Direct Transformation of Fungal Biomass from Submerged Cultures into Biodiesel

3173r 2010 American Chemical Society pubs.acs.org/EF

Energy Fuels 2010, 24, 3173–3178 : DOI:10.1021/ef9015872Published on Web 04/02/2010

Direct Transformation of Fungal Biomass from Submerged Cultures into Biodiesel

Gemma Vicente,† L. Fernando Bautista,† Francisco J. Guti�errez,† Rosalı́a Rodrı́guez,† Virginia Martı́nez,†

Rosa A. Rodrı́guez-Fr�ometa,‡ RosaM. Ruiz-V�azquez,‡ Santiago Torres-Martı́nez,‡ and Victoriano Garre*,‡

†Department of Chemical and Environmental Technology, ESCET, Universidad Rey Juan Carlos, C/Tulip�an s/n,28933 M�ostoles, Madrid, Spain, and ‡Departamento de Gen�etica y Microbiologı́a (Unidad Asociada al IQFR-CSIC),

Facultad de Biologı́a, Universidad de Murcia, 30071 Murcia, Spain

Received July 29, 2009. Revised Manuscript Received March 11, 2010

Diminishing fossil fuel reserves and the increase in their consumption indicate that strategies need to bedeveloped to produce biofuels from renewable resources. Biodiesel offers advantages over other petro-leum-derived fuel substitutes, because it is comparatively environmentally friendly and an excellent fuel forexisting diesel engines. Biodiesel, which consists of fatty acid methyl esters (FAMEs), is usually obtainedfrom plant oils. However, its extensive production from oil crops is not sustainable because of the impactthis would have on food supply and the environment. Microbial oils are postulated as an alternative toplant oils, but not all oleaginous microorganisms have ideal lipid profiles for biodiesel production. On theother hand, lipid profiles could be modified by genetic engineering in some oleaginous microorganisms,such as the fungusMucor circinelloides, which has powerful genetic tools. We show here that the biomassfrom submerged cultures of the oleaginous fungus M. circinelloides can be used to produce biodiesel byacid-catalyzed direct transformation, without previous extraction of the lipids. Direct transformation,which should mean a cost savings for biodiesel production, increased lipid extraction and demonstratedthat structural lipids, in addition to energy storage lipids, can be transformed into FAMEs.Moreover, theanalyzed properties of the M. circinelloides-derived biodiesel using three different catalysts (BF3, H2SO4,and HCl) fulfilled the specifications established by the American standards and most of the Europeanstandard specifications.

1. Introduction

Society is facing an unprecedented situation with regardto the fundamental sources of its raw materials and energy.Petroleum, the fuel that has drivenmodern society for the lastcentury, is showing signs of scarcity.1,2 Many renewable fuelalternatives are under study,3 but ethanol and biodiesel arealready available in petrol stations. Biodiesel, which consistsof fatty acid methyl esters (FAMEs), has many advantages,such as high energy density, great lubricity, fast biodegrada-tion rate, and reduced emissions of sulfur, aromatic com-pounds, andparticulatematter.4However, biodiesel adoptionis complicated because it competes with the food industry forthe main raw material input, plant oils, and the worldwidesupply of plant oils is limited by land andwater availability.4,5

Moreover, a rapid expansion in biodiesel production capacityis being observed in not only developed countries, e.g., UnitedStates and European Union, but also developing countries.To meet the demand of this industry, oil sources other than

crop oils should be quickly developed.6 One way to increaseworld oil production thatwould cause a low ecosystem impactis to use lipids from oleaginous microorganisms (also calledsingle-cell oils), which present many significant advanta-ges over plants. Oleaginous microorganisms, such as yeasts,fungi, bacteria, and microalgae, can accumulate high levels oflipids7-14 (Table 1) and do not require arable land, so thatthey donot competewith foodproduction.More particularly,photosynthetic microalgae have attracted attention and invest-ment because they capture carbon dioxide in lipids using sun-light. However, their growth in bioreactor systems is proble-matic because of the light supply requirement.6,15 Oleaginousyeasts and fungi have also been considered as potential oilsources for biodiesel production because they accumulate largeamounts of lipids. Among these microorganisms, particularattention has been dedicated to various oleaginous zygomyce-tes species, such as Mortierella isabelina and Cunninghamella

*To whom correspondence should be addressed: Departamento deGen�etica y Microbiologı́a, Facultad de Biologı́a, Universidad de Murcia,30071 Murcia, Spain. Telephone: þ34-868887148. Fax: þ34-868883963.E-mail: [email protected].(1) Grant, L. Science 2005, 309, 52–54.(2) Vasudevan, P. T.; Briggs, M. J. Ind. Microbiol. Biotechnol. 2008,

35, 421–430.(3) Wackett, L. P. Microb. Biotechnol. 2008, 1, 211–225.(4) Durrett, T. P.; Benning., C.; Ohlrogge, J. Plant J. 2008, 54, 593–

607.(5) Pinzi, S.; Garcia, I. L.; Lopez-Gimenez, F. J.; Luque de Castro,

M. D.; Dorado, G.; Dorado, M. P. Energy Fuels 2009, 23, 2325–2341.(6) Li, Q.; Du,W.; Liu, D.Appl.Microbiol. Biotechnol. 2008, 80, 749–

756.

(7) Meng, X.; Yang, J.; Xu, X.; Zhang, L.; Nie, Q.; Xian,M.RenewableEnergy 2009, 34, 1–5.

(8) Chisti, Y. Biotechnol. Adv. 2007, 25, 294–306.(9) Illman, A. M.; Scragg, A. H.; Shales, S. W. Enzyme Microb.

Technol. 2000, 27, 631–635.(10) Gouda, M. K.; Omar, S. H.; Aouad, L. M. World J. Microbiol.

Biotechnol. 2008, 24, 1703–1711.(11) Papanikolaou, S.; Komaitis, M.; Aggelis, G. Bioresour. Technol.

2004, 95, 287–291.(12) Fakas, S.; Papanikolaou, S.; Galiotou-Panatoyou, M.; Komaitis,

M.; Aggelis, G. J. Appl. Microbiol. 2008, 105, 1062–1070.(13) Fakas, S.; Papanikolaou, S.; Batsos, A.; Galiotou-Panatoyou,

M.; Mallouchos, A.; Aggelis, G. Biomass Bioenergy 2009, 33, 573–580.(14) Vicente, G.; Bautista, L. F.; Rodrı́guez, R.; Guti�errez, F. J.;

S�adaba, I.; Ruiz-V�azquez, R. M.; Torres-Martı́nez, S.; Garre, V.Biochem. Eng. J. 2009, 48, 22–27.

(15) Rittmann, B. E. Biotechnol. Bioeng. 2008, 100, 203–212.

3174

Energy Fuels 2010, 24, 3173–3178 : DOI:10.1021/ef9015872 Vicente et al.

echinulata, whichmay accumulate up to 86 and 57%of lipids indry biomass, respectively.11-13 These fungi are able to growandaccumulate large amounts of lipids in cultures containing rawglycerol derived from biodiesel production as a carbon source.Glycerol is themajorbyproductof thebiodiesel production, andits recycling to produce oleaginous microbial biomass couldsignificantly decrease the cost of biodiesel production.13

Biodiesel is conventionally produced by transesterificationof extracted triacylglycerides with methanol, but a single-stepmethod has been developed that transforms lipids present indried microbial biomass into FAMEs, without previous lipidextraction.16 This method combines the lipid extraction, theacid-catalyzed transesterificationof the extracted saponifiablelipids, and the acid-catalyzed esterification of the extractedfree fatty acids in one step and was initially proposed becauseof the substantial reduction in both time and solvents that thistechnique offers for analytical purposes.17 Similar proceduresthat avoid the lipid extraction step have already been deve-loped.13,18-20 However, most of them involve a previoustransmethylation step and do not include an acid-catalyzedtransesterification and esterification.13,18,19

Biodiesel quality depends upon the fatty acid compositionof raw materials, and consequently, not all microorganismscan be used as a feedstock for biodiesel production.4,5 Thus, acareful characterization of the lipid composition of eachmicrobial candidate should be carried out before its adoption

by the industry. One way to generate microorganisms withideal lipid composition for biodiesel production could be bymeans of genetic manipulation of key genes.4,5 However,microorganisms considered thus far as a feedstock for biodie-sel production lack appropriate genetic engineering tech-niques to improve fatty acid profiles that would producehigh-quality biodiesel.16 Besides, their genomes have not beensequenced, which makes it even more difficult to improvestrategies based on genetic manipulation.

In contrast, the oleaginous fungus Mucor circinelloides,which was used for the first commercial production of micro-bial lipids,21 has its genome sequenced anda large collectionofgenetic engineering techniques for its manipulation. Thesetechniques include the expression of genes using autoreplica-tive plasmids and inactivation of genes by disruption22 or genesilencing (RNAi).23 In addition, the regulation of lipid accu-mulation in this fungus has been extensively studied fordecades,24,25 and key genes have been identified.26 Moreover,the possibility to manipulate lipid accumulation in M. circi-nelloides using genetic engineering techniques has been recen-tly proven. Thus, overexpression of malic enzyme, which hasbeen postulated to be the rate-limiting step for fatty acidbiosynthesis in M. circinelloides, led to a 2.5-fold increase inlipid accumulation.27

TheM.circinelloides lipids extracted formyceliumgrown in asolid medium have been suggested as a suitable feedstock toproduce biodiesel.14 Biodiesel was produced by acid-catalyzedtransesterification/esterification because of its high free fattyacid content (31.6 ( 1.3%) following two different app-roaches: transformation of extracted microbial lipids andacid-catalyzed direct transformation of microbial dry mass.The FAME yield was significantly higher in the direct transfor-mation than in the two-stepprocess,with theFAMEpurity alsobeing higher in the direct method. However, growth in a solidmedium is unfeasible for the industry,which shouldusebiomassfrom submerged cultures. Therefore, we describe here thecharacterization of the lipids accumulated by M. circinelloidesmycelia grown in submerged liquid cultures and the acid-catalyzed direct transformation of theM. circinelloides biomassinto biodiesel, without previous extraction of those lipids. Inaddition,we also show that thebiodiesel obtained complieswiththe current existing standards, theASTMD6751 standard in theUnited States and most of the specifications in the EN 14213and 14214 standards in the European Union.

2. Experimental Section

2.1. Strains and Growth Conditions. The strain MU241,28

derived from R7B29 after replacement of its leuA mutant alleleby a wild-type allele, was used as a wild-type strain to producefungal biomass. For biomass production, 1 � 105 spores/mL

Table 1. Oleaginous Microorganisms Used for

Single-Cell Oil Production

(16) Liu, B.; Zhao, Z. B. J. Chem. Technol. Biotechnol. 2007, 82, 775–780.(17) Lewis, T.; Nichols, P. D.; McMeekin, T. A. J. Microbiol.

Methods 2000, 43, 107–116.(18) Rodrı́guez-Ruiz, J.; Belarbi, E.-H.; Garcı́a S�anchez, J. L.; L�opez

Alonso, D. Biotechnol. Technol. 1998, 12, 689–691.(19) Weete, J.D.; Shewmaker, F.;Gandhi, S.R. J.Am.Oil Chem.Soc.

1998, 75, 1367–1372.(20) Johnson, M. B.; Wen, Z. Energy Fuels 2009, 23, 5179–5183.

(21) Ratledge, C. Biochimie 2004, 86, 807–815.(22) Navarro, E.; Lorca-Pascual, J. M.; Quiles-Rosillo, M. D.; Nicol�as,

F. E.; Garre, V.; Torres-Martı́nez, S.; Ruiz-V�azquez, R. M. Mol. Genet.Genomics 2001, 266, 463–470.

(23) Nicol�as, F. E.; Torres-Martı́nez, S.; Ruiz-V�azquez,R.M.EMBOJ. 2003, 22, 3983–3991.

(24) Aggelis, G.; Ratomahenina, R.; Arnaud, A.; Galzy, P.; Martin-Privat, P.; Perraud, J. P.; Pina, M.; Graille, J.Oleagineux 1988, 43, 311–317.

(25) Aggelis, G.; Pina, M.; Graille, J. Oleagineux 1990, 45, 229–232.(26) Wynn, J. P.; bin Abdul, H. A.; Ratledge, C. Microbiology 1999,

145, 1911–1917.(27) Zhang, Y.; Adams, I. P.; Ratledge, C. Microbiology 2007, 153,

2013–2025.(28) Silva, F.; Navarro, E.; Pe~naranda,A.;Murcia-Flores, L.; Torres-

Martı́nez, S.; Garre, V. Mol. Microbiol. 2008, 70, 1026–1036.(29) Roncero, M. I. G. Carlsberg Res. Commun. 1984, 49, 685–690.

3175

Energy Fuels 2010, 24, 3173–3178 : DOI:10.1021/ef9015872 Vicente et al.

were inoculated in a 500 mL flask with 100 mL of YNB2XGliquid medium (20 g/L glucose, 1.5 g/L ammonium sulfate,1.5 g/L glutamic acid, 0.5 g/L yeast nitrogen base without aminoacids and ammonium sulfate, 1 mg/L nicotinic acid, and 1mg/Lthiamine at pH 4.5) and incubated in the dark for 24, 48, 72, or96 h at 26 �C and 250 rpm. Culture pH was measured every 24 hand manually adjusted by the addition of 1 M NaOH.

2.2. Analysis of Cell Lipids. Mycelia harvested by filtrationusing Whatman Paper No. 1 were dried between paper towels,frozen in liquid nitrogen, lyophilized, weighed to estimate drymass, and ground using a mortar and pestle. Cell lipids wereextracted as previously described.30

Characterization of cell lipids was performed following stan-dard methods when possible. Free fatty acids, tri-, di-, andmonoglycerides, FAMEs, carotenoids, sterol esters, sterols andtocoferols, retinoids and polar lipids in microbial oil were identi-fiedandquantifiedbyTLCanalysis.Chromatographic separationwas developed in 20 � 20 cm silica-coated aluminum plates(Alugram Sil G/UV,Macherey-Nagel GmbH, D€uren, Germany)using a solvent mixture of 88% (v) n-hexane, 11% (v/v) diethylether, and 1% (v/v) glacial acetic acid. Visualization was carriedout by staining with iodine. Digital image analyses of stainingplates were performed with Un-Scan-It Gel 6.1 software (SilkScientific, Inc., Orem, UT), and the lipid compositions werequantified by the corresponding calibration curves.

Free fatty acid content in the lipid fraction extracted from themicroorganisms was measured following a colorimetric proce-dure31 based on the formation of cupric soaps and further quan-tification of the chromophore complex by absorbance at 715 nmin a Cary 500 spectrophotometer (Varian, Inc., Palo Alto, CA).

The phosphorus content in microbial oil was determined byinductively coupled plasma-optical emission spectrometry(ICP-OES) using aVistaAXmodel (Varian, Inc.). The analysiswas performed according to EN 14107:2003 standard.

Fatty acid profiles of microbial, rapeseed, and sunflower oilswere performed by gas chromatography (GC) in a CP-3800gas chromatograph (Varian, Inc.) fitted with a flame ioniza-tion detector (FID) and TRB-FFAP capillary column (30 mlength, 0.32 mm internal diameter, and 0.25 μm film thickness,Teknokroma, Barcelona, Spain). Prior to GC analysis, the oilsamples were transformed into their corresponding methylesters by saponification in 0.5 M KOH in methanol solution(30 min at 90 �C) followed by treatment with 14% borontrifluoride in methanol (10 min at 90 �C) and extraction withn-hexane/water. Finally, 3 μL of the organic phase containingFAMEs was injected into the capillary column, where theseparation was achieved using a temperature ramp (1 �C/min)from 150 to 240 �C at a flow rate of 1 mL/min (injector tempe-rature, 180 �C; detector temperature, 280 �C; injection mode,splitless). Identification of chromatographic peaks was per-formed by a comparison to a FAME standard mixture (refe-rence 07131-1AM, Supelco, Bellefonte, PA) and quantificationbymeans of external standards and their corresponding calibra-tion curve. The iodine numberwas calculated as described in EN14214:2003 standard from the free fatty acid profile.

2.3. Direct Acid-Catalyzed Transesterification/Esterification

Reactions. M. circinelloides biomass was transesterified/ester-ified by stirring (900 rpm) with a solution of the catalyst(BF3, H2SO4, or HCl) in a closed container at 65 �C for 8 h. Inthis direct process, a 10:1 methanol/chloroform (v/v) mixturewas used as a reagent-solvent system, where the appropriateamount of the corresponding acid catalyst was dissolved. Theobtainedmixturewas dilutedwithwater and then extractedwithhexane and diethyl ether using a centrifuge. The solvents wereremoved in a rotary evaporator, and the residue (FAMEs) was

weighed to calculate the yield and then analyzed to determine itsquality as biodiesel, following standard methods according toEuropean Union specifications (EN 14214).

3. Results and Discussion

3.1. Biomass Production and Lipid Characterization. Toproduce biodiesel, M. circinelloides biomass was obtainedfromtheprototrophic strainMU241grown ina liquidmedium(YNB2XG) containing glucose as a carbon source (20 g/L). Inour experimental conditions, the fungus grew very quicklybecause it consumed all of the available glucose and stoppedgrowing in the first 48 h after inoculation (Figure 1). Similarfast growth has been observed in not only M. circinelloides,26

but also other Mucorales, such asM. isabellina.32 Lipid accu-mulation was high in the first analyzed time (24 h) and onlyincrease slightly afterward. Although culture kinetic compar-isons are difficult, particularly when different strains or cultureconditions are used, similar lipid accumulation kinetics werepreviously observed in cultures ofM. circinelloides.26 In addi-tion, the fatty acid profile of the lipid extracted fromM. circinelloides did not change significantly with the fermen-tation time (data not shown).

After 96 h of growth, the fungus was clearly in stationaryphase and no further increases in lipidswere expected. In thattime, a 4.17 ( 0.25 g/L fungal biomass with a total lipidcontent of 22.9( 0.9% drymass was obtained. Nonetheless,not all lipids obtained from microbial biomass are suitablefor making biodiesel. Only saponifiable lipids and free fattyacids (also referred to as oils) can be converted into FAMEs,which can be used as biodiesel if they complywith the currentstandards (ASTM D6751 in the United States or EN 14213and 14214 in the European Union). The saponifiable lipidsand free fatty acids (including energy storage and structurallipids) were 98.0 ( 1.3% of the total lipids extracted fromM. circinelloides biomass, with the main components beingtriglycerides, polar lipids (phospholipids, sphingolipids, andsaccharolipids), and free fatty acids (Table 2). In particular,the quantity of sphingolipids and saccharolipids producedby M. circinelloides was very high (around 54% of totallipids). The amount of neutral lipids (mono-, di-, and trigly-cerides) accumulated by M. circinelloides was 23.8%. Neu-tral lipids were comprised of mainly triglycerides (22.6 (1.3%). In addition, the proportion of phospholipids in this

Figure 1.Kinetics of biomass production (2), lipid biosynthesis (O),and glucose consumption (b) inM. circinelloides cultures. Data arepresented as mean values from duplicate experiments.

(30) Folch, J.; Lees, M.; Stanley, G. H. S. J. Biol. Chem. 1957, 226,497–509.(31) Lowry, R. R.; Tinsley, I. J. J. Am. Oil Chem. Soc. 1976, 53, 470–

472.(32) Papanikolaou, S.; Galiotou-Panatoyou, M.; Fakas, S.; Komaitis,

M.; Aggelis, G. Eur. J. Lipid Sci. Technol. 2007, 109, 1060–1070.

3176

Energy Fuels 2010, 24, 3173–3178 : DOI:10.1021/ef9015872 Vicente et al.

fungus was 16%. Significantly lower proportions of struc-tural lipids (sphingolipids, saccharolipids, and phospho-lipids) were observed in the biomass from stationary culturesof other oleaginous fungi, such asCunninghamella echinulata,33

whereas the amount of neutral lipids (storage lipids) washigher at this stage. The level of neutral lipids (storage lipids)increased with time during the cultivation of this fungus,which means a decrease in the relative proportion of all ofthe structural lipids with this variable. In fact, the amount ofstructural lipids in a microorganism is concrete, and there-fore, it has to keep constant with time. In contrast, lipidaccumulation inM. circinelloideswas 18.9% at 24 h, increas-ing only slightly after this time (Figure 1). In this case, thequantity of neutral lipids did not change significantly withthe fermentation time, which justifies the relative high pro-portion of phospholipids, sphingolipids, and saccharolipidsat the stationary stage. Although free fatty acid levels werestill high (3.6 ( 0.6%), they were substantially reduced incomparison to those observed in biomass from solidmedium(31.6 ( 1.3%).14 The non-saponifiable lipid fraction, whichconsisted of small amounts of carotenoids, sterols, tocopher-ols, and retinoids (Table 2), was also reduced in these cultureconditions (1.96%) in comparison to the solid medium(13.5%), probably because of the absence of light.22 Theseresults suggest that the fungal biomass from liquid cultures inthe dark shows better characteristics for biodiesel produc-tion than that from solid cultures.

3.2. Biodiesel Production. The high concentration of freefatty acids (3.6 ( 0.6%) in M. circinelloides determines thatan acid-catalyzed process is more suitable for producingbiodiesel than an alkali one to avoid yield losses from freefatty acid neutralization.34 Methods for simultaneous lipidextraction and transesterification involving a previous trans-methylation step have been previously used with zygomy-cetes fungi, but they were avoided because of their lowyields.13 Therefore, the acid-catalyzed direct transformationmethod16,17 (Figure 2) was applied to driedmycelial biomassusingmethanol and chloroform as solvents andH2SO4,HCl,and BF3 as acid catalysts, all of which are commonly used in

esterification or transesterification reactions.35-38Operatingconditions (temperature, time, and solvent ratio) were pre-viously optimized usingM. circinelloides biomass from solidmedium.14 Using optimal reaction conditions (8 h at 65 �C),biodiesel yields were 18.9, 18.9, and 18.4% relative to the drymass of M. circinelloides, using H2SO4, HCl, and BF3,respectively. These yields were even slightly higher than thecorresponding theoretical yield calculated for this micro-organism (18.1%), indicating that acid-catalyzed direct tran-sesterification/esterification of fungal biomass can be app-lied to M. circinelloides biomass from submerged culturesbecause it improves the amount of total lipids extracted incomparison to the conventional methods for lipid extractionfrom microorganisms.30,39 This observation is supportedby previous works describing increased recovery of fattyacids from microorganisms by direct transterification tech-niques.17,40 Interestingly, these results also indicate thatsaponifiable lipids other than triglycerides, such as phospho-lipids, sphingolipids, and saccharolipids (Table 2), are trans-formed into FAMEs by this method and should be consi-dered as substrates for FAME obtention.

At the end of the procedure, methanol and chloro-form were recovered and recirculated through the process(Figure 2).

3.3. Quality Analysis of the Biodiesel. The quality of thebiodiesel produced in the one-step procedure was deter-mined according to the EN 14214 specifications, and theresults were compared to the corresponding specified bio-diesel limits in standards EN 14213 (European Union), EN14214 (European Union), and ASTM D6751 (UnitedStates). Dependent upon the catalyst, the ester contentranged between 99.0 and 99.2% (Table 3), which is signifi-cantly higher than the corresponding specified minimumvalue in the EuropeanUnion standard (96.5%). These valueswere higher and the reaction was faster than those repor-ted for other oleaginous microorganisms, in which an acid-catalyzed direct transformation method was also used.16

Futhermore, the amounts of all byproduct analyzed werebelow the maximum allowed values for American andEuropean standards. Thus, the contents of individual glyce-rides (mono-, di-, and triglycerides) were within the biodieselspecifications, indicating that the transesterification andesterification reactions were complete. The free glycerolcontent was lower than the two standard limits, indicatingthat the glycerol residues were eliminated during the purifi-cation treatment. Besides, the individual glyceride and freeglycerol levels were below the established limits. The totalglycerol content also met all of the standards. The acidvalues, which depend upon the free fatty acid content,were also within the specifications in all reactions. In addi-tion, non-saponifiable lipids were not detected in theM. circinelloides-derived biodiesel, which means that thesetypes of lipids were also eliminated during the purificationstage. Nonetheless, the biodiesel obtained had small quan-tities of polar lipids, which were lower than 0.9% in all cases(Table 3). These compounds are residuals of nonconvertedpolar lipids, and they are not considered in the biodieselspecifications established thus far.

Table 2. Composition of the Lipids Extracted from

M. circinelloides after 96 h of Growth

(33) Fakas, S.; Papanikolaou, S.; Galiotou-Panatoyou, M.; Komaitis,M.; Aggelis, G. Appl. Microbiol. Biotechnol. 2006, 73, 676–683.(34) Vicente, G.; Martı́nez, M.; Aracil, J. Energy Fuels 2006, 20, 394–

398.(35) Formo, M. W. J. Am. Oil Chem. Soc. 1954, 31, 548–559.(36) Freedman, B.; Pryde, E. H.;Mounts, T. L. J. Am. Oil Chem. Soc.

1984, 61, 1638–1643.

(37) Canakci, M.; Van Gerpen, J. Trans. ASAE 1999, 42, 1203–1210.(38) Canakci, M.; Van Gerpen, J. Trans ASAE 2003, 46, 945–954.(39) Bligh, E. G.; Dyer,W. J.Can. J. Biochem. Physiol. 1959, 37, 911–

917.(40) Dionisi, F.; Golay, P. A.; Elli, M.; Fay, L. B. Lipids 1999, 34,

1107.

3177

Energy Fuels 2010, 24, 3173–3178 : DOI:10.1021/ef9015872 Vicente et al.

The fatty acid profile for the FAMEs obtained fromM. circinelloideswas compared to those produced for rapeseed,sunflower, palm,41 and soy41 oils (Table 4), which are the mostcommonly used raw materials by the biodiesel industry inEurope and theUnitedStates.Microbial oils usually differ frommost vegetable oils in being quite rich in polyunsaturatedfatty acids.8 However, the content of these fatty acids in the

biodiesel obtained from M. circinelloides was within theEuropean Union specifications because the specified limit(1%) only includes polyunsaturated fatty acids with four ormore double bonds, which are absent in M. circinelloides-derived biodiesel. FAMEs from M. circinelloides contained12.7 and 22.5% of linoleic (two double bonds) and linolenic(three double bonds) acids, respectively, which would havelow oxidative stability. In fact, the linolenic acid methyl estercontent in the M. circinelloides-derived biodiesel was abovethe specified limit, 12%, in the European standards. On the

Figure 2. Schematic diagram of the process for biodiesel production from fungal biomass.

Table 3. Quality Control of M. circinelloides-Derived Biodiesela

catalyst

property BF3 H2SO4 HCl EU standard EN 14214 U.S. standard ASTM D6751

monoglyceride content (wt %) nd nd nd 0.8 maximum nsdiglyceride content (wt %) nd nd nd 0.2 maximum nstriglyceride content (wt %) nd nd nd 0.2 maximum nsfree glycerol (wt %) 0.0020 0.0032 0.0030 0.02 maximum 0.02 maximumtotal glycerol (wt %) 0.0020 0.0032 0.0030 0.25 maximum 0.24 maximumacid value (mg of KOH/g) nd 0.40 nd 0.5 maximum 0.5 maximumnon-saponifiable lipids (wt %) nd nd nd ns nspolar lipids (wt %) 0.8 0.8 0.9 ns nsester content (wt %) 99.2 99.0 99.1 96.5 minimum ns

a nd, not detected; ns, not a specified limit.

Table 4. Fatty Acid Composition in Biodiesel from M. circinelloides, Rapeseed, Sunflower, Palm, and Soy Oils

content (wt %)

fatty acid M. circinelloides oil rapeseed oil sunflower oil palm oil41 soy oil41

lauric acid 12:0 nd nd nd 0.1 ndmyristic acid 14:0 1.6 0.1 nd 0.7 ndmyristoleic acid 14:1 0.6 nd nd nd ndpentadecanoic acid 15:0 2.5 nd nd nd ndpalmı́tic acid 16:0 20.7 5.0 6.3 36.7 11.3palmitoleic acid 16:1 1.1 nd 0.2 0.1 0.1stearic acid 18:0 7.0 1.6 2.2 6.6 3.6oleic acid 18:1 28.0 36.3 20.6 46.1 24.9linoleic acid 18:2 12.7 19.8 52.8 8.6 53.0linolenic acid 18:3 22.5a 7.8b 3.5b 0.3b 6.1b

arachidic acid 20:0 0.3 0.1 1.6 0.4 0.3gadoleic acid 20:1 nd 9.1 0.3 0.2 0.3behenic acid 22:0 0.4 nd 7.2 0.1 nderucic acid 22:1 0.07 20.2 5.1 nd 0.3lignoceric acid 24:0 1.2 nd 0.2 0.1 0.1nervonic acid 24:1 nd nd nd nd ndother 1.3 nd nd nd ndiodine value (g of I2/100 g) 106.0 107.7 122.4 55.6 129.7

aThe γ-linolenic acid isomer was obtained. bThe R-linolenic acid isomer was obtained.

(41) Ramos,M. J.; Fern�andez,C.M.;Casas,A.; Rodrı́guez, L.; P�erez,A. Bioresour. Technol. 2008, 100, 261–268.

3178

Energy Fuels 2010, 24, 3173–3178 : DOI:10.1021/ef9015872 Vicente et al.

other hand, the high degree of unsaturation inherent tomethyl esters from these fatty acids would evidence excellentfuel properties at low temperatures, which is an advantage inwinter operation.42 Moreover, all of these fatty acids arecommon in industrial vegetable oils, and in particular, sun-flower and soy oils are also very rich in polyunsaturated fattyacids. Thus, the calculated iodine value, which is a measureof the total unsaturation level, for the M. circinelloides-derived biodiesel (106.0 mg of I2/g) was far below the speci-fied limit (120 mg of I2/g) in the European Union standardsand also met the United States standards because thesespecifications do not include the iodine value as a qualityparameter. In comparison to the vegetable oils, the iodinevalue was very similar to the one obtained in biodiesel fromrapeseed oil (107.7 mg of I2/g), which is the preferred rawmaterial for biodiesel production in Europe.

4. Conclusions

The results shown here indicate that M. circinelloidesbiomass from submerged cultures may be a suitable feedstockfor biodiesel production. Moreover, the analyzed propertiesof the M. circinelloides-derived biodiesel fulfilled the specifi-

cations established by the current existing standards, ASTMD6751 in the United States and EN 14213 and 14214 inthe European Union. In addition, efficient biodiesel produc-tion by direct transformation of fungal biomass without lipidextraction is technically feasible in M. circinelloides, whichrepresents a starting point for developing this process on anindustrial scale. However, biodiesel yields should be increasedto make the industrial process economical, which could beattained by the genetic manipulation of this fungus. In thissense, efforts are nowdedicated tooverexpress genes that codefor enzymes postulated to be rate-limiting steps for fattyacid biosynthesis in oleaginous fungi.26 Other strategies arefocused on the generation of strains with enhanced ability touse crop residues or industrial byproduct, avoiding competi-tion with the food supply, with low linolenic acid levels oroverexpressing genes involved in saponifiable lipid biosynthe-sis. Particularly interesting is the generation of strains withlow free fatty acid levels because they could be used forbiodiesel production by using a base-catalyzed technology,which is the common way to produce biodiesel on an indus-trial scale.

Acknowledgment. We thank J. A. Madrid for technical assis-tance. This work was funded by the D. G. de Investigaci�on yPolı́tica Cientı́fica (Comunidad Aut�onoma de la Regi�on deMurcia, Spain), Project BIO-BMC 07/01-0005.

(42) Vicente,G.;Martı́nez,M.; Aracil, J.Bioresour. Technol. 2004, 92,297–305.

Experimental Investigations into the Insecticidal, Fungicidal, and BactericidalProperties of Pyrolysis Bio-oil from Tobacco Leaves Using a Fluidized BedPilot Plant

Christina J. Booker,†,‡ Rohan Bedmutha,†,§ Tiffany Vogel,†,‡ Alex Gloor,†,‡ Ran Xu,†,§

Lorenzo Ferrante,†,§ Ken K.-C. Yeung,†,‡ Ian M. Scott,| Kenneth L. Conn,| Franco Berruti,†,§ andCedric Briens*,†,§

Institute for Chemicals and Fuels from AlternatiVe Resources (ICFAR), 22312 Wonderland Road North,RR#3, Ilderton, Ontario N0M 2A0, Canada, Faculty of Science, The UniVersity of Western Ontario,1151 Richmond Street, London, Ontario N6A 5B9, Canada, Faculty of Engineering, The UniVersity ofWestern Ontario, 1151 Richmond Street, London, Ontario N6A 5B9, Canada, and Agriculture andAgri-Food Canada, 1391, Sandford Street, London, Ontario N5V 4T3, Canada

Tobacco bio-oil, gases, and char were produced through pyrolysis of tobacco leaves using a fluidized bedpilot plant under varying temperature (350, 400, 450, 500, 550, and 600 °C) and residence time (5, 10, and17 s) conditions. The optimized condition for the production of bio-oil was found to be at 500 °C at a vaporresidence time of 5 s, giving a bio-oil yield of 43.4%. The Colorado Potato Beetle (CPB) Leptinotarsadecemlineata L. (Coleoptera: Chrysomelidae), a destructive pest toward potato crops, and three microorganisms(Streptomyces scabies, ClaVibacter michiganensis, and Pythium ultimum), all problematic in Canadianagriculture, were strongly affected by tobacco bio-oil generated at all pyrolysis temperatures. Nicotine-freefractions of the tobacco bio-oil were prepared through liquid-liquid extraction, and high mortality rates forthe CPB and inhibited growth for the microorganisms were still observed. A potential pesticide from tobaccobio-oil adds value to the biomass as well as the pyrolysis process.

1. Introduction

Pyrolysis is one of the thermo-chemical processes that is usedextensively worldwide to convert biomass into liquid bio-oil,char, and gases. This process is carried out in the absence ofoxygen.1 However, the pyrolysis oil normally contains a highproportion of oxygenates, reflecting the oxygen content of theoriginal substrates. With the current focus on environmentallyfriendly energy prospects and renewable energy resources,significant research is being directed toward bio-oils. Bio-oil isconsidered a CO2 neutral alternative to fossil fuels with lowemissions of the undesirable components SO2, NOx, and soot.2

Despite these advantages, bio-oil has several undesirable proper-ties as a fuel, including high viscosity, low heating value, poorvolatility, and coking. Refining bio-oil to a satisfactory levelfor commercial use has been performed, but currently uses toomuch energy and occurs at too high a cost to be economicallyviable.3

An additional, potentially lucrative prospect for bio-oil is asa source for valuable chemicals. These chemicals could be foundin the original biomass, such as nicotine in tobacco bio-oil, orcould be created during the pyrolysis process, such as phenolsor new chemicals yet to be identified. One of the many potentialapplications of these chemicals is as a pesticide. The search foreffective and safe pesticides is a continuing challenge as speciesquickly adapt to most pesticides that are applied.

In this Article, tobacco bio-oil is generated through pyrolysisunder a wide range of operating conditions and analyzed forpesticide properties toward a variety of species of concern in

Canada. One of the reasons this biomass was selected for anal-ysis is that tobacco farmers across the world, and in particularin Canada, are suffering as demand for their crop continues todecline. It is well-known that smoking tobacco has a significant,negative impact on human health. Transitioning out of tobaccofarming, however, is difficult due to the specified nature of theequipment used, and therefore many farmers are left with excesscrop every year, which currently goes to waste. Thus, findingalternative, healthy, high value applications to this highlyabundant product is an important research area. Already, tobaccobiomass is being investigated for unique, high value applications,such as for medical or industrial proteins,4-6 and in the case ofthis research, as a natural pesticide. Because tobacco’s pesticideproperties are well-known, converting tobacco leaves to naturalpesticides in the form of bio-oil could provide additional incometo farmers.

Tobacco biomass has been characterized,7-9 but very limitedwork has been published on the pyrolysis of tobacco for theproduction of bio-oil. One study concentrated on the productionof fuel gases but did not perform liquid analysis,10 while anotherstudy performed liquid analysis but failed to analyze the bio-oil for nicotine.11

The potential pesticide activity of bio-oil is an excitingresearch area that has yet to be fully explored. Recently, bio-oil has been studied for its wood preservative qualities12 andspecifically for its antifungal properties.13 Two species of fungiwere tested and found to have inhibited growth patterns in thepresence to bio-oil from wood biomass. In contrast, this researchArticle investigates the pesticide characteristics of bio-oil fromtobacco biomass, not only for antifungal activity, but also forantibacterial and insecticidal activity. The pyrolysis of thistobacco biomass is also investigated.

* To whom correspondence should be addressed. E-mail: [email protected].

† ICFAR.‡ Faculty of Science, The University of Western Ontario.§ Faculty of Engineering, The University of Western Ontario.| Agriculture and Agri-Food Canada.

Ind. Eng. Chem. Res. 2010, 49, 10074–1007910074

10.1021/ie100329z 2010 American Chemical SocietyPublished on Web 09/14/2010

2. Experimental Section

2.1. Materials. Finely ground tobacco leaves were providedby Agriculture and Agri-Food Canada, London, ON. Tobaccoleaves were obtained from tobacco crops in 2006 and dried at60 °C. Dried tobacco leaves were then ground using a blender/mixing mill and sieved. The Sauter mean diameter of the tobaccoparticles used for pyrolysis was 60 µm.

2.2. Methods. 2.2.1. Pilot Plant Design for Pyrolysis. Allpyrolysis experiments were carried out using a fluidized bedpilot plant14 (Figure S1, Supporting Information). The heart ofthe plant was an atmospheric fluid bed reactor, 0.078 m indiameter, with a 0.52 m long cylindrical section, and equippedwith an expanded section made up of a 0.065 m long truncatedcone with an upper diameter of 0.168 m, topped by a second,0.124 m long, cylindrical section. The total volume of thisconfiguration was 6.09 × 10-3 m3. This assembly provided thelowest vapor residence time (5 s). Two different freeboardextensions were used to increase the vapor residence time to10 and 17 s. A filter capable of withstanding high temperatureswas installed at the gas exit of each of the extensions. Eachfilter was made up of a perforated pipe connected to the gasexit covered by a fiberglass pad and wrapped inside a finestainless steel mesh cover. The resulting filter was, in all cases,0.076 m in diameter and 0.178 m long. Although not ideal, thesehot filters have been used in the initial phase of the project withthe objective of avoiding the use of a hot cyclone for the charseparation, which would be impossible to properly size due tothe variety of physical characteristics of the chars expected fromthe different feedstocks.

The fluidizing nitrogen was injected through a perforated copperdistributor plate with 33 holes, 0.5 mm in diameter, equally spacedacross the cross section. The reactor was equipped with 18thermowells for temperature measurements and control (type Kthermocouples).

An innovative pulsating automatic feeder was used forbiomass injection to the reactor. It quickly dispersed the injectedbiomass into the core of the fluidized bed.

2.2.2. Bio-oil Production. Tobacco, when injected into thereactor, produced vapors that exited at the top of the reactorthrough the hot filter section and flowed into three condensersin series through a line traced with Raychem Chemelex heatingcable to avoid early, undesirable condensation (as shown inFigure S1). Persistent aerosols were then separated in a cylin-drical demister packed with cotton wool. The demister wasweighed before and after the experiment. The exact yield oftobacco bio-oil was obtained from the mass of oil collected inthe three condensers and the demister.

Pyrolysis was initially carried out at six different temperaturesfrom 350 to 600 °C and at three different residence times (5,10, and 17 s). Each test was conducted with 700 g of tobaccoleaves. Fluidizing and atomizing nitrogen volumetric flow rateswere precisely controlled using “Mass Trak” flow-meters fromSierra Instruments Inc., to keep the nominal vapor residencetime constant at all temperatures. Tobacco bio-oils produced atall these temperatures separated into two separate phases, anorganic and an aqueous one.

Pyrolysis of tobacco leaves was subsequently carried outunder the best reactor conditions for high bio-oil yield (discussedin Results and Discussion section and found to be at atemperature of 500 °C and a vapor residence time of 5 s) todetermine the accurate liquid, gas, and char yields.

2.2.3. Characterization of Product Gases. Gases weresampled in plastic bags at three different time intervals. Tomeasure the product gas composition, a Hewlett-Packard 5890

series II gas chromatograph (GC) was used. A RESTEK ShinCarbon ST (micro packed), 2 m length column with 1 mm i.d.and 1.58 mm o.d., was used to separate the gas mixture. Athermal conductive detector (TCD) was used to detect thecomposition of the gas mixture, which consisted of N2, H2, CO,CO2, and CH4. To measure product gas yields accurately, N2

was selected as an internal standard gas. Argon was selected asthe GC carrier gas. A standard gas mixture with a fixed com-position of H2, CO, CO2, and CH4 was used to calibrate thesystem. The injector was maintained at 150 °C, and the TCDwas maintained at 275 °C. A gas sample volume of 0.5 µL wasinjected with a 100 µL Hamilton syringe. Upon injection, theoven temperature was held at 35 °C for 180 s, then increasedat 10 °C/min to 150 °C, and finally increased at 20 °C/min to250 °C. The temperature was then held constant at 250 °C for330 s.

2.2.4. Characterization of Char. The differential pressuredrop across the fluidized bed was measured at minimumfluidization conditions before and after each experiment. Theincrease in the reading of the differential pressure drop wasproportional to the increase in bed weight. This system wascalibrated for very accurate measurement of the char yields.

2.2.5. Characterization of Bio-oil. The bio-oil was charac-terized through GC-MS analysis of the various fractionsexamined for biological assays (see below). A HP 6890 Seriesgas chromatography system with a mass selective detector wasused to analyze the bio-oil fractions. All experiments wereperformed on an HP-5MS, 30 m column obtained from AgilentTechnologies with an i.d. of 0.25 mm and a film of 0.25 µm.The injector temperature and auxiliary temperature were main-tained at 300 °C. The oven temperature began at 60 °C for 2min, and then increased at 10 °C/min to 280 °C and was heldfor 6 min. A threshold of 150 was used, with a mass to chargescan range of 50-300 at a rate of 2.98 scans/s.

2.2.6. Bio-oil Pesticide Characterization. Pesticide activitytests with the bio-oil were performed on a variety of problematicspecies of microorganisms and one insect. All tobacco bio-oilsamples used for the biological tests were produced at a vaporresidence time of 5 s and at different pyrolysis temperatures,as specified for each assay.

2.2.6.1. Bio-oil Sample Preparation for Pesticide Analysis.To initially determine which microorganisms were negativelyaffected by the tobacco bio-oil, a cocktail of naturally separated,organic phases and a cocktail of the aqueous phases of the bio-oils produced from 350 to 600 °C were prepared in acetone(375 mg/mL, one solution of all pyrolysis temperatures). Bio-oil samples from each pyrolysis temperature were then preparedseparately in acetone (375 mg/mL, one solution for eachpyrolysis temperature). Raw tobacco bio-oil at each pyrolysistemperature was used for the CPB tests.

Two different liquid-liquid extraction techniques were usedto generate nicotine-free and nicotine-containing fractions ofthe tobacco bio-oil. One method was used for the microorganismassays and generated six unique fractions (also analyzed throughGC-MS), while the other method was used for the insect assaysand generated two distinct fractions. The reason for the twomethods was that two separate researchers performed theserespective tests. Even so, the end result successfully allowedfor nicotine-free fractions to be tested on both the microorganismand the CPB.

The fractionation method used for the microorganism tests,which generated six unique fractions, is illustrated in Figure 1.The organic phase of the tobacco bio-oil pyrolyzed at 450 °Cwas dissolved in ether at a concentration of 175 mg/mL. This

Ind. Eng. Chem. Res., Vol. 49, No. 20, 2010 10075

fraction was sterile filtered with a 2.5 cm diameter, 45 µm poresize, syringe filter with a nylon membrane (Whatman, NJ)(Fraction Z). The remaining residue was dissolved in acetone(approximately 102 mg/mL) and was also sterile-filtered, givinga very dark brown solution (Fraction I). Fraction Z was thenfractionated into its aqueous (Fraction A) and organic (FractionB) components with a water/ether extraction. An additionalwater/ether separation was then performed with Fraction Bwhere the water phase was acidified with HCl to a pH of 4-5.This step caused some components, such as the compoundnicotine, to become charged and move into the aqueous phase.An organic, ether phase (Fraction C) and a charged, aqueousphase were generated. The acidic phase was then adjusted topH 9 (to move the majority of nicotine back into an organicphase) and a final aqueous/ether extraction made an organicphase (Fraction D) and an aqueous phase (Fraction E). Dilutionfactors were calculated for each fraction, and the volume ofsample used for the biological assays was appropriately adjusted.Each fraction was analyzed using GC-MS (Figure S2).

To generate a nicotine-free and a nicotine-containing fractionfor the insect tests, liquid-liquid extraction was performed withdiethyl ether and dichloromethane (DCM) (Figure 2). Theprocedure outlined by Oasmaa et al.15 was used as it closelymatched past literature methods for nicotine extraction fromtobacco plants.16,17 A bio-oil mixture from all pyrolysis tem-peratures (15-20 g) was first passed through a filter paper(Whatman’s #4) to remove the solid lignin residue. This residuewas washed with two, 5 mL portions of diethyl ether followedby two, 5 mL portions of DCM. The filtrate was then extractedwith 20-30 mL of diethyl ether followed by 20-30 mL ofDCM. All organic phases were combined, and the solvent wasevaporated using a rotary evaporator (BUCHI R-114). The

organic fraction recovered was a moderately viscous brown oil,quite similar to the bio-oil itself. The aqueous fraction wasorange and had low viscosity.

2.2.6.2. Biological Assays for Pesticide Activity. 2.2.6.2.1. Mi-croorganism Assays. The disk diffusion assay was used to test11 fungi and 4 bacteria for growth inhibition in the presence ofthe tobacco bio-oil samples. All species are problematicmicroorganisms in Canada. See Table S1 for the list of species,their source, and the type of media on which they weremaintained. Samples and control solutions were added to sterile,6 mm diameter filter paper disks and allowed to air-dry beforebeing placed onto freshly inoculated plates. For bacteria tests,the plates were inoculated by streaking the entire surface withfreshly grown bacteria to generate a lawn of growth. One orthree paper disks were placed into the center of the plate or ina triangular formation on the plate, depending upon the ex-periment. For fungi tests, a plug of a fresh culture was addedabout 1 cm away from the disks on a fresh plate. After the plateswere incubated at 24 °C for 3 days, the results were recorded.A region of no growth around the disk indicated inhibition (witha minimum measurement of inhibition being 6 mm, the diameterof the disk). Triplicate experiments were performed.

2.2.6.2.2. Insect Assays. These tests were carried out by theleaf disk application, similar to the procedure outlined bySengonca.18 Bio-oil fractions and control solutions were spreadon both sides of a potato leaf disk with a cotton-tippedapplicator. Three leaves were tested for each fraction; however,most tests were repeated on multiple dates to ensure accuracy.The potato plants (var. Cal White) were grown on site at theSouthern Crop Protection and Food Research Centre (SCPFRC),Agriculture and Agri-Food Canada, London, Ontario, with theleaves cut to a diameter of 4 cm. The leaves were allowed todry after sample application. After drying, the leaves weretransferred to a Gelman Petri dish. Five, second instar insecticidesusceptible strain Colorado Potato Beetle (CPB) larvae rearedat SCPFRC were then transferred to the leaf. Mortality rateswere recorded after 24 and 48 h intervals. Adjusted percentmortality values are reported, which take into account the naturalmortality levels of the CPB in the control treatments. Controltreatments involved simply placing the beetles on leaf diskswithout any oil present. If a specific test involved dilution ofthe bio-oil, the control leaf disks were coated with the solventused.

3. Results and Discussion

3.1. Tobacco Pyrolysis. The effects of pyrolysis temperatures(350-600 °C) and residence times (5, 10, and 17 s) on the liquidyield are as shown in Figure 3. Tobacco bio-oil yields were astrong function of temperature and residence time. The greatestyield peaked at 500 °C for all residence times. It could also beobserved that bio-oil yield increased as the residence timedecreased, for all temperatures. Comparable results were foundwhen this reactor was used to pyrolyze grape seeds and skins,for at a 5 s vapor residence time, the optimum pyrolysis temper-ature was also found to be 500 °C.14

As shown in Table 1, for a residence time of 5 s and a reactiontemperature of 500 °C, the bio-oil yield was the highest (43.4%),followed by the char yield (29.4%) and the gas yield (22.4%).The mass balance on the pyrolysis products was close to 95%,which was within the margin of error. Calculations showed thatthe heat of combustion of the gases produced was 508 J/g ofbiomass fed. It was assumed that the water produced by com-bustion was condensed. The heat of combustion value fortobacco was on the lower side as compared to other feedstocks,

Figure 1. Bio-oil fractionation scheme for microorganism assay testing andGC-MS analysis. Shaded boxes indicate fractions tested in microorganismassays.

Figure 2. Extraction scheme for nicotine-free tobacco bio-oil fractions forinsect assays.

10076 Ind. Eng. Chem. Res., Vol. 49, No. 20, 2010

such as coffee grounds and pinewood, pyrolyzed in the samepilot plant at the same temperature.

The higher liquid yield at lower residence time can be at-tributed to the fact that lower residence time minimizessecondary reactions19 such as thermal cracking, repolymeriza-tion, and recondensation to maximize liquid yields. It is alsovery well-known that higher temperature favors gasification(higher gas yields and lower liquid and char yields). Thus, theresults obtained are consistent with the existing literature onvarious other biomass feedstocks.20

3.2. Bio-oil Activity toward Pest Species. 3.2.1. InitialPesticide Discovery. Initial tests with tobacco bio-oil demon-strated clear pesticide activity toward a selection of microorgan-ism species and the Colorado Potato Beetle.

To determine which microorganism species were inhibitedby the tobacco bio-oil, naturally separated organic (375 mg/mL organic phase in acetone) and aqueous (used directly withoutdilution) phase mixtures from all pyrolysis temperatures (350-550°C) were assayed against 11 fungi and 4 bacteria (Table S1).These species were selected for analysis because of theirdestructive properties toward agriculture in Canada and wereavailable for testing through Agriculture and Agri-Food Canada.No inhibition was found from the aqueous phases of the tobaccobio-oil. In contrast, the organic phases of the tobacco bio-oilshowed clear inhibition for two bacteria, Streptomyces scabies(S. scabies) and ClaVibacter michiganensis sub. sp. michigan-ensis (C. michiganensis), and one fungus, Pythium ultimum (P.ultimum).

Pythium ultimum is a fungus that affects plants as a seedlingdamping-off disease.21 Plants affected include eggplant, pepper,lettuce, tomato, and cucumber. ClaVibacter michiganensis killsyoung plants and deforms fruits, primarily tomatoes.22 Strep-tomyces scabies is a common potato scab disease that infectspotatoes and makes them unmarketable.23 Finding inhibition forS. scabies is particularly exciting because, currently, no safepesticide exists on the market that can control this widespreaddisease.

This discovery of tobacco bio-oil affecting only three mi-croorganism species (and not the remaining 12) is particularly

interesting. This selective inhibition suggests that the activecomponents in the bio-oil are not destructive to all living things,which is an important quality for a potential pesticide.

The Colorado Potato Beetle was also found to be negativelyaffected by the presence of the tobacco bio-oil. Early testsconfirmed high mortality rates for the CPB, and further experimentswere performed to investigate one of the key pyrolysis parameters:the pyrolysis temperature.

3.2.2. Investigation into the Effect of Pyrolysis Tempera-ture on Pesticide Activity. Bio-oil produced at each pyrolysistemperature successfully inhibited the growth of each of thethree microorganisms (Figure 4).

As the pyrolysis temperature increased to 550 °C, the activityof the bio-oil seemed to decrease. This could be due to the activecomponents being cracked into smaller, inactive componentsat this high temperature. At 450 °C, the greatest inhibition wasobserved for all three species. For this reason, as well as thefact that this temperature was close to 500 °C (the pyrolysistemperature with the greatest percent yield of bio-oil), the bio-oil pyrolized at 450 °C was selected for continued investigation.It is important to note that, although these bio-oil samples wereprepared to a specific concentration, the observed variations inthe activity with pyrolysis temperature could be affected by theamount of water in each bio-oil sample. The water was notremoved from the sample to avoid removing other, potentiallyimportant chemicals in the process. Nevertheless, each bio-oilsample was found to successfully inhibit the growth of eachspecies.

Similar to the microorganism pattern of inhibition, the CPBwas found to be strongly affected by bio-oil produced at allpyrolysis temperatures (Figure 5). The potency of each bio-oilwas quite strong given the high mortality levels seen. The 48 hresults show that 100% of the beetles tested at each pyrolysistemperature died when in the presence of the tobacco bio-oil.Although the 24 h results seem to demonstrate some changesin toxicity with pyrolysis temperature, these changes are onlyminor.

It was possible that the toxicity effect of the bio-oils towardthe CPB was caused solely by the high quantities of nicotinein the bio-oil. Nicotine is a moderately effective insecticideagainst the CPB with an LD50 of 61 µg per CPB.24 Sufficientquantities of nicotine could be present in the bio-oil to accountfor the observed activity. Thus, the bio-oil was separated intonicotine-free and nicotine-containing fractions to determine theeffect of nicotine in the observed activity.

Figure 3. Effect of temperature and residence time on the liquid bio-oilyield. For experimental details, see Methods section.

Table 1. Pyrolysis Product Split at a Vapor Residence Time of 5 sand Pyrolysis Temperature of 500 °C

liquid yield (wt %) gas yield (wt %) char yield (wt %)

43.4 22.4 29.4H2 0.7CO 27CH4 2.8CO2 69.5

Figure 4. Effect of pyrolysis temperature on the diameter of inhibition forthe three affected microorganism species. Error bars indicate ( standarddeviation (σ) of replicate measurements within an experiment (total length2σ).

Ind. Eng. Chem. Res., Vol. 49, No. 20, 2010 10077

3.2.3. Investigation into the Activity of the Nicotine-FreeFractions of Tobacco Bio-oil. The fractionation scheme shownin Figure 1 was used to generate the six fractions tested on thethree microorganisms, as shown in Figure 6. As expected,Fraction Z (the initial fraction) had high activity toward themicroorganisms. However, high levels of nicotine were alsofound in Fraction Z (Figure S1), so much so that few otherchemicals could be observed in the chromatograms of thisfraction.

The fractionation scheme successfully generated a nicotine-free fraction, Fraction C, which was confirmed by the absenceof a nicotine peak in the GC-MS data. This fraction was alsostrongly active (as shown in Figure 6). Phenol and a variety ofits derivatives were found to be in high concentration in thisfraction. Although phenolic compounds are known to havepesticide properties,25,26 10 of the most abundant compoundsin this fraction were quantitatively tested by chemical standards,and it was found that these most abundant phenolic compoundswere not present in high enough concentrations to be responsiblefor the observed activity.27

Fraction D, which contains nicotine, was also found to beactive. However, when nicotine standards were tested to matchand even double the concentration of nicotine found in FractionD, no inhibition was observed. It is interesting to note thatnicotine is the most abundant and almost the only peak detectedby GC-MS in this fraction. Therefore, the active componentsin Fraction D cannot be detected by our GC-MS analysismethod. These active components either have a higher boiling

point than 280 °C (the highest temperature in our GC program)or cannot be detected by an electron impact MS detector.

A nicotine-free fraction was also found to be active in theCPB assays. The organic fraction showed greater activity overthe aqueous (nicotine-free) fraction. After 24 h of testing, theorganic fraction obtained 100% mortality rates, while after 48 hof testing, the aqueous fraction obtained a maximum of 80%mortality for the CPB (Figure 7). It is also worth noting that,although the aqueous phase did not result in 100% mortality tothe CPB, application of the aqueous phase to the leaf resultedin a greatly reduced appetite for the beetle. Using the aqueousphase at 2% concentration or higher, the beetles would eat littleto none of the leaf. Studies have shown that 24 h starvation ofthe CPB does not prove fatal; however, starvation does causeincreased susceptibility to applied insecticides.28 Whether or notthe chemical agent that causes mortality is the same as the agentthat is causing starvation is not known, but the starvation isaiding the insecticidal activity of the aqueous fraction.

Further investigation into the nicotine content of the organicfraction was performed. Nicotine standards were tested at theconcentration found in the organic fraction. Dilution tests ofthe organic phase and the equivalent nicotine standard demon-strated that the potency of the samples was the same whenmeasured at 48 h. However, the 24 h results demonstrated thatthe organic fraction worked faster at causing death in the CPBthan the nicotine standards. This indicates that additional, non-nicotine components are acting in the organic fraction.

The assays performed on the CPB and the three microorgan-isms clearly indicate that tobacco bio-oil contains potent, non-nicotine components with insecticidal and antibiotic activity.Multiple, active components must be present in the tobacco bio-oil as liquid-liquid extraction produced multiple, active frac-tions. Some of these active compounds cannot be detected byGC-MS.

4. Conclusions

Pyrolysis experiments demonstrated that the liquid bio-oilyield was a strong function of temperature and vapor residencetime. The maximum bio-oil yield was found at a reactor tem-perature of 500 °C and the lowest residence time, 5 s.

Bio-oil was found to have valuable pesticide characteris-tics toward three problematic microorganisms as well as theColorado Potato Beetle, a major agricultural pest. Bio-oil pro-duced at all pyrolysis temperatures was effective at inhibiting

Figure 5. Effect of pyrolysis temperature on the adjusted percent mortalityof the Colorado Potato Beetle at 24 and 48 h.

Figure 6. Measured diameters of inhibition for three microorganisms bythe six tobacco bio-oil fractions (see Figure 1 for fractionation scheme)after 3 days of growth. Fraction C is nicotine-free. Error bars indicate (standard deviation (σ) of replicate measurements within an experiment (totallength 2σ).

Figure 7. Dilution tests comparing the aqueous (nicotine-free) fraction andthe organic (nicotine containing) fraction prepared as illustrated in Figure2. Results for both fractions were recorded at 24 and 48 h.

10078 Ind. Eng. Chem. Res., Vol. 49, No. 20, 2010

the growth or causing mortality in the microorganisms andColorado Potato Beetle, respectively.

Nicotine was found to be active toward the Colorado PotatoBeetle, but had no effect on the microorganisms. Nicotine-freefractions of tobacco bio-oil were found to be active toward theColorado Potato Beetle and three microorganisms. Multiplecomponents are likely responsible for this activity. Thesecomponents were not lethal to all of the microorganisms thatwere examined, demonstrating that these chemicals may onlybe toxic to selective species, which is a desirable quality in apotential pesticide.

As the demand for tobacco is decreasing, the search for othervaluable products from this resource is increasing. A naturalpesticide that targets problematic species is a very valuable find.Further investigation into the active components and the poten-tial applicability of using tobacco bio-oil as a natural pesticidewill continue.

Acknowledgment

We wish to express our gratitude to the Ontario Centres ofExcellence (OCE), the Natural Sciences and Research Councilof Canada (NSERC), Agri-Therm Canada, the Institute forChemicals and Fuels from Alternative Resources (ICFAR),Agriculture and Agri-Food Canada (through ABIN), and theUniversity of Western Ontario for their generous support of thisresearch.

Supporting Information Available: Additional figures andtable. This material is available free of charge via the Internetat http://pubs.acs.org.

Literature Cited

(1) Bridgewater, A. V.; Czernik, S.; Piskorz, J. An overview of fastpyrolysis. In Progress in Thermochemical Biomass ConVersion; Bridge-water, A. V., Ed.; Blackwell Science Ltd.: Oxford, 2001; pp 977-997.

(2) Zhang, Q.; Chang, J.; Wang, T. J.; Xu, Y. Review of biomasspyrolysis oil properties and upgrading research. Energy ConVers. Manage.2007, 48, 87–92.

(3) Czernik, S.; Bridgewater, A. V. Overview of applications of biomassfast pyrolysis oil. Energy Fuels 2004, 18, 590–598.

(4) Joensuu, J. J.; Brown, K. D.; Conley, A. J.; Clavijo, A.; Menassa,R.; Brandle, J. E. Expression and purification of an anti-Foot-and-mouthdisease virus single chain variable antibody fragment in tobacco plants.Transgenic Res. 2009, 18, 685–696.

(5) Menassa, R.; Du, C. G.; Yin, Z. Q.; Ma, S. W.; Poussier, P.; Brandle,J.; Jevnikar, A. M. Therapeutic effectiveness of orally administeredtransgenic low-alkaloid tobacco expressing human interleukin-10 in a mousemodel of colitis. Plant Biotechnol. J. 2007, 5, 50–59.

(6) Ma, S. W.; Huang, Y.; Davis, A.; Yin, Z. Q.; Mi, Q. S.; Menassa,R.; Brandle, J. E.; Jevnikar, A. M. Production of biologically active humaninterleukin-4 in transgenic tobacco and potato. Plant Biotechnol. J. 2005,3, 309–318.

(7) Sheng, L. Q.; Ding, L.; Tong, H. W.; Yong, G. P.; Zhou, X. Z.;Liu, S. M. Determination of nicotine-related alkaloids in tobacco andcigarette smoke by GC-FID. Chromatographia 2005, 62, 63–68.

(8) Shen, J. C.; Shao, X. G. Determination of tobacco alkaloids by gaschromatography-mass spectrometry using cloud point extraction as apreconcentration step. Anal. Chim. Acta 2006, 561, 83–87.

(9) Cai, J. B.; Liu, B. Z.; Lin, P.; Su, Q. D. Fast analysis of nicotinerelated alkaloids in tobacco and cigarette smoke by megabore capillary gaschromatography. J. Chromatogr., A 2003, 1017, 187–193.

(10) Encinar, J. M.; Beltran, F. J.; Gonzalez, J. F.; Moreno, M. J.Pyrolysis of maize, sunflower, grape and tobacco residues. J. Chem. Technol.Biotechnol. 1997, 70, 400–410.

(11) Demirbas, A. Analysis of liquid products from biomass via flashpyrolysis. Energy Sources 2002, 24, 337–345.

(12) Mourant, D.; Riedl, B.; Rodrigue, D.; Yang, D. Q.; Roy, C. Phenol-formaldehyde-pyrolytic oil resins for wood preservation: A rheological study.J. Appl. Polym. Sci. 2007, 106, 1087–1094.

(13) Mohan, D.; Shi, J.; Nicholas, D. D.; Pittman, C. U.; Steele, P. H.;Cooper, J. E. Fungicidal values of bio-oils and their lignin-rich fractionsobtained from wood/bark fast pyrolysis. Chemosphere 2008, 71, 456–465.

(14) Xu, R.; Ferrante, L.; Briens, C.; Berruti, F. Flash pyrolysis of graperesidues into biofuel in a bubbling fluid bed. J. Anal. Appl. Pyrolysis 2009,86, 58–65.

(15) Oasmaa, A.; Kuoppala, E. Fast pyrolysis of forestry residue. 3.Storage stability of liquid fuel. Energy Fuels 2003, 17, 1075–1084.

(16) Sheen, S. J. Detection of nicotine in foods and plant materials. J.Food Sci. 1988, 53, 1572–1573.

(17) Shin, H. S.; Kim, J. G.; Shin, Y. J.; Jee, S. H. Sensitive and simplemethod for the determination of nicotine and cotinine in human urine, plasmaand saliva by gas chromatography-mass spectrometry. J. Chromatogr., B2002, 769, 177–183.

(18) Sengonca, C.; Liu, B.; Zhu, Y. J. Pestic. Sci. 2005, 79, 3–8.(19) Gercel, H. F.; Putun, E. Fast pyrolysis of sunflower-pressed bagasse:

Effects of sweeping gas flow rate. Energy Sources 2002, 24, 451–460.(20) Bridgwater, A. V. Biomass fast pyrolysis. Therm. Sci. 2004, 8, 21–

49.(21) Abbasi, P. A.; Lazarovits, G. Effects of AG3 phosphonate formula-

tions on incidence and severity of Pythium damping-off of cucumberseedlings under growth room, microplot, and field conditions. Can. J. PlantPathol. 2005, 27, 420–429.

(22) Gleason, M. L.; Braun, E. J.; Carlton, W. M.; Peterson, R. H.Survival and dissemination of ClaVibacter michiganensis sub. sp. michagen-ensis in tomatoes. Phytopathology 1991, 81, 1519–1523.

(23) Wang, A. X.; Lazarovits, G. Role of seed tubers in the spread ofplant pathogenic Streptomyces and initiating potato common scab disease.Am. J. Potato Res. 2005, 82, 221–230.

(24) Mota-Sanchez, D.; Hollingworth, R. M.; Grafius, E. J.; Moyer, D. D.Resistance and cross-resistance to neonicotinoid insecticides and spinosadin the Colorado potato beetle, Leptinotarsa decemlineata (Say) (Coleoptera:Chrysomelidae). Pest Manage. Sci. 2006, 62, 30–37.

(25) EI DuPont de Numours & Co. Cresols, ortho-, meta-, and para-.NTIS report no. OTS0205862; EI DuPont de Numours & Co., 1983.

(26) Anonymous. Final report on the safety assessment of sodiump-chloro-m-cresol, p-chloro-m-cresol, chlorothymol, mixed cresols, m-cresol,o-cresol, p-cresol, isopropyl cresols, thymol, o-cymen-5-ol, and carvacrol.Int. J. Toxicol. 2006, 25, 29–127.

(27) Booker, C. J.; Bedmutha, R.; Scott, I. M.; Conn, K.; Berruti, F.;Briens, C.; Yeung, K. K.-C. Bioenergy II: Characterization of the pesticideproperties of tobacco bio-oil. Int. J. Chem. Reactor Eng. 2010, 8, Art. 26.

(28) MacQuarrie, C. J. K.; Boiteau, G. Effect of diet and feeding historyon flight of Colorado potato beetle, Leptinotarsa decemlineata. Entomol.Exp. Appl. 2003, 107, 207–213.

ReceiVed for reView February 11, 2010ReVised manuscript receiVed August 15, 2010

Accepted September 1, 2010

IE100329Z

Ind. Eng. Chem. Res., Vol. 49, No. 20, 2010 10079