly, remaining above8 h incubation. The increase of the catalytic activity ofd lipases after treatment with hydrophobic organic sol-een observed previously and is attributed to residuallecules maintaining the lipase in its open conformation,ilitating the access of the substrates to the active site in

    assay [35,3739]. In ethanol and in solvent-free reac-, the residual activities remained above 80% after 24 hand above 72% after 48 h incubation. These results con-e lipolytic activity of the fermented solids has similarorganic media to that of immobilized B. cepacia lipaseample, the residual lipolytic activity of a lipase from

  • D. Soares et al. / Biochemical Engineering Journal 81 (2013) 15 23 19

    100

    120

    140

    y (%

    )

    Fig. 3. StabilitResidual activ210 mmol L1

    oleic acid and ditions: 500 mand 200 rpm. Adetermined byrepresent the

    B. cepacia im120% after i

    3.4. Ester p

    High yietransesterilipase from[30,33,40]. ication ofessentially immobilizesolids of Bwhether it wwhile main

    In the so82% after 8with n-hexaconcentratifor the solvobtained foof the solveratio of subneed for theprocessing

    3.5. Effect o

    The effecture (9, 12 mass of olerate of prodincrease in However, a did not incrsolids madeof 12% ferm

    3.6. Esterifatty acid so

    We evalusing free fa

    0

    20

    40

    60

    80

    100

    967248240

    Subs

    trat

    e con

    vers

    ion

    (%)

    Time (h)

    ffect of the fermented solids content on the esterication reaction in afree system. Key: () 9%; () 12%; () 15% (mass of fermented solids astage of the mass of oleic acid). Reaction conditions: oleic acid 1890 mmol,5670 mmol, 40 C, 200 rpm. Values plotted represent the mean of triplicate

    the standard error of the mean.

    btained through the hydrolysis of soybean oil, soybean soap-cid oil, beef tallow and waste cooking oil (denoted as FA-SO,O, FA-BT and FA-WCO, respectively). Esterication activityee fatty acids was not directly related to the acyl chain length). Th

    gdfstivitimons [41l EP-

    reas7.0 U

    the rroprlecter.

    fect o

    peracatio

    LTE. cepO2-Pcia li0

    20

    40

    60

    80

    483624120

    Res

    idua

    l ac

    tivit

    Time (h)y of the fermented solids after incubation in different systems. Key:ity in () ethanol; () n-hexane; () 70 mmol L1 of oleic acid andof ethanol in n-hexane (i.e. reaction mixture containing solvent); ()ethanol in a 1:3 molar ratio (i.e. solvent-free reaction mixture). Con-g of dry fermented solids in 10 mL of solution, incubated at 40 Cctivities were determined relative to an initial activity of 84 U gdfs1

    the titrimetric method, using triolein as the substrate. Values plottedmean of triplicate analyses the standard error of the mean.

    mobilized on a macroporous resin ranged from 80 toncubation for 4 h in methanol, ethanol and acetone [36].

    roduction in the presence and absence of n-hexane

    lds of methyl or ethyl esters have been reported forcation and esterication reactions catalyzed by the

    B. cepacia LTEB11 in systems containing solventsIn previous investigations of our group into the ester-

    oleic acid with ethanol in n-heptane, we obtained95100% ester yield in 13 h using B. cepacia LTEB11d on Accurel [33,40] and 94% in 18 h with dry fermented. cepacia [30]. In the current work, we investigatedould be possible to remove the solvent from the system

    taining acceptable conversion rates.lvent-free reaction mixture, the conversion was only8 h, compared to a 92% conversion after 8 h obtainedne as the solvent (Table 2). However, due to the higherons within the reaction mixture, the ester productivityent-free system was essentially twice as high as thatr the reaction with n-hexane. Additional advantagesnt-free system are that, rstly, it uses a much higherstrate to fermented solids and, secondly, it avoids the

    recovery and recycling of the solvent, thereby lowering

    Fig. 4. Esolvent-a percenethanol analyses

    acids ostock aFA-SSAwith fr(Fig. 5(12.7 Ution acPseudosystemAccure

    TheSSAO (duringan appfore sereacto

    3.7. Efreactor

    Temestericepaciaoil by Phol (SiP. cepacosts.

    f the amount of fermented solids on esterication

    t of the amount of fermented solids in the reaction mix-and 15% of fermented solids, expressed relative to theic acid) was evaluated in the solvent-free system. Theuction of ethyl oleate increased signicantly with anthe amount of fermented solids from 9 to 12% (Fig. 4).further increase in the fermented solids, from 12 to 15%,ease the reaction rate signicantly. Higher amounts of

    it difcult to agitate the reaction mixture. The additionented solids was used in the remaining studies.

    cation activity of the fermented solids on differenturces

    uated the esterication activity of the fermented solidstty acids of different chain lengths, as well as free fatty

    Fig. 5. EsteriConditions: 10210 mmol L1

    represent the e highest activities were observed with palmitic acid1) and caprylic acid (11.7 U gdfs1). Higher esterica-es with palmitic acid have been obtained previously foras cepacia lipase (LPS A001526) in AOT microemulsion] and B. cepacia lipase (LPS AR01520) immobilized on100 [42].onably high esterication activity obtained with FA-

    gdfs1) is encouraging, since it is a byproduct generatedening of soybean oil [8,4346]. Its low cost may make itiate feedstock for biodiesel production [8]. It was there-d as the substrate for the experiments in the packed-bed

    f temperature on biodiesel synthesis in a packed-bed

    tures from 40 to 60 C have been utilized for the trans-n of soybean oil catalyzed by fermented solids from B.B11 [14], transesterication of beef tallow and babassuacia lipase immobilized in polysiloxane-polyvinyl alco-VA) [47], esterication of lauric acid by an immobilizedpase from Amano [35] and regioselective acylation of

    FA-SOFA-SSAO

    FA-BT15129630

    C6:0FA -C8:0FA -

    C12:0FA -C14:0FA -C16:0FA -C18:0FA -C18:1FA -C18:2FA -

    FA-WCO

    Esterification activity (U gdfs-1)

    Free

    fatty

    aci

    ds

    cation activity of the fermented solids against different fatty acids. mL of reaction mixture with n-hexane (70 mmol L1 of fatty acid andof ethanol), 800 mg of fermented solids, 40 C, 200 rpm. Values plottedmean of duplicate analyses the standard error of the mean.

  • 20 D. Soares et al. / Biochemical Engineering Journal 81 (2013) 15 23

    Table 2Esterication reactions carried out in shake asks in the presence and in the absence of solvent.

    Solventa Solvent-free

    Mass (mg) Molar concentration (mmol L1) Mass (mg) Amount (mmol)

    Reaction mixtureOleic acid 198 70 5760 20.4Ethanol 97 210 2813 61.2Fermented solids 500 500

    Results

    Ratio fermented solids/oleic acid (g/g) 2.5 0.09Results

    Conversion in time taken 92% in 8 h 82% in 88 hProductivity 50 mg gdfs1 h1 118 mg gdfs1 h1

    Reactions were done in shake asks with 10 mL of reaction mixture (molar ratio of oleic acid: ethanol of 1:3), at 40 C and 200 rpm.a Sufcient n-hexane was added to give a total volume of 10 mL.

    andrographolide by immobilized B. cepacia lipase from Amano [48].This temperature range was therefore tested for the estericationof FA-SSAO with ethanol in the packed-bed reactor. The initial reac-tion rate (based on the conversions obtained at 12 h) increased withincreasing temperature (Fig. 6). However, at 60 C the conversionsfor reaction times above 24 h were lower than those obtained atthe lower temperatures, probably due to a higher degree of dena-turation. At

    31 h; at the85% after 40

    3.8. Operatpacked-bed

    We usedcycles of 48cycle, the cthe value ofan attemptcycles, we rcase the coduring six c

    3.9. Compo

    The propester compwork by es16.5% of ethand 44.2% o

    Fig. 6. Effect o40 C; () 50 Cfatty acids fromacid to ethanomean of dupli

    obtained by de Sousa et al. [22] and Chen et al. [50], which gavebiodiesel properties that met the Brazilian [51] and European [52]standards, respectively.

    4. Discussion

    This work makes two contributions in the area of hydroester-n. Firstly, this is the rst study of hydroesterication usingn sorst ss be

    twoof faeen his icondtionpackns hutiotentiep inck.

    ntrib

    of thlow-AO

    onve

    rsio

    n (%

    ) 50 C, a 92% conversion of FA-SSAO was obtained at other temperatures the conversions were still below

    h of reaction.

    ional stability of the fermented solids in the reactor

    the same fermented solids in successive esterication h in the packed-bed reactor at 50 C (Fig. 7). In the fthonversion remained above 84% (which is above 90% of

    92% conversion that was obtained in the rst cycle). In to keep the original conversion for a larger number ofedid the reutilization experiment at 45 C (Fig. 7). In thisnversion remained above 90% of the initial conversionycles.

    sition of the ethyl esters

    erties of biodiesel are directly inuenced by the fattyosition [49]. The ethyl ester that was obtained in thisterifying FA-SSAO with ethanol contained, by mass,yl palmitate, 4.2% of ethyl stearate, 33.4% of ethyl oleatef ethyl linoleate. This composition is similar to those

    60

    80

    100

    sion

    (%)

    icatiosoybeais the step hamakescation have bacids, ttem. Seproducout in reactiocontribthe potion stfeedsto

    4.1. Co

    Oneto use that SS0

    20

    40

    483624120

    Con

    ver

    Time (h)f temperature on esterication in the packed-bed reactor. Key: (); () 60 C. Reaction conditions: 12 g of dry fermented solids, 100 g of

    soybean soapstock acid oil and 50 g of ethanol (i.e. molar ratio of fattyl of 1:3), recirculation rate of 5 mL min1. Values plotted represent thecate analyses the standard error of the mean.

    Rel

    ativ

    e c

    Fig. 7. Operatsame fermentcycles. Key: re12 g of dry ferm50 g of ethano5 mL min1.apstock acid oil (SSAO) as the feedstock. Secondly, thistudy of hydroesterication in which the esterication

    en undertaken using enzymatic catalysis. The work also contributions in the area of lipase-catalyzed esteri-tty acids. Firstly, although lipolytic fermented solidsused previously to catalyze the esterication of fattys the rst study that has done this in a solvent-free sys-ly, lipase-catalyzed transesterication reactions for the

    of biodiesel in solvent-free systems have been carrieded-bed bioreactors, but lipase-catalyzed estericationave not previously been studied in this system. Thesens, which are discussed below, combine to demonstrateal of using fermented solids to catalyze the esterica-

    a hydroesterication process that uses SSAO as the

    utions in the area of hydroesterication

    e strategies to reduce the cost of biodiesel production isvalue feedstocks. In the current work we demonstratedis an adequate feedstock for hydroesterication, with

    5060708090100

    5060708090

    100110

    conv

    ersio

    n (%

    )010203040

    010203040

    87654321

    Ori

    gina

    l

    Cycles

    ional stability of the fermented solids in the packed-bed reactor. Theed solids were used to catalyze the esterication reaction in 48-haction temperature of () 45 C and () 50 C. Reaction conditions:ented solids, 100 g of fatty acids from soybean soapstock acid oil and

    l (i.e. molar ratio of fatty acid to ethanol of 1:3), recirculation rate of

  • D. Soares et al. / Biochemical Engineering Journal 81 (2013) 15 23 21

    Table 3Studies of biodiesel production by hydroesterication with an enzymatic step.

    Reference Cavalcanti-Oliveira et al.[21]

    de Sousa et al. [22] Talukder et al. [23] This work

    Hydrolysis sCatalyst (% w t

    TemperatureSubstrates (r

    t oil

    Conversion i

    Estericatio

    Catalyst (% wTemperatureReaction mix

    Conversion i

    VEEG: vegetab lyst 1

    the free fatsubcritical wSoapstock ivegetable othe originalthe soybeanbreaking anthe acid oilacids, 28% t[8]. It sells f[43].

    Our worfrom soybeterication.that hydroebiodiesel frothe productthat have aone used w

    Additionenzymatic estericatiohydrolysis/justify this smild condiHowever, urequired, esused by Caalthough a to oil volumconditions ucally involv(200 C) [21

    The hydinverts thetial hydrolyis catalyzechemical-hThis strategsoapstocks salts [43], sstrategy, thto these cowere not fa[21] in theirwere of rela

    ysis it to ceivtudieave iing ]. Altthe ms ope

    CH/n comcatals relclessiblet-freesteri

    the gions

    pro-bedat r

    by fel cone coougin btion

    potethe step Enzymatic Enzymatic eight of oil) Thermomyces lanuginos

    (liquid lipase, 2.3%)VEEG from physic nu(10%)

    & pressure 60 C & 1 atm 40 C & 1 atm atio in v:v) Water:soybean oil (1:1) TrisHCl

    0.1 mol L1:physic nu(9:1)

    n time taken 89% in 48 h 98% in 2 h

    n step Chemical Chemical

    eight of fatty acid) Niobic acid (20%) Niobic acid (20%) & pressure 200 C & 24 atm 200 C & 34 atm ture (molar ratio) FA/methanol (1:3) FA/methanol (1:3)

    n time taken 92% in 1 h 97% in 2 h

    le enzyme extract from germinated seeds; FA: fatty acids from hydrolysis; *Amber

    ty acids that were obtained after its hydrolysis withater being efciently converted into their ethyl esters.

    s a byproduct from the alkaline neutralization step ofil rening and represents about 6% of the volume of

    crude vegetable oil [43]. In a typical industrial process, soapstock is acidied with sulfuric acid for emulsiond then separates into two phases, an aqueous phase and

    phase. The acid oil contains, by weight, 59% free fattyriacylglycerol, and around 5% di- and monoacylglycerolor approximately half the cost of rened vegetable oils

    k demonstrates, for the rst time, that acid oil derivedan soapstock can be used as a feedstock for hydroes-

    Despite the fact that previous authors have recognizedsterication is a promising technology for producingm low-value feedstocks, two of the previous studies ofion of biodiesel esters in hydroesterication processesn enzymatic step have used relatively pure oils, whileaste cooking oil (Table 3).ally, the three studies listed in Table 3 all involve anoil hydrolysis step followed by a chemically catalyzedn step. This will be referred to as the enzymatic-chemical-esterication (EH/CE) strategy. The authorstrategy by pointing out that the hydrolysis step involvestions of temperature and pressure (3060 C, 1 atm).nder these conditions long reaction times are normallypecially at the volumetric ratio of water to oil of 1:1valcanti-Oliveira et al. [21] and Talukder et al. [23],98% conversion in 2 h has been achieved with a wateretric ratio of 9:1 [22]. In addition, unlike the mild

    hydrolcatalysstep rethree sdate htion be[2123mode, batche

    Ourtages ilipase-involve48-h cybe possolventransecation,limitatcationpackedthan thalyzedoriginawith th

    Alththe maproduchas theuct of sed in the hydrolysis step, the esterication step typi-es either high pressures (2434 atm) and temperatures,22] or the use of solvents such as isooctane [23].roesterication process studied in the present work

    strategy of these previous studies, in that the ini-sis step is a chemical step while the estericationd enzymatically. This will be referred to as theydrolysis/enzymatic-esterication (CH/EE) strategy.y has two advantages over the EH/CE strategy. Firstly,can contain contaminants, such as sodium or potassiumulfur, phosphorous and metal ions [53]. In the EH/CEe enzymes used in the initial hydrolysis step are exposedntaminants and may be inactivated. These problemsced by de Sousa et al. [22] and Cavalcanti-Oliveira et al.

    studies of the EH/CE strategy because the oils they usedtively high quality. In contrast, in the CH/EE strategy the

    air drying, tof the lipaslized lipaseCosts couldsunower s[30,59].

    4.2. Contrib

    The currcontaining solvent-freemented soln-heptane can be achiumetric proEnzymatic Subcritical waterCandida rugosa (0.05%) Free of catalyst

    30 C & 1 atm 250 C & 60 atmWaste cooking oil:water(1:1)

    Water:soybean soapstockacid oil (1:1)

    100% in 10 h 95% in 1 h

    Chemical Enzymatic

    *Amberlyst 15 (100%) Fermented solid (12%)60 C & 1 atm 50 C & 1 atmFA/methanol (4:1), inisooctane

    FA/ethanol (1:3)

    99% in 2 h 93% in 31 h

    5: acidic styrene-divinylbenzene sulfonated ion-exchange resin.

    s carried out with subcritical water, such that there is nosuffer inactivation, and the lipases in the estericatione a contaminant-free fatty acid stream. Secondly, thes using the EH/CE strategy that have been published tonvolved batch operation, with the enzymatic prepara-used to catalyze the hydrolysis of a single batch of oilhough we also used the packed-bed bioreactor in batchaintenance of over 80% conversion in seven successivens up the possibility of continuous operation.EE hydroesterication process also has some advan-parison to lipase-catalyzed transesterication. Firstly,

    yzed transesterication in solvent-free media typicallyatively long reaction times (Table 4). Although we used

    for the esterication reaction, once optimized, it should to reduce the reaction time to a few hours, even in

    media, since esterication is typically much faster thancation [13,14,54]. Secondly, in enzymatic transesteri-lycerol absorbs onto the catalyst, causing mass transfer[5,5558], while this problem is avoided in hydroesteri-cesses. This explains why the operational stability of the

    reactor for the esterication of FA-SSAO was highereported for the transesterication of soybean oil cat-rmented solids from B. cepacia LTEB11 [14], where theversion of 95% was maintained only for three cycles,nversion decreasing to 62% after six cycles.h the high cost of commercial lipases remains as one ofarriers against their use in industrial processes for the

    of biodiesel, the use of fermented solids in our processntial to reduce lipase costs signicantly. The nal prod-olid-state fermentation is simply subjected to a mild

    hus avoiding the costs of recovery and immobilizatione that are associated with the production of immobi-s that are produced by submerged liquid fermentation.

    be decreased further if it were possible to replace theeed meal with a cheaper inducer of lipase production

    utions in the area of esterication of fatty acids

    ent work represents the rst time that fermented solidslipases have been used to catalyze esterication in a

    system. Previous studies of esterication with fer-ids have involved the use of organic solvents such as[30,60] and n-hexane [31]. Although high conversionseved in short times in the presence of solvents, the vol-ductivity of such reactions is lower, due to the much

  • 22 D. Soares et al. / Biochemical Engineering Journal 81 (2013) 15 23

    Table 4Studies of solvent-free biodiesel production by lipases in packed-bed reactors.

    Microorganism Support Reaction mixture Conversion in time taken System Reference

    Burkholderia cepacia LTEB11 Fermented solid FA-SSAO + ethanol 93% in 31 h Batch This workBurkholderiaRhizopus oryCandida antaCandida antaCandida anta ethanPseudomonaaRecombina ethan

    a Aspergillus suppo

    lower substhen it is esthis extra st

    Althougbiodiesel uhave all invrepresents solvent-freePacked-bedavoiding thagitated reawith the sa(w/w), 74 hthis convers

    5. Conclus

    The currthe costs ohydroesteristock acid othe second sfree systemcontains lip80% can be involving szation of thlaboratory.

    Acknowled

    This resselho NacioBrazilian gotechnologyreira Pinto,research scSoares, for tplant for th

    References

    [1] http://ww[2] M. Canak

    effects on[3] Y. Watan

    waste edi78 (2001

    [4] Y.M. Chenfrom wasConvers.

    [5] S.F.A. Halbiodieselusing ressour. Tech

    . Mehenseste

    Banko-edib. Haasm acid03) 97Loterothesis353arava

    n of trmass awas

    alystsKim, erogeion anama, Fukudducinchem.C. Salriege

    nted s10) 13ama,uda, A

    packe (201an, A.m wasses, BCanak

    fatty. Berchd oil w617. Wa

    tion t710Minamdiesel924. Cavady of

    its ap1 (20

    de Soion ofzymeal. B: .R. T

    diesells 24 . Soar-ther cepacia LTEB11 Fermented solid Soybean oil + ethanolzae Polyurethane foam Soybean oil + methanol rctica Novozyme 435 Acrylic resin Rapeseed oil + methanol rctica Novozyme 435 Acrylic resin Waste oil + methanol rctica Novozyme 435 Acrylic resin Rapeseed and soybean oils + ms cepacia Fe3O4 nanoparticle Soybean oil + methanol nt Aspergillus oryzae Polyurethane foam Rapeseed and soybean oils + m

    oryzae expressing Fusarium heterosporum lipase was immobilized within biomass

    trate concentrations. In addition, if solvents are used,sential to recover them at the end of the process, andep increases process costs.h packed-bed reactors have been used to producesing lipases in solvent-free systems, previous studiesolved transesterication (Table 4). The present studythe rst time that lipase-catalyzed esterication in a

    system has been carried out in a packed-bed reactor. reactors allow high catalyst to substrate ratios, whilee mixing problems when the same ratios are used inctors. For example, in the current work, at 40 C andme ratio of fermented solid to fatty acids, namely 12%

    were required for 84% conversion in shake asks, whileion was reached in only 43 h in the packed-bed reactor.

    ions

    ent work has proposed a potential new route to lowerf production of biodiesel using lipases, namely thecation of low-value feedstocks, such as soybean soap-il. The rst step involves hydrolysis in subcritical water;tep involves esterication of the fatty acids in a solvent-, using a reactor packed with a fermented solid thatases produced by B. cepacia. Conversion to ester of overobtained in each of seven cycles of 48 h. Further studiescale-up of the packed-bed bioreactor and characteri-e biodiesel are currently under investigation in our

    gements

    earch was supported by a grant from CNPq (Con-nal de Desenvolvimento Cientco e Tecnolgico), avernment agency for the advancement of science and. Diniara Soares, Alan Guilherme Gonc alves, Andrei Fer-

    David Mitchell and Nadia Krieger also thank CNPq forholarships. We are also grateful to Ubaldino Rodrigueshe supply of the fatty feedstocks and the use of the pilote subcritical water hydrolysis step.

    w.anp.gov.br (accessed 14.01.13).

    [6] L.Ctra

    [7] I.B.non

    [8] M.Jfro(20

    [9] E. Syn535

    [10] N. SisoBio

    [11] A. Kcat

    [12] M. hetcat

    [13] S. HH. proBio

    [14] T.FN. Kme(20

    [15] S. HFukin a102

    [16] J. Yfrolipa

    [17] M. free

    [18] H.Jsee171

    [19] Z.-Mduc102

    [20] E. bio247

    [21] E.DStuand201

    [22] J.S.cat(enCat

    [23] M.MbioFue

    [24] U.Rdowci, H. Sanli, Biodiesel production from various feedstocks and their the fuel properties, J. Ind. Microbiol. Biotechnol. 35 (2008) 431441.abe, Y. Shimada, A. Sugihara, Y. Tominaga, Enzymatic conversion ofble oil to biodiesel fuel in a xed-bed bioreactor, J. Am. Oil Chem. Soc.) 703707., B. Xiao, J. Chang, Y. Fu, P.M. Lv, X.W. Wang, Synthesis of biodieselte cooking oil using immobilized lipase in xed bed reactor, EnergyManage. 50 (2009) 668673.im, A.H. Kamaruddin, W.J.N. Fernando, Continuous biosynthesis of

    from waste cooking palm oil in a packed bed reactor: optimizationponse surface methodology (RSM) and mass transfer studies, Biore-nol. 100 (2009) 710716.

    2008.[25] European

    determinCommitt

    [26] AOCS Of[27] AOCS Of[28] A. Tiss, F.

    and activ[29] R.R. Lowr

    Am. Oil C[30] M.L.M. Fe

    Esterica95% in 46 h Batch [14]91% in 50 h Batch [13]99% in 72 h Batch [54]90%, time not mentioned Continuous [55]

    ol 9699%, time not mentioned Continuous [56]>88%, 192 h residence time Continuous [57]

    ol 96% in 14 h Batch [58]

    rt particles (BSPs), made of reticulated polyurethane foam.

    r, D.V. Sagar, S.N. Naik, Technical aspects of biodiesel production byrication a review, Renew. Sustain. Energy Rev. 10 (2006) 248268.vic-Ilic, O.S. Starnenkovic, V.B. Veljkovic, Biodiesel production fromle plant oils, Renew. Sustain. Energy Rev. 16 (2012) 36213647., P.J. Michalski, S. Runyon, A. Nunez, K.M. Scott, Production of FAME

    oil, a by-product of vegetable oil rening, J. Am. Oil Chem. Soc. 80102., Y.J. Liu, D.E. Lopez, K. Suwannakarn, D.A. Bruce, J.G. Goodwin,

    of biodiesel via acid catalysis, Ind. Eng. Chem. Res. 44 (2005)63.nan, S. Puhan, G. Nagarajan, N. Vedaraman, An experimental compar-ansesterication process with different alcohols using acid catalysts,Bioenergy 34 (2010) 9991005.hima, K. Matsubara, K. Honda, Development of heterogeneous base

    for biodiesel production, Bioresour. Technol. 99 (2008) 34393443.C. DiMaggio, S. Yan, H. Wang, S.O. Salley, K.Y.S. Ng, Performance ofneous ZrO2 supported metaloxide catalysts for brown grease esteri-d sulfur removal, Bioresour. Technol. 102 (2011) 23802386.H. Yamaji, T. Fukumizu, T. Numata, S. Tamalampudi, A. Kondo, H. Noda,a, Biodiesel-fuel production in a packed-bed reactor using lipase-g Rhizopus oryzae cells immobilized within biomass support particles,. Eng. J. 34 (2007) 273278.um, P. Villeneuve, B. Barea, C.I. Yamamoto, L.C. Cocco, D.A. Mitchell,r, Synthesis of biodiesel in column xed-bed bioreactor using the fer-olid produced by Burkholderia cepacia LTEB11, Process Biochem. 45481354.

    S. Tamalampudi, A. Yoshida, N. Tamadani, N. Kuratani, H. Noda, H.. Kondo, Process engineering and optimization of glycerol separationd-bed reactor for enzymatic biodiesel production, Bioresour. Technol.1) 1041910424.

    Li, Y. Xu, T.P.N. Ngo, S. Phua, Z. Li, Efcient production of biodieselte grease: one-pot esterication and transesterication with tandemioresour. Technol. 123 (2012) 332337.ci, J. Van Gerpen, Biodiesel production from oils and fats with high

    acids, Trans. ASAE 44 (2001) 14291436.mans, S. Hirata, Biodiesel production from crude Jatropha curcas L.ith a high content of free fatty acids, Bioresour. Technol. 99 (2008)

    21.ng, J.-S. Lee, J.-Y. Park, C.-Z. Wu, Z.-H. Yuan, Novel biodiesel pro-echnology from soybean soapstock, Korean J. Chem. Eng. 24 (2007)30.i, S. Saka, Kinetics of hydrolysis and methyl esterication for

    production in two-step supercritical methanol process, Fuel 85 (2006)83.lcanti-Oliveira, P.R. Silva, A.P. Ramos, D.A.G. Aranda, D.M.G. Freire,soybean oil hydrolysis catalyzed by Thermomyces lanuginosus lipaseplication to biodiesel production via hydroesterication, Enzyme Res.10) 18.usa, E.D. Cavalcanti-Oliveira, D.A.G. Aranda, D.M.G. Freire, Appli-

    lipase from the physic nut (Jatropha curcas L.) to a new hybrid/chemical) hydroesterication process for biodiesel production, J. Mol.Enzym. 65 (2010) 133137.alukder, J.C. Wu, L.P.L. Chua, Conversion of waste cooking oil to

    via enzymatic hydrolysis followed by chemical esterication, Energy(2010) 20162019.es, Processo de ciso contnua a alta presso tendo vaporizador dem como fonte geradora de energia trmica, Patent BR 0604172-8 A, Standard EN 14103, Fat and oil derivatives, fatty acid methyl esters,ation of ester and linolenic acid methyl ester contents, CEN, Europeanee for Standardization, 2003.cial Methods and Recommended Practices, Cd 3-25, fourth ed., 1990.cial Methods and Recommended Practices, Ca 5a-40, fourth ed., 1989.

    Carriere, R. Verger, Effects of gum arabic on lipase interfacial bindingity, Anal. Biochem. 294 (2001) 3643.y, J.I. Tinsley, Rapid colorimetric determination of free fatty acids, J.hem. Soc. 53 (1976) 470472.rnandes, E.B. Saad, J.A. Meira, L.P. Ramos, D.A. Mitchell, N. Krieger,tion and transesterication reactions catalysed by addition of

  • D. Soares et al. / Biochemical Engineering Journal 81 (2013) 15 23 23

    fermented solids to organic reaction media, J. Mol. Catal. B: Enzym. 44 (2007)813.

    [31] A. Martinez-Ruiz, H.S. Garcia, G. Saucedo-Castaneda, E. Favela-Torres, Organicphase synthesis of ethyl oleate using lipases produced by solid-state fermen-tation, Appl. Biochem. Biotechnol. 151 (2008) 393401.

    [32] D. Alberton, D.A. Mitchell, J. Cordova, P. Peralta-Zamora, N. Krieger, Produc-tion of a fermented solid containing lipases of Rhizopus microsporus and itsapplication in the pre-hydrolysis of a high-fat dairy wastewater, Food Technol.Biotechnol. 48 (2010) 2835.

    [33] T.F.C. Salum, A.M. Baron, E. Zago, V. Turra, J. Baratti, D.A. Mitchell, N. Krieger,An efcient system for catalyzing ester synthesis using a lipase from anewly isolated Burkholderia cepacia strain, Biocatal. Biotransform. 26 (2008)197203.

    [34] S.T. Pan, X. Liu, Y.D. Xie, Y.Y. Yi, C. Li, Y.J. Yan, Y. Liu, Esterication activityand conformation studies of Burkholderia cepacia lipase in conventional organicsolvents, ionic liquids and their co-solvent mixture media, Bioresour. Technol.101 (2010) 98229824.

    [35] Y. Liu, X.A. Zhang, H. Tan, Y.J. Yan, B.H. Hameed, Effect of pretreatment by differ-ent organic solvents on esterication activity and conformation of immobilizedPseudomonas cepacia lipase, Process Biochem. 45 (2010) 11761180.

    [36] T. Liu, Y. Liu, X. Wang, Q. Li, J. Wang, Y. Yan, Improving catalytic performanceof Burkholderia cepacia lipase immobilized on macroporous resin NKA, J. Mol.Catal. B: Enzym. 71 (2011) 4550.

    [37] F. Secundo, S. Spadaro, G. Carrea, P.L.A. Overbeeke, Optimization of Pseu-domonas cepacia lipase preparations for catalysis in organic solvents,Biotechnol. Bioeng. 62 (1999) 554561.

    [38] J.C. Wu, B.D. Song, A.H. Xing, Y. Hayashi, M.M.R. Talukder, S.C. Wang, Esterica-tion reactions catalyzed by surfactant-coated Candida rugosa lipase in organicsolvents, Process Biochem. 37 (2002) 12291233.

    [39] E. Su, D. Wei, Improvement in lipase-catalyzed methanolysis of triacylglycerolsfor biodiesel production using a solvent engineering method, J. Mol. Catal. B:Enzym. 55 (2008) 118125.

    [40] A.M. Baron, E.C. Zago, D.A. Mitchell, N. Krieger, S.P.I.L., Simultaneous produc-tion and immobilization of lipase from Burkholderia cepacia LTEB11, Biocatal.Biotransform. 29 (2011) 1924.

    [41] H. Stamatis, A. Xenakis, E. Dimitriadis, F.N. Kolisis, Catalytic behavior of Pseu-domonas cepacia lipase in W/O microemulsions, Biotechnol. Bioeng. 45 (1995)3341.

    [42] G. Pencreach, J.C. Baratti, Properties of free and immobilised lipase fromBurkholderia cepacia in organic media, Appl. Microbiol. Biotechnol. 52 (1999)276280.

    [43] M.J. Haas, S. Bloomer, K. Scott, Simple, high-efciency synthesis of fatty acidmethyl es

    [44] M.J. Haasbiodieselproduced

    [45] M.J. Haaslow value86 (2005)

    [46] C.Y. Lin, Y.W. Lin, Fuel characteristics of biodiesel produced from a high-acidoil from soybean soapstock by supercritical-methanol transesterication, Ener-gies 5 (2012) 23702380.

    [47] P.C.M. Da Ros, G.A.M. Silva, A.A. Mendes, J.C. Santos, H.F. de Castro, Evalua-tion of the catalytic properties of Burkholderia cepacia lipase immobilized onnon-commercial matrices to be used in biodiesel synthesis from different feed-stocks, Bioresour. Technol. 101 (2010) 55085516.

    [48] Z.G. Chen, R.X. Tan, M. Huang, Efcient regioselective acylation of andro-grapholide catalyzed by immobilized Burkholderia cepacia lipase, ProcessBiochem. 45 (2010) 415418.

    [49] G. Knothe, A.C. Matheaus, T.W. Ryan, Cetane numbers of branched and straight-chain fatty esters determined in an ignition quality tester, Fuel 82 (2003)971975.

    [50] G. Chen, M. Ying, W. Li, Enzymatic conversion of waste cooking oils into alter-native fuel-biodiesel, Appl. Biochem. Biotechnol. 132 (2006) 911921.

    [51] Brasil, Resoluco ANP n. 7, de 19 de marc o de 2008. Estabelece a especicac o dobiodiesel a ser comercializado pelos diversos agentes econmicos autorizadosem todo territrio nacional, Dirio Ocial da Unio, Braslia, DF, 2008.

    [52] European Standard EN 14214, Automotive fuels Fatty acid methyl esters(FAME) for diesel engines Requirements and test methods, CEN, EuropeanCommittee for Standardization, 2008.

    [53] W.D. Morgan, Production of alkyl ester for use as biodiesel comprises obtainingvegetable oil soapstock as feedstock, performing saponication, acidulationand separation steps to yield fatty acid layer, and esterifying the fatty acidlayer, Patent US0036705A1, 2009.

    [54] M. Hajar, S. Shokrollahzadeh, F. Vahabzadeh, A. Monazzami, Solvent-freemethanolysis of canola oil in a packed-bed reactor with use of Novozym 435plus loofa, Enzyme Microb. Technol. 45 (2009) 188194.

    [55] Y. Watanabe, Y. Shimada, A. Sugihara, H. Noda, H. Fukuda, Y. Tominaga, Contin-uous production of biodiesel fuel from vegetable oil using immobilized Candidaantarctica lipase, J. Am. Oil Chem. Soc. 77 (2000) 703707.

    [56] S. Hama, S. Tamalampudi, A. Yoshida, N. Tamadani, N. Kuratani, H. Noda, H.Fukuda, A. Kondo, Enzymatic packed-bed reactor integrated with glycerol-separating system for solvent-free production of biodiesel fuel, Biochem. Eng.J. 55 (2011) 6671.

    [57] X. Wang, X. Liu, C. Zhao, Y. Ding, P. Xu, Biodiesel production in packed-bed reac-tors using lipase-nanoparticle biocomposite, Bioresour. Technol. 102 (2011)63526355.

    [58] A. Yoshida, S. Hama, N. Tamadani, H. Fukuda, A. Kondo, Improved performanceof a packed-bed reactor for biodiesel production through whole-cell biocatal-ysis employing a high-lipase-expression system, Biochem. Eng. J. 63 (2012)7680.

    amili. Mitc

    usincerol aomanid statrobioters from soapstock, J. Am. Oil Chem. Soc. 77 (2000) 373379., K.M. Scott, T.L. Alleman, R.L. McCormick, Engine performance of

    fuel prepared from soybean soapstock: a high quality renewable fuel from a waste feedstock, Energy Fuels 15 (2001) 12071212., Improving the economics of biodiesel production through the use of

    lipids as feedstocks: vegetable oil soapstock, Fuel Process. Technol. 10871096.

    [59] D. CD.Ationgly

    [60] D. RSolMicos-Neto, C. Bugay, A.P. de Santana, T. Joslin, L.M. de Souza, G.L. Sassaki,hell, N. Krieger, Production of rhamnolipids in solid-state cultiva-g a mixture of sugarcane bagasse and corn bran supplemented withnd soybean oil, Appl. Microbiol. Biotechnol. 89 (2011) 13951403.o, R. Gandol, R. Gualandris, A. Romano, F. Starace, R. Villa, F. Molinari,e fermentation for the production of lipolytic fungal enzymes, Ann.l. 57 (2007) 561564.

    Biodiesel production from soybean soapstock acid oil by hydrolysis in subcritical water followed by lipase-catalyzed ester...1 Introduction2 Materials and methods2.1 Raw materials2.2 Hydrolysis of fat feedstocks2.3 Microorganism2.4 Solid-state fermentation2.5 Drying and preparation of the fermented solid2.6 Activity measurement with the fermented solid2.7 Stability of the fermented solid2.8 Preliminary studies of biodiesel synthesis2.9 Solvent-free esterification in a packed-bed reactor2.10 GC analysis

    3 Results3.1 Production of fermented solids3.2 Stability of the hydrolytic activity during drying of the fermented solid3.3 Stability of the fermented solid in organic solvents3.4 Ester production in the presence and absence of n-hexane3.5 Effect of the amount of fermented solids on esterification3.6 Esterification activity of the fermented solids on different fatty acid sources3.7 Effect of temperature on biodiesel synthesis in a packed-bed reactor3.8 Operational stability of the fermented solids in the packed-bed reactor3.9 Composition of the ethyl esters

    4 Discussion4.1 Contributions in the area of hydroesterification4.2 Contributions in the area of esterification of fatty acids

    5 ConclusionsAcknowledgementsReferences