Evaluation of alkyl esters from Camelina sativa oil as biodiesel and as blend components in ultra low-sulfur diesel fuel

Bioresource Technology 101 (2010) 646–653

Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior tech

Evaluation of alkyl esters from Camelina sativa oil as biodiesel and as blendcomponents in ultra low-sulfur diesel fuel q

Bryan R. Moser *, Steven F. VaughnUnited States Department of Agriculture, Agricultural Research Service, National Center for Agricultural Utilization Research, 1815 N. University St., Peoria, IL 61604, USA

a r t i c l e i n f o a b s t r a c t

Article history:Received 15 December 2008Received in revised form 10 August 2009Accepted 12 August 2009Available online 8 September 2009

Keywords:BiodieselCamelina sativaDieselFuel propertiesMethyl esters

0960-8524/$ - see front matter Published by Elsevierdoi:10.1016/j.biortech.2009.08.054

q Disclaimer: Product names are necessary to reporhowever, the USDA neither guarantees nor warrantsand the use of the name by USDA implies no appexclusion of others that may also be suitable.

* Corresponding author. Tel.: +1 309 681 6511; faxE-mail address: [email protected] (B.R. M

Methyl and ethyl esters were prepared from camelina [Camelina sativa (L.) Crantz] oil by homogenousbase-catalyzed transesterification for evaluation as biodiesel fuels. Camelina oil contained high percent-ages of linolenic (32.6 wt.%), linoleic (19.6 wt.%), and oleic (18.6 wt.%) acids. Consequently, camelina oilmethyl and ethyl esters (CSME and CSEE) exhibited poor oxidative stabilities and high iodine values ver-sus methyl esters prepared from canola, palm, and soybean oils (CME, PME, and SME). Other fuel prop-erties of CSME and CSEE were similar to CME, PME, and SME, such as low temperature operability, acidvalue, cetane number, kinematic viscosity, lubricity, sulfur and phosphorous contents, as well as surfacetension. As blend components in ultra low-sulfur diesel fuel, CSME and CSEE were essentially indistin-guishable from SME and soybean oil ethyl ester blends with regard to low temperature operability, kine-matic viscosity, lubricity, and surface tension.

Published by Elsevier Ltd.

1. Introduction

Biodiesel, defined as an alternative fuel composed of mono-al-kyl esters of long-chain fatty acids prepared from vegetable oilsor animal fats, has attracted considerable interest as a substituteor blend component for conventional petroleum diesel fuel (petro-diesel). Biodiesel possesses significant technical advantages overpetrodiesel, such as derivation from renewable feedstocks, dis-placement of imported petroleum, inherent lubricity, essentiallyno sulfur content, superior flash point and biodegradability, re-duced toxicity, as well as a reduction in most exhaust emissions.Important disadvantages of biodiesel include inferior oxidativeand storage stability, lower volumetric energy content, inferiorlow temperature operability, and higher NOx exhaust emissions(Mittelbach and Remschmidt, 2004; Knothe et al., 2005;McCormick et al., 2006; Moser, 2009a). Additionally, the high costof commodity vegetable oils, such as soybean oil in the UnitedStates, represents a serious threat to the economic viability ofthe biodiesel industry (Paulson and Ginder, 2007; Retka-Schill,2008a). Presently, feedstock acquisition accounts for up to 85% ofbiodiesel production costs (Paulson and Ginder, 2007; Retka-Schill,2008a).

Ltd.

t factually on available data;the standard of the product,roval of the product to the

: +1 309 681 6340.oser).

Feedstocks for biodiesel production vary considerably withlocation according to climate and availability. Generally, the mostabundant lipid in a particular region is the most common feed-stock. Thus, rapeseed oil is predominantly used in Europe, palmoil predominates in tropical countries, and soybean oil and animalfats are primarily used in the United States (Mittelbach and Rems-chmidt, 2004; Knothe et al., 2005; Moser, 2009a). However, manyof these oils are prohibitively expensive and have competingfood-related uses. Consequently, the development of alternativefeedstocks for biodiesel production that meet all or most of the fol-lowing criteria has attracted considerable research attention: highoil content, low agricultural inputs, favorable fatty acid composi-tion, compatibility with existing farm equipment and infrastruc-ture, production in a sustainable fashion in off-season or inagriculturally undesirable lands, definable growth seasons, anduniform seed maturation rates. Selected recent examples of biodie-sel produced from feedstock oils that meet at least a majority ofthe above criteria include moringa (Rashid et al., 2008), jatropha(Tiwari et al., 2007), field pennycress (Moser et al., 2009a), andwild mustard (Jham et al., 2009).

Camelina [Camelina sativa (L.) Crantz], also known as false flaxor gold-of-pleasure, is a spring annual broadleaf oilseed herb ofthe Brassicaceae family that grows well in temperate climates.Camelina has several positive agronomic attributes: low agricul-tural inputs, cold-weather tolerance, short growing season(85–100 days), compatibility with existing farm equipment, andgrows well in semiarid regions and in low-fertility or saline soils.These qualities are unusual for an oilseed crop (Putnam et al.,1993; Retka-Schill, 2008b; Sawyer, 2008). Other more common

B.R. Moser, S.F. Vaughn / Bioresource Technology 101 (2010) 646–653 647

oilseed crops, such as canola, soybean, rapeseed, and sunflower,have higher water, pesticide, and fertilizer requirements (Budinet al., 1995; Sawyer, 2008). Moreover, camelina, unlike soybean,thrives in cool, arid climates and is nicely adapted to the morenortherly regions of North America, Europe, and Asia. As such, itmay serve as a rotational crop for winter wheat, which would facil-itate disrupting undesirable weed and pest cycles (Retka-Schill,2008b). Camelina yields an average of 420–640 L/ha and the pro-tein and fiber content in its meal byproduct is comparable to thatof soybean meal (Retka-Schill, 2008b; Sawyer, 2008). The seeds ofcamelina contain 28–40 wt.% of vegetable oil (Putnam et al., 1993;Budin et al., 1995), which is far superior to that typically found insoybeans (18–22 wt.%). Camelina has traditionally been used forrelatively high value products such as culinary oil, cosmetics, andanimal feed (Sawyer, 2008).

Examples of camelina oil appearing in the scientific literatureinclude its direct use as a fuel for diesel transport engines(Bernardo et al., 2003) and as a chemical feedstock in the enzy-matic production of long-chain esters (Steinke et al., 2000a,b,2001). Camelina oil may also serve as an industrial source of a-linolenic acid (9Z,12Z,15Z-octadecatrienoic acid), as it generallycontains 30–45 wt.% of this constituent (Leonard, 1998; Ni Eidhinet al., 2003; Krist et al., 2006; Sawyer, 2008). Moreover, camelinaoil was examined as a feedstock for biodiesel production (Frohlichand Rice, 2005). We follow up in the current study with a morethorough examination of the fuel properties of methyl and ethylesters prepared from alcoholysis of camelina oil, along with theirevaluation as blend components in ultra low-sulfur diesel fuel(ULSD, <15 ppm S). A comparison with canola and palm oil methylesters (CME and PME, respectively), along with soybean oil methyland ethyl esters (SME and SEE, respectively), was also of interest.Important fuel properties elucidated of the alkyl esters, both neatand blended with ULSD, included low temperature operability, oxi-dative stability, cetane number, acid value, sulfur and phosphorouscontents, lubricity, iodine value, surface tension, and kinematicviscosity.

Although biodiesel can be used in modern unmodified dieselengines in the neat form, it is more commonly encountered as ablend component in petrodiesel, such as B20 (20% biodiesel by vol-ume in petrodiesel). Currently, blends of up to B5 are allowed inASTM D975 (ASTM, 2008a) (Table 1), the US diesel fuel standard.In addition, ASTM D7467 (ASTM, 2008b) (Table 1) was recentlyadopted for blends from B6 to B20 in petrodiesel. Biodiesel mustbe certified as acceptable according to ASTM D6751 (ASTM,

Table 1Selected specifications from biodiesel and ULSD fuel standards.

Biodiesel ULSD

ASTM D6751 EN 14214 ASTM D975 ASTM D7467

vol.% BD 100 100 0–5 6–20CP (�C) Report – Not

specifiedaNotspecifieda

PP (�C) – – Notspecifieda

Notspecifieda

CFPP (�C) – Variableb Notspecifieda

Notspecifieda

OSI, 110 �C (h) 3 min 6 min – 6 mint, 40 �C (mm2/s) 1.9–6.0 3.5–5.0 1.9–4.1 1.9–4.1Lub, 60 �C (lm) – – 520 max 520 maxAV (mg KOH/g) 0.50 max 0.50 max – 0.30 maxIV (g I2/100 g) – 120 max – –CN 47 min 51 min 40 min 40 minS (ppm) 15 max 10 max 15 max 15 maxP (mass%) 0.001 max 0.001 max – –

a There are no requirements for low temperature operability in ASTM D975 orD7467, only guidance.

b Variable by location and time of year.

2008c) (Table 1), the B100 standard, before its use as a fuel or blendcomponent. Moreover, EN 14214 (Committee for Standardization,2003) (Table 1) and EN 590 (Committee for Standardization,2004) are the European Union standards for biodiesel and petro-diesel (up to B5), respectively.

2. Experimental section

2.1. Materials

Camelina seeds were purchased from Marx Foods (AtlanticHighlands, NJ). ULSD, described as fungible by the manufacturer,was donated by a major petrochemical company that wished to re-main anonymous. Conductivity and corrosion inhibitor additiveswere added by the manufacturer, but no drag reducing, lubricity,low temperature, or antioxidant additives were added. The com-mercial adsorbent Magnesol� was purchased from The DallasGroup of America, Inc. (Whitehouse, NJ). Refined, bleached, anddeodorized canola, palm, and soybean oils were purchased fromKIC Chemicals, Inc (New Platz, NY). Fatty acid methyl (FAME) andethyl ester standards were purchased from Nu-Chek Prep, Inc. (Ely-sian, MN). All other chemicals and reagents were obtained fromSigma–Aldrich Corp (St. Louis, MO).

2.2. Camelina oil extraction

Camelina seeds (1000 g) were ground in a coffee grinder and oilwas extracted with hexane for 24 h in a Soxhlet apparatus. Hexanewas removed by rotary evaporation under reduced pressure(10 mbar; 30 �C). For determination of total oil content, 10.0 g trip-licates of ground seed were extracted (Soxhlet) for 24 h, and afterhexane was removed by rotary evaporation (10 mbar; 30 �C) theweight of the residual oil was calculated. The percentage of oilrecovered from the samples was 30.5 wt.% (rM ± 0.4%).

2.3. Methanolysis of camelina oil

Methanolysis of camelina oil was conducted in a 500 mL three-necked round bottom flask connected to a reflux condenser and amechanical magnetic stirrer set at 1200 rpm. Initially, camelinaoil (180 g, 200 mL, 0.202 mol) and methanol (49 mL, 1.21 mol)were added to the flask and heated to 60 �C (internal reaction tem-perature monitored by digital temperature probe), followed byaddition of 1.0 wt.% NaOCH3 (25 wt.% in CH3OH). After 1.5 h ofreaction the mixture was equilibrated to room temperature andtransferred to a separatory funnel. The lower glycerol phase wasremoved by gravity separation (>2 h settling time) followed by re-moval of methanol from the upper crude methyl ester phase by re-duced pressure (10 mbar; 30 �C) rotary evaporation. The crudemethyl esters were treated with 2.5 wt.% Magnesol� at 65 �C for25 min with continuous stirring (500 rpm), followed by filtrationto remove the adsorbent to provide C. sativa oil methyl esters(CSME, 160 g, 89 wt.% yield based on original mass of camelinaoil). Canola, palm, and soybean oil methyl esters were preparedin a similar fashion in essentially quantitative (>98 wt.%) yield.Free and total glycerol contents of CSME, CME, PME, and SME weredetermined according to ASTM D6584.

2.4. Ethanolysis of camelina oil

Ethanolysis of camelina oil was accomplished largely as de-scribed in Section 2.3. Initially, camelina oil (180 g, 200 mL,0.202 mol) and ethanol (110 mL, 1.82 mol) were added to the reac-tion flask and heated to 70 �C (internal reaction temperature mon-itored by digital temperature probe), followed by addition of

Table 2Fatty acid composition (wt.%) of camelina oil (CSO), with canola (CO), palm (PO), andsoybean (SBO) oils shown for comparison.

Fatty acida CSO CO PO SBO

C12:0 Trace 0.3C14:0 0.1 1.1C16:0 6.8 4.6 41.9 10.5C18:0 2.7 2.1 4.6 4.1C20:0 1.5 0.7 0.3 TraceC22:0 0.2 0.3 TraceC16:1 9 Trace 0.2 0.2C18:1 9 18.6 64.3 41.2 22.5C18:1 11 1.1 1.6C18:2 9, 12 19.6 20.2 10.3 53.6C18:3 9, 12, 15 32.6 7.6 0.1 7.7C20:1 11 12.4 Trace TraceC20:2 11, 14 1.3 TraceC20:3 11, 14, 17 0.8C22:1 13 2.3C24:1 15 Trace

a For example, C18:1 9 signifies an 18 carbon fatty acid chain with one doublebond located at carbon 9 (methyl 9Z-octadecenoate; methyl oleate). All doublebonds are cis.

648 B.R. Moser, S.F. Vaughn / Bioresource Technology 101 (2010) 646–653

1.0 wt.% KOH (1.8 g). After 1.5 of reaction the mixture was cooledto room temperature and ethanol was removed by rotary evapora-tion (10 mbar; 35 �C) followed by removal of glycerol by gravityseparation (>2 h settling time). The crude ethyl esters were treatedwith 2.5 wt.% Magnesol� as described in Section 2.3 to afford puri-fied C. sativa oil ethyl esters (CSEE, 151 g, 84 wt.% yield). Soybeanoil ethyl esters were prepared in a similar fashion in high yield(94 wt.%). Free and total glycerol contents of CSEE and SEE weredetermined according to ASTM D6584.

2.5. Fatty acid profile by GC

Fatty acid methyl esters (FAME) of camelina, canola, palm, andsoybean oils were separated (triplicates, means reported) using aVarian (Walnut Creek, CA) 8400 GC equipped with an FID detectorand a SP2380 (Supelco, Bellefonte, PA) column (30 m � 0.25 mmi.d., 0.20 lm film thickness). Carrier gas was He at 1 mL/min. Theoven temperature was initially held at 150 �C for 15 min, then in-creased to 210 �C at 2 �C/min, followed by an increase to 220 �Cat 50 �C/min, which was then held for 10 min. The injector anddetector temperatures were set to 240 �C and 270 �C, respectively.FAME peaks were identified by comparison to the retention timesof reference standards.

2.6. Fuel properties of alkyl esters and blends with ULSD

Cloud and pour point (CP and PP, respectively) determinations(triplicates, means reported) were made according to ASTMD5773 and ASTM D5949, respectively, using a model PSA-70SPhase Technology Analyzer (Richmond, B.C., Canada). Cloud andpour points were rounded to the nearest whole degree (�C). For agreater degree of accuracy, PP measurements were done with aresolution of 1 �C instead of the specified 3 �C increment. Cold filterplugging point (CFPP) was measured (triplicates, means reported)in accordance with ASTM D6371 utilizing a model FPP-5Gs ISLAutomatic CFPP Analyzer (Houston, TX).

Kinematic viscosity (t, mm2/s) was determined (triplicates,means reported) with Cannon–Fenske viscometers (Cannon Instru-ment Co., State College, PA) at 40 �C in accordance with ASTMD445.

Lubricity (Lub, lm) determinations (duplicates, means re-ported) were performed at 60 �C (±1 �C) according to ASTMD6079 using a high-frequency reciprocating rig (HFRR) lubricitytester (PCS Instruments, London, England) from Lazar Scientific(Granger, IN). Reported wear scar (lm) values were the result ofmeasuring the maximum lengths of the x- and y-axes of each wearscar with a Prior Scientific (Rockland, Massachusetts, USA) Epimatmodel M4000 microscope, followed by calculating the average ofthese maximum values.

Oil stability index (OSI, h) was measured (triplicates, means re-ported) following EN 14112 with a Metrohm USA, Inc. (Riverview,FL) model 743 Rancimat instrument. For biodiesel (B100) samples,the flow rate of air through 3 ± 0.01 g of sample was 10 L/h with ablock temperature of 110 �C and a correction factor (DT) of 1.5 �C.The glass conductivity measuring vessel contained 50 ± 0.1 mL ofdeionized water. For biodiesel samples blended with ULSD, theflow rate of air through 7.5 ± 0.1 g of sample was 10 L/h with ablock temperature of 110 �C (DT = 1.5 �C). The conductivity mea-suring vessel contained 60 ± 0.1 mL of deionized water. Longerreaction vessels and gas inlet tubes were used for the blends toreduce ULSD evaporation. OSI was determined as the inflectionpoint of a computer-generated plot of conductivity (lS/cm) ofdeionized water versus time (h). Experiments were terminatedat 24 h.

Acid value (AV, mg KOH/g) titrations (triplicates, means re-ported) were performed as described in American Oil Chemists’

Society (AOCS) official method Cd 3d-63 using a Metrohm 836 Titr-ando (Westbury, NY) autotitrator equipped with a model 801 stir-rer and a Metrohm 6.0229.100 Solvotrode. However, the officialmethod was modified for scale to use 2 g of sample and 0.02 MKOH. The titration endpoint was automatically determined bythe instrument and visually verified using a phenolphthalein indi-cator. Iodine value (IV) was calculated from the fatty acid profileaccording to AOCS Cd 1c-85.

Surface tension (c, mN/m) was determined (quintuplets, meansreported) at 24 ± 1 �C and 40 ± 1 �C with a Sita t60 bubble pressuretensiometer (Dresden, Germany). Dilutions and temperature con-trol were handled by a Cat Ingenieurbüro M. Zipperer GmbH (Stau-fen, Germany) M26 stir plate and a model l10MC burette. A bubblelifetime of at least 4 s was used so that dynamic effects were not afactor in the measurements. The instrument was calibrated usingpure water. Additionally, the surface tensions of several organicsolutions were measured, and found to agree with literaturevalues.

Sulfur (S, ppm) and phosphorous (P, mass%) were measured byMagellan Midstream Partners, L.P. (Kansas City, KS) according toASTM standards D5453 and D4951, respectively. Derived cetanenumber (DCN) was determined by Southwest Research Institute(San Antonio, TX) utilizing an Ignition Quality TesterTM (IQT) fol-lowing ASTM D6890.

3. Results and discussion

3.1. Composition and physical properties of camelina oil

The oil content of camelina seeds was 30.5 wt.%, which was inagreement with the range reported in previous studies (Budinet al., 1995; Leonard, 1998; Sawyer, 2008). The primary fatty acididentified in camelina oil was linolenic acid (C18:3; Table 2) at32.6 wt.%, with other unsaturated fatty acids such as linoleic(C18:2; 19.6 wt.%), oleic (C18:1; 18.6 wt.%), and eicosenoic(C20:1; 12.4 wt.%) acids also detected in significant quantities.The majority of the remaining fatty acids aside from erucicacid (C22:1; 2.3 wt.%) were saturated species such as palmitic(C16:0; 6.8 wt.%), stearic (C18:0; 2.7 wt.%), and arachidic (C20:0;1.5 wt.%) acids. These results are in close agreement with prior re-ports on the fatty acid profile of camelina oil (Budin et al., 1995;Frohlich and Rice, 2005; Krist et al., 2006). The relatively high poly-unsaturated (88.7 wt.%) and trienoic (33.4 wt.%) acid contents of

Table 3Fatty acid trends and physical properties of camelina, canola, palm, and soybean oils.

CSO CO PO SBO

R saturatesa 11.3 7.7 48.2 14.6R unsaturatesb 88.7 92.3 51.8 85.4R polyunsaturatesc 54.3 27.8 10.4 61.3R C20+d 16.5 Trace TraceR trienese 33.4 7.6 0.1 7.7CP (�C) �10 (1)f �9 (1) 11 (1) �7 (1)PP (�C) �17 (1) �21 (1) 8 (1) �9 (1)OSI, 110 �C (h) 2.2 (0.2) 10.3 (0.2) 24.7 (0.2) 8.3 (0.1)t, 40 �C (mm2/s) 28.28

(0.05)36.28(0.05)

41.69(0.02)

31.49(0.02)

Lub, 60 �C (lm) 128 (2) 137 (3) 148 (4) 124 (1)AV (mg KOH/g) 2.06 (0.04) 0.04 (0.03) 0.15 (0.03) 0.03 (0.02)IV (g I2/100 g) 151 110 54 134c, 24 �C (mN/m) 32.3 (0.1) 34.2 (0.1) 33.8 (0.1) 34.2 (0.1)c, 40 �C (mN/m) 30.7 (0.1) 32.2 (0.1) 32.0 (0.1) 32.2 (0.1)

a R saturates (wt.%) = C12:0 + C14:0 + C16:0 + C18:0 + C20:0 + C22:0.b R unsaturates (wt.%) = C16:1 + C18:1 + C18:2 + C18:3 + C20:1 + C20:2 + C20:3 +

C22:1 + C24:1.c R polyunsaturates (wt.%) = C18:2 + C18:3 + C20:2 + C20:3.d R C20+ (wt.%) = C20:0 + C22:0 + C20:1 + C20:2 + C20:3 + C22:1 + C24:1.e R trienes (wt.%) = C18:3 + C20:3.f Values in parentheses are standard deviations from the reported means.

B.R. Moser, S.F. Vaughn / Bioresource Technology 101 (2010) 646–653 649

camelina oil resulted in an IV (151) that was substantially higherthan the other oils listed in Table 3.

The AV of crude camelina oil was 2.06 mg KOH/g (Table 3). Re-fined, bleached and deodorized (RBD) canola, palm, and soybeanoils exhibited very low AV (<0.16 mg KOH/g, Table 3). The oil sta-bility index (OSI) of camelina oil, determined at 110 �C according tothe Rancimat method (EN 14112) was lowest among the oils listedin Table 3 with a value of 2.2 h and was consistent with its highpolyunsaturated fatty acid content. The CP and PP of camelina oilwere �10 �C and �17 �C, respectively. With the exception of palmoil, these values did not vary significantly from the other oils listedin Table 3. The kinematic viscosity (40 �C) of camelina oil(28.28 mm2/s) was noticeably lower than the values obtained forpalm, canola, and soybean oils (41.69, 36.28, and 31.49 mm2/s,respectively, Table 3). The wear scar generated by camelina oilaccording to ASTM D6079 (60 �C) was 128 lm (shorter wear scarsindicate better lubricity), which with the exception of palm oil(148 lm) did not vary significantly from the other oils in Table 3.The surface tension at 24 �C and 40 �C of camelina oil (32.3 and30.7 nN/m, respectively) was similar to the values obtained for ca-nola, palm, and soybean oils (Table 3).

Table 4Fuel properties of camelina, canola, palm, and soybean oil alkyl esters, along with ultra lo

CSME CSEE CME

CP (�C) 3 (1)b 2 (1) 0 (1)PP (�C) �4 (1) �4 (1) �9 (1)CFPP (�C) �3 (1) �3 (1) �7 (1)OSI, 110 �C (h) 2.5 (0.1) 2.9 (0.1) 6.4 (0.1)t, 40 �C (mm2/s) 4.15 (0.01) 4.48 (0.01) 4.42 (0.23)Lub, 60 �C (lm) 122 (3) 145 (5) 169 (1)AV (mg KOH/g) 0.31 (0.01) 0.41 (0.02) 0.01 (0.01)IV (g I2/100 g) 151 144 110S (ppm) 3 3 2P (mass%) 0 0 0CN 52.8 nd ndc, 24 �C (mN/m) 31.0 (0.1) 30.8 (0.1) 30.9 (0.1)c, 40 �C (mN/m) 29.5 (0.1) 29.3 (0.1) 29.4 (0.1)

a CSME: camelina oil methyl esters; CSEE: camelina oil ethyl esters; CME: canola oilsoybean oil ethyl esters, ULSD: ultra low-sulfur diesel fuel.

b Values in parentheses are standard deviations from the reported means.c nd = not determined.

3.2. Preparation and quality of alkyl esters from camelina oil

Camelina oil was subjected to homogenous base-catalyzedtransesterification employing classic conditions described previ-ously (Freedman et al., 1984; Moser, 2009a) to afford camelinaoil methyl (CSME) and ethyl (CSEE) esters. The yields of CSME(89 wt.%) and CSEE (84 wt.%) were not quantitative. Under idealconditions, base-catalyzed transesterifications normally proceedto completion (Tiwari et al., 2007). The culprit was most likelythe free fatty acid content of camelina oil, as measured by AV. Freefatty acids react with homogenous base catalysts such as sodiumhydroxide and methoxide to form soap (sodium salt of fatty acid)and water (or methanol in the case of sodium methoxide), thusirreversibly quenching the catalyst and reducing product yield(Lotero et al., 2005).

The AVs of CSME and CSEE were 0.31 and 0.41 mg KOH/g,respectively (Table 4). Although these values were higher thanthose obtained from methanolysis of RBD canola, palm, and soy-bean oils (Table 4), they were below the maximum prescribed lim-its in ASTM D6751 and EN 14214 (<0.50 g KOH/g; Table 1). The AVof SEE was 0.37 mg KOH/g (Table 4), which was comparable toCSME and CSEE. Additionally, the free and total glycerol contentsof alkyl esters from camelina, canola, palm, and soybean oils werebelow the maximum allowable limits specified in ASTM D6751 andEN 14214.

3.3. Physical properties of camelina oil alkyl esters

The low temperature operability of CSME and CSEE was mea-sured through CP, PP, and CFPP determination. As seen in Table4, CSME provided CP, PP, and CFPP values of 3, �4, and �3 �C,respectively, which were indistinguishable from CSEE. Methyl es-ters prepared from canola oil yielded superior low temperatureproperties, as indicated by lower CP, PP, and CFPP. The values ob-tained for SME and SEE were similar to CSME and CSEE. As ex-pected, PME exhibited noticeably diminished low temperatureoperability in comparison to CSME and CSEE (Moser et al., 2008;Moser, 2008a). Strategies to improve the low temperature opera-bility of CSME and CSEE may include the use of additives, blendingwith biodiesel produced from other feedstocks, crystallization frac-tionation, transesterification with long- or branched-chain alco-hols, and/or blending with petrodiesel (Knothe et al., 2005;Moser, 2008a, 2009a; Moser et al., 2008).

The OSI values of CSME and CSEE were 2.5 and 2.9 h (Table 4),respectively. Methyl esters from canola (6.4 h), palm (10.3 h) and

w-sulfur diesel fuela.

PME SME SEE ULSD

17 (1) 0 (1) 0 (1) �18 (1)15 (1) �3 (1) �4 (1) �23 (1)12 (1) �4 (1) �5 (1) �17 (1)10.3 (0.1) 5.0 (0.1) 6.0 (0.1) >244.58 (0.01) 4.12 (0.01) 4.41 (0.02) 2.30 (0.01)172 (1) 135 (1) 137 (3) 571 (5)0.01 (0.01) 0.01 (0.01) 0.37 (0.01) 054 134 127 Ndc

2 1 1 80 0 0 ndnd nd nd 41.430.5 (0.1) 31.0 (0.1) 30.4 (0.1) 27.3 (0.1)29.0 (0.1) 29.5 (0.1) 29.1 (0.1) 25.9 (0.1)

methyl esters; PME: palm oil methyl esters; SME: soybean oil methyl esters; SEE:

650 B.R. Moser, S.F. Vaughn / Bioresource Technology 101 (2010) 646–653

soybean (5.0 h) oils were more stable to oxidation than CSME andCSEE as a result of the high trienoic fatty acid content of camelinaoil (33.4 wt.%; Table 3). The rate of autoxidation is dependant onthe number and location of methylene-interrupted double bondscontained within fatty acid alkyl esters that comprise biodiesel.Polyunsaturated materials, especially trienoic constituents, areparticularly vulnerable to autoxidation, as evidenced by the rela-tive rates of oxidation of the unsaturates: 1 for ethyl oleate, 41for ethyl linoleate, and 98 for ethyl linolenate (Holman and Elmer,1947). In addition, the OSI values of methyl esters of oleic, linoleic,and linolenic acids were reported as 2.5, 1.0, and 0.2 h, respectively(Moser, 2009b). The combined trienoic fatty acid contents of cano-la, palm, and soybean oils were 7.6, 0.1, and 7.7 wt.%, respectively(Table 3). Palm oil methyl esters were especially resistant to oxida-tion as a result of their high saturated (48.2 wt.%) and low polyun-saturated (10.4 wt.%) fatty ester composition. Both CSME and CSEEwere unsatisfactory according to oxidative stability specifications(Table 1) contained within ASTM D6751 (OSI >3 h) and EN 14214(OSI > 6 h). Accordingly, addition of antioxidant additives (Moser,2008b) or blending with more oxidatively stable feedstocks (Mo-ser, 2008a) would be necessary to yield satisfactory oxidative sta-bility from esters of camelina oil.

The relatively high polyunsaturated fatty acid content of CSMEand CSEE resulted in IVs of 151 and 144, respectively (Table 4).Although ASTM D6751 does not contain an IV specification, EN14214 limits IV to a maximum value of 120 (Table 1). Conse-quently, CSME and CSEE, along with SME (IV 134) and SEE (IV127) were unsatisfactory according to EN 14214. In order for came-lina oil to be acceptable as a feedstock with regard to IV in loca-tions where the EN 14214 standard is applied, blending withfeedstocks that contain a greater percentage of saturated fattyacids would be necessary (Moser, 2008a).

The kinematic viscosities (40 �C) of CSME and CSEE were 4.15and 4.48 mm2/s (Table 4), which were satisfactory according toASTM D6751 (1.9–6.0 mm2/s; Table 1) and EN 14214 (3.5–5.0 mm2/s; Table 1). Among the methyl esters, an inverse relation-ship was elucidated between kinematic viscosity and polyunsatu-rated FAME content: PME (most viscous; lowest polyunsaturatedester content) > CME > CSME > SME (least viscous; highest polyun-saturated content). This was expected, as it is known that increas-ing levels of unsaturation result in lower kinematic viscosities. Forexample, the reported kinematic viscosities (40 �C) of methyl es-ters of stearic, oleic, linoleic, and linolenic acids were 5.85, 4.51,3.65, and 3.14 mm2/s, respectively (Knothe and Steidley, 2005a).As anticipated, the kinematic viscosities of the ethyl esters (CSEEand SEE) were higher than the corresponding methyl esters (CSMEand SME).

The DCN of CSME was above the minimum limits of 47 and 51specified in ASTM D6751 and EN 14214, respectively, with a valueof 52.8 (Table 4). For comparison, a previous study reported theDCN of SME as 54.0 (Knothe et al., 2006). Results generated byASTM D6890 (DCN) generally correlate with CN determination byATSM D613. The ASTM D6890 method is approved as an alterna-tive to the more traditional CN method (ASTM D613) specified inASTM D6751.

The sulfur contents of CSME and CSEEE were 3 ppm (Table 4),which were satisfactory according to the specified maximumallowable limits in ASTM D6751 and EN 14214 of 15 and10 ppm, respectively. SME was essentially free of sulfur (1 ppm),along with CME (2 ppm) and PME (2 ppm). In addition, phospho-rous content is limited in ASTM D6751 and EN 14214 to a maxi-mum value of 0.001 mass%. None of the methyl esters containedphosphorous (Table 4).

The lubricities (60 �C) of CSME and CSEE were essentially indis-tinguishable from esters prepared from other oils (Table 4).Although lubricity is not specified in ASTM D6751 or EN 14214,

it is nonetheless an important fuel property. Fuels with poorlubricity can cause failure of diesel engine parts that rely on lubri-cation from fuels, such as fuel pumps and injectors (Knothe et al.,2005; Knothe and Steidley, 2005b). As such, lubricity specificationsare included in petrodiesel standards in the United States (ASTMD975; <520 lm by ASTM D6079; Table 1) and Europe (EN 590;<460 lm). As expected, all alkyl esters were considerably belowthese maximum allowable limits, which was in agreement withprevious studies that indicated biodiesel possessed inherentlygood lubricity (Knothe and Steidley, 2005b; Moser et al., 2008;Jham et al., 2009; Moser et al., 2009a). The wear scar generatedby CSME (122 lm; Table 4) was significantly shorter than the cor-responding scars for CME (169 lm), PME (172 lm), SME (135 lm),and SEE (137 lm), which was not surprising since increasing levelsof unsaturation result in superior lubricity (shorter wear scars). Forexample, the reported wear scars produced by methyl esters ofstearic, oleic, linoleic, and linolenic acids were 322, 290, 236, and183 lm, respectively (Knothe and Steidley, 2005b). As previouslystated, CSME and CSEE contained high percentages of in trienoicmethyl esters (C18:3 and C20:3).

Although surface tension is not specified in either ASTM D6751or EN 14214, it is nevertheless an important fuel property that af-fects atomization in combustion chambers in compression–igni-tion (diesel) engines (Ejim et al., 2007). The surface tensions at24 �C and 40 �C of CSME (31.0 and 29.5 mN/m) and CSEE (30.8and 29.3 mN/m) were similar to SME and slightly higher thanCME and PME (Table 4). Both CSME/EE and SME/EE contained high-er polyunsaturated fatty ester content (see Table 3) than CME andPME. Increasing levels of unsaturation were reported to increasesurface tension (Doll et al., 2007). Additionally, longer chainlengths among otherwise similar molecules are reported to in-crease surface tension (Allen et al., 1999), which further explainedwhy CSME and CSEE had higher surface tensions than CME andPME. Although CSME/EE had slightly higher surface tensions thanCME, SME/EE, and PME, it is expected that the differences are notsignificant enough to negatively influence fuel atomization.

3.4. Evaluation of camelina oil alkyl esters as blend components inULSD

Because biodiesel is completely miscible with ULSD, it can beused as a blend component in any proportion in ULSD. However,ASTM D975 and D7467 allow up to 5 and 20 vol.% biodiesel (Table1), respectively. The European petrodiesel standard, EN 590, allowsfor up to 5 vol.% biodiesel as well. Biodiesel and ULSD are notchemically similar: biodiesel is composed of long-chain fatty acidalkyl esters whereas ULSD is a complex mixture of aliphatic andaromatic hydrocarbons (12 or more carbons) that distill from crudepetroleum oil in a similar temperature range (250–350 �C). Fuelproperties determined in the current study that were affected byblending included lubricity, oxidative stability, low temperatureoperability, kinematic viscosity, and surface tension. Other proper-ties influenced by blending include exhaust emissions, flash point,sulfur content, and cetane number. As seen in Table 4, neat ULSDexhibited a DCN of 41.4, which was satisfactory according to ASTMD975 but significantly below the value obtained for CSME. The sul-fur content of ULSD was acceptable with regard to ASTM D975with a value of 8 ppm. Blending CSME with ULSD resulted in asmall reduction in sulfur content, as evidenced by values of 8, 8,7, and 7 ppm for the B2, B5, B10, and B20 blends. The results forDCN and sulfur content of ULSD were in agreement with a previousstudy (Moser et al., 2009b). Flash point and exhaust emissionswere not measured because biodiesel is reported to be superiorto petrodiesel with regard to these properties (Mittelbach andRemschmidt, 2004; Knothe et al., 2005). Furthermore, most regu-lated exhaust emissions (except for NOx) of petrodiesel are

B.R. Moser, S.F. Vaughn / Bioresource Technology 101 (2010) 646–653 651

reduced upon blending with biodiesel (McCormick et al., 2006;Moser et al., 2009b).

Biodiesel was considerably less stable to oxidation than ULSD,as evidenced by an OSI value of greater than 24 h for ULSD (Table5). Consequently, blends of CSME, CSEE, SME, and SEE exhibitedprogressively lower OSI values as the percentage of biodiesel wasincreased from 0 to 20 vol.% (Table 5). Comparison at the B20 levelrevealed that the CSME/EE blends were less stable to oxidationthan the SME/EE blends, which was consistent with the results ob-tained for neat methyl and ethyl esters discussed in Section 3.3.The petrodiesel–biodiesel blend standard, ASTM D7467 (Table 1),which covers B6 to B20 blends, specifies a minimum OSI value of6.0 h. As seen in Table 5, the B20 blend of CSEE failed to meet thisrequirement (5.8 h). Accordingly, employment of antioxidant addi-tives would be required in the case of B20 CSEE-petrodiesel blends.The petrodiesel standard, ASTM D975 (Table 1) does not contain anoxidative stability specification. As indicated in Table 5, all B2 andB5 blends of CSME/EE and SME/EE, with the exception of B5 CSEE,exhibited OSI values greater than 24 h.

Biodiesel is generally inferior to ULSD with regard to low tem-perature operability (Moser et al., 2008, 2009b). Illustrative of thisis a comparison of neat ULSD (CP �18 �C; PP �23 �C; CFPP �18 �C;Table 5) to the neat biodiesel fuels listed in Table 4. Consequently,blends of CSME, CSEE, SME, and SEE in ULSD displayed progres-sively higher CP, PP, and CFPP values as the percentage of biodieselwas increased from 0 to 20 vol.% (Table 5). Taking into consider-ation standard deviation, the behavior at each blend level amongCSME, CSEE, SME, and SEE blends did not vary significantly. Forexample, the CFPP of B5 blends with CSME, CSEE, SME, and SEEwere �15, �16, �17, and �16 �C, respectively. Neither ASTMD975 nor D7467 specify limits on low temperature operability. In-stead, only guidance is provided.

Petrodiesel is less viscous than biodiesel, as evidenced by thekinematic viscosity (40 �C) of ULSD (2.30 mm2/s; Table 5) versusthe biodiesel fuels (Table 4). As a result, blends of CSME, CSEE,SME, and SEE in ULSD afforded progressively higher kinematic vis-cosities as the percentage of biodiesel was increased from 0 to20 vol.% (Table 5). This result was in agreement with previous re-ports that examined the kinematic viscosities of SME and PME

Table 5Oxidative stability (OSI, h), low temperature operability (CP, PP, CFPP, �C), kinematic viscosiof camelina and soybean oil methyl and ethyl esters blended with ultra low-sulfur diesel

Vol.% OSI(h) CP(�C) PP(�C) CFPP(�C)

0 >24 �18 (1)a �23 (1) �17 (1)

Camelina oil methyl esters2 >24 �16 (1) �23 (1) �17 (1)5 >24 �14 (1) �23 (1) �15 (1)10 16.6 (0.1) �13 (1) �22 (1) �15 (1)20 8.9 (0.1) �11 (1) �20 (1) �14 (1)

Camelina oil ethyl esters2 >24 �15 (1) �23 (1) �17 (1)5 17.8 (1.1) �14 (1) �23 (1) �16 (1)10 9.1 (0.3) �13 (1) �22 (1) �15 (1)20 5.8 (0.3) �11 (1) �20 (1) �14 (1)

Soybean oil methyl esters2 >24 �18 (1) �22 (1) �17 (1)5 >24 �16 (1) �22 (1) �17 (1)10 >24 �14 (1) �21 (1) �17 (1)20 17.1 (0.2) �12 (1) �17 (1) �16 (1)

Soybean oil ethyl esters2 >24 �15 (1) �22 (1) �16 (1)5 >24 �15 (1) �22 (1) �16 (1)10 19.5 (0.4) �14 (1) �22 (1) �13 (1)20 13.1 (0.2) �13 (1) �17 (1) �11 (1)

a Values in parentheses are standard deviations from the reported means.

blends in ULSD (Moser et al., 2008) and SME blends in low-sulfurdiesel fuel (<500 ppm S) (Doll et al., 2008). The behavior at eachblend level among the CSME, CSEE, SME, and SEE blends did notvary significantly with regard to viscosity. For example, the kine-matic viscosities of the B5 blends of CSME, CSEE, SME, and SEEwere 2.38, 2.38, 2.37, and 2.38 mm2/s, respectively. All of theblends exhibited kinematic viscosities that were satisfactoryaccording to the petrodiesel (ASTM D975) and petrodiesel–biodie-sel blend (ASTM D7467) standards, which both specify a rangefrom 1.9–4.1 mm2/s (Table 1).

As previously stated, petrodiesel (untreated with lubricity-enhancing additives) was inferior to biodiesel with regard tolubricity. In fact, the wear scar generated by ULSD (571 lm; Table5) was significantly longer than biodiesel (<173 lm; Table 4) andin excess of the maximum allowable wear scar lengths specifiedin ASTM D6751 and D7467 (Table 1; <520 lm) and EN 590(<460 lm). Accordingly, blends of CSME, CSEE, SME, and SEE inULSD exhibited progressively better lubricities (shorter wear scars)as the percentage of biodiesel was increased from 0 to 20 vol.% (Ta-ble 5). In fact, even at the B2 blend level, CSME/EE and SME/EEblends were satisfactory according to the aforementioned petro-diesel standard (ASTM D975), providing wear scars below275 lm in all cases. Blends equal to or greater than the B5 blendlevel afforded wear scars below 200 lm (Table 5). These resultswere in agreement with a previous report that examined thelubricity behavior of SME and PME blends in ULSD from 0.1 to20 vol.% (Moser et al., 2008). Taking into consideration standarddeviation, the lubricity behavior at each blend level among theCSME, CSEE, SME, and SEE blends did not vary significantly. Forexample, the wear scars generated by the B10 blends of CSME,CSEE, SME, and SEE were 142, 169, 154, and 159 lm, respectively.

The surface tension of ULSD at 24 and 40 �C (27.3 and 25.9 mN/m; Table 5) was superior (lower value) to biodiesel (>29 mN/m; Ta-ble 4). Accordingly, blends of CSME, CSEE, SME, and SEE in ULSDexhibited progressively higher surface tensions as the percentageof biodiesel was increased from 0 to 20 vol.% (Table 5), whichwas in agreement with a previous study (Doll et al., 2008). How-ever, as the percentage of CSME, for example, in ULSD was in-creased from B2 (27.4 mN/m at 24 �C) to B20 (27.8 mN/m at

ty (mm2/s, 40 �C), lubricity (Lub, 60 �C), and surface tension (c, mN/m, 24 �C and 40 �C)fuel.

t(mm2/s) Lub(lm) c, 24 �C(mN/m) c, 40 �C(mN/m)

2.30 (0.01) 571 (5) 27.3 (0.1) 25.9 (0.1)

2.33 (0.01) 248 (8) 27.4 (0.1) 26.1 (0.1)2.38 (0.01) 147 (5) 27.5 (0.1) 26.0 (0.1)2.42 (0.01) 142 (7) 27.6 (0.1) 26.2 (0.1)2.58 (0.01) 136 (6) 27.8 (0.1) 26.5 (0.1)

2.33 (0.01) 260 (6) 27.4 (0.1) 26.0 (0.1)2.38 (0.01) 182 (3) 27.4 (0.1) 26.1 (0.1)2.48 (0.01) 169 (8) 27.5 (0.1) 26.2 (0.1)2.69 (0.02) 154 (5) 27.8 (0.1) 26.4 (0.1)

2.31 (0.01) 271 (7) 27.4 (0.1) 26.0 (0.1)2.37 (0.01) 198 (4) 27.4 (0.1) 26.0 (0.1)2.41 (0.01) 154 (3) 27.6 (0.1) 26.1 (0.1)2.54 (0.01) 143 (4) 27.8 (0.1) 26.4 (0.1)

2.31 (0.01) 262 (6) 26.0 (0.1) 25.9 (0.1)2.38 (0.01) 191 (5) 27.4 (0.1) 26.1 (0.1)2.47 (0.01) 159 (4) 27.5 (0.1) 26.2 (0.1)2.61 (0.01) 149 (5) 27.7 (0.1) 26.4 (0.1)

652 B.R. Moser, S.F. Vaughn / Bioresource Technology 101 (2010) 646–653

24 �C), a minimal increase in surface tension was noticed. Takinginto consideration standard deviation, the surface tension behaviorat each blend level among CSME, CSEE, SME, and SEE blends did notvary significantly. For example, the surface tensions (24 �C) of theB10 blends with CSME, CSEE, SME, and SEE were 27.6, 27.5, 27.6,and 27.5 mN/m, respectively. The minimal increase in surface ten-sion among the blends in comparison to neat ULSD was not consid-ered substantial enough to negatively impact fuel atomization.

4. Conclusions

Camelina oil, which was obtained in 30.5 wt.% from dried seeds,contained a high percentage of polyunsaturated fatty acids(54.3 wt.%), as indicated by its linolenic (32.6 wt.%) and linoleic(19.6 wt.%) acid contents. Other fatty acids identified in camelinaoil included oleic (18.6 wt.%), gondoic (12.4 wt.%), palmitic(6.8 wt.%), stearic (2.7 wt.%), and erucic (2.3 wt.%) acids, with lesseramounts of other constituents also detected.

Methyl and ethyl esters were prepared in moderately highyields (89 and 84 wt.%, respectively) from camelina oil by homog-enous base-catalyzed transesterification. The AV of crude camelinaoil (2.06 mg KOH/g) was attributed to non-quantitative alkyl esteryields.

The oxidative stability of biodiesel prepared from camelina oilwas unsatisfactory according to ASTM D6751 and EN 14214 as aresult of its high polyunsaturated fatty acid content. The high IVsof CSME and CSEE were also in excess of the limit contained inEN 14214. The low temperature operability of CSME and CSEEwas similar to biodiesel prepared from canola and soybean oils,and superior to PME. The kinematic viscosity, AV, DCN, lubricity,sulfur and phosphorous contents, as well as the surface tensionsof CSME and CSEE were satisfactory according to ASTM D6751and EN 14214, where applicable.

Evaluation of CSME and CSEE as blend components in ULSD re-vealed that blends at the B20 level were not satisfactory with re-gard to the oxidative stability specification contained withinASTM D7467. Otherwise, camelina-ULSD blends were indistin-guishable from soybean-ULSD blends with regard to low tempera-ture performance, kinematic viscosity, lubricity, and surfacetension. Camelina oil alkyl esters, like soybean oil alkyl esters,may be used as lubricity-enhancing additives in petrodiesel.

In summary, CSME and CSEE are acceptable as biodiesel fuelsprovided that antioxidant additives are employed. Overall, theproperties of camelina oil alkyl esters, both neat and blended withpetrodiesel, were similar to other commonly encountered biodieselfuels, such as soybean, canola, and palm oil methyl esters.

Acknowledgements

The authors acknowledge Benetria N. Banks, Ray K. Holloway,Erin L. Walter, and Jennifer R. Koch for excellent technicalassistance.

References

Allen, C.A.W., Watts, K.C., Ackman, R.G., 1999. Predicting the surface tension ofbiodiesel fuels from their fatty acid composition. J. Am. Oil Chem. Soc. 76, 317–323.

ASTM, 2008a. Standard specification for diesel fuel oils. In: Annual Book of ASTMStandards, ASTM International, West Conshohocken, Method D975-08a.

ASTM, 2008b. Standard specification for diesel fuel oil, biodiesel blend (B6 to B20).In: Annual Book of ASTM Standards, ASTM International, West Conshohocken,Method D7467-08a.

ASTM, 2008c. Standard specification for biodiesel fuel (B100) blend stock fordistillate fuels. In: Annual Book of ASTM Standards, ASTM International, WestConshohocken, Method D6751-08a.

Bernardo, A., Howard-Hildige, R., O’Connell, A., Nichol, R., Ryan, J., Rice, B., Roche, E.,Leahy, J.J., 2003. Camelina oil as a fuel for diesel transport engines. Ind. CropsProd. 17, 191–197.

Budin, J.T., Breene, W.M., Putnam, D.H., 1995. Some compositional properties ofCamelina (Camelina sativa L. Crantz) seeds and oils. J. Am. Oil Chem. Soc. 72,309–315.

Committee for Standardization, 2003. Automotive Fuels – Fatty Acid Methyl Esters(FAME) for Diesel Engines – Requirements and Test Methods. EuropeanCommittee for Standardization, Brussels, Method EN 14214:2003.

Committee for Standardization, 2004. Automotive Fuels – Diesel – Requirementsand Test Methods. European Committee for Standardization, Brussels, MethodEN 590:2004.

Doll, K.M., Moser, B.R., Erhan, S.Z., 2007. Surface tension studies of alkyl esters andepoxidized alkyl esters relevant to oleochemically based fuel additives. EnergyFuels 21, 3044–3048.

Doll, K.M., Sharma, B.K., Suarez, P.A.Z., Erhan, S.Z., 2008. Comparing biofuelsobtained from pyrolysis, of soybean oil or soapstock, with traditional soybeanbiodiesel: density, kinematic viscosity, and surface tensions. Energy Fuels 22,2061–2066.

Ejim, C.E., Fleck, B.A., Amirfazli, A., 2007. Analytical study for atomization ofbiodiesels and their blends in a typical injector: surface tension and viscosityeffects. Fuel 86, 1534–1544.

Freedman, B., Pryde, E.H., Mounts, T.L., 1984. Variables affecting the yields of fattyesters from transesterified vegetable oils. J. Am. Oil Chem. Soc. 61, 1638–1643.

Frohlich, A., Rice, B., 2005. Evaluation of Camelina sativa oil as a feedstock forbiodiesel production. Ind. Crops Prod. 21, 25–31.

Holman, R.A., Elmer, O.C., 1947. The rates of oxidation of unsaturated fatty acids andesters. J. Am. Oil Chem. Soc. 24, 127–129.

Jham, G.N., Moser, B.R., Shah, S.N., Holser, R.A., Dhingra, O.D., Vaughn, S.F., Berhow,M.A., Winkler-Moser, J.K., Isbell, T.A., Holloway, R.K., Walter, E.L., Natalino, R.,Anderson, J.C., Stelly, D.M., 2009. Wild Brazilian mustard (Brassica juncea L.)seed oil methyl esters as biodiesel fuel. J. Am. Oil Chem. Soc. 86, 917–926.

Knothe, G., Steidley, K.R., 2005a. Kinematic viscosity of biodiesel fuel componentsand related compounds. Influence of compound structure and comparison topetrodiesel fuel components. Fuel 84, 1059–1065.

Knothe, G., Steidley, K.R., 2005b. Lubricity of components of biodiesel andpetrodiesel. The origin of biodiesel lubricity. Energy Fuels 19, 1192–1200.

Knothe, G., Van Gerpen, J., Krahl, J., 2005. The Biodiesel Handbook. AOCS Press,Urbana.

Knothe, G., Sharp, C.A., Ryan III, T.W., 2006. Exhaust emissions of biodiesel,petrodiesel, neat methyl esters, and alkanes in a new technology engine. EnergyFuels 20, 403–408.

Krist, S., Stuebiger, G., Bail, S., Unterweger, H., 2006. Analysis of volatile compoundsand triacylglycerol composition of fatty seed oil gained from flax and false flax.Eur. J. Lipid Sci. Technol. 108, 48–60.

Leonard, E.C., 1998. Camelina oil: a-linolenic source. Inform 9, 830–838.Lotero, E., Liu, Y., Lopez, D.E., Suwannakarn, K., Bruce, D.A., Goodwin Jr., J.G., 2005.

Synthesis of biodiesel via acid catalysis. Ind. Eng. Chem. Res. 44, 5353–5363.

McCormick, R.L., Williams, A., Ireland, J., Brimhall, M., Hayes, R.R., 2006. Effects ofBiodiesel Blends on Vehicle Emissions. Fiscal Year 2006 Annual Operating PlanMilestone 10.4, NREL Milestone Report 540-40554. <http://www.nrel.gov/docs/fy07osti/40554.pdf>. (accessed 8.10.09).

Mittelbach, M., Remschmidt, C., 2004. Biodiesel – A Comprehensive Handbook.Martin Mittelbach, Graz.

Moser, B.R., 2008a. Influence of blending canola, palm, soybean, and sunflower oilmethyl esters on fuel properties of biodiesel. Energy Fuels 22, 4301–4306.

Moser, B.R., 2008b. Efficacy of myricetin as an antioxidant additive in methyl estersof soybean oil. Eur. J. Lipid Sci. Technol. 110, 1167–1174.

Moser, B.R., 2009a. Biodiesel production, properties, and feedstocks. In vitro Cell.Dev. Biol.-Plant 45, 229–266.

Moser, B.R., 2009b. Comparative oxidative stability of fatty acid alkyl esters byaccelerated methods. J. Am. Oil Chem. Soc. 86, 699–706.

Moser, B.R., Cermak, S.C., Isbell, T.A., 2008. Evaluation of castor and lesquerella oilderivatives as additives in biodiesel and ultra low sulfur diesel fuels. EnergyFuels 22, 1349–1352.

Moser, B.R., Knothe, G., Vaughn, S.F., Isbell, T.A., 2009a. Production and evaluation ofbiodiesel from field pennycress (Thlaspi arvense L.) oil. Energy Fuels 23, 4149–4155.

Moser, B.R., Williams, A., Haas, M.J., McCormick, R.L., 2009b. Exhaust emissionsand fuel properties of partially hydrogenated soybean oil methyl estersblended with ultra low sulfur diesel fuel. Fuel Process. Technol. 90, 1122–1128.

Ni Eidhin, D., Burke, J., O’Beirne, D., 2003. Oxidative stability of x3-rich camelina oiland camelina oil-based spread compared with plant and fish oils and sunflowerspread. J. Food Sci. 68, 345–353.

Paulson, N.D., Ginder, R.G., 2007. Working Paper 07-WP 448, Center for Agriculturaland Rural Development, Iowa State University, May.

Putnam, D.H., Budin, J.T., Field, L.A., Breene, W.M., 1993. Camelina: a promising lowinput oilseed. In: Janick, J., Simon, J. (Eds.), New Crops. John Wiley and Sons,New York, pp. 314–322.

Rashid, U., Anwar, F., Moser, B.R., Knothe, G., 2008. Moringa oleifera oil: a possiblesource of biodiesel. Bioresour. Technol. 99, 8175–8179.

Retka-Schill, S., 2008a. Walking a tightrope. Biodiesel Mag. 5 (3), 64–70.Retka-Schill, S., 2008b. Oilseed comes of age. Biodiesel Mag. 5 (11), 44–49.Sawyer, K., 2008. Is there room for camelina? Biodiesel Mag. 5 (7), 83–87.Steinke, G., Kirchhoff, R., Mukherjee, K.D., 2000a. Lipase-catalyzed alcoholysis of

crambe and camelina oil for the preparation of long-chain esters. J. Am. OilChem. Soc. 77, 361–366.

B.R. Moser, S.F. Vaughn / Bioresource Technology 101 (2010) 646–653 653

Steinke, G., Schonwiese, S., Mukherjee, K.D., 2000b. Alkali-catalyzed alcoholysis ofcrambe oil and camelina oil for the preparation of long-chain esters. J. Am. OilChem. Soc. 77, 367–371.

Steinke, G., Weitkamp, P., Klein, E., Mukherjee, K.D., 2001. High-yield preparation ofwax esters via lipase-catalyzed esterification using fatty acids and alcohols fromcrambe and camelina oils. J. Agric. Food Chem. 49, 647–651.

Tiwari, A.K., Kumar, A., Raheman, H., 2007. Biodiesel production from jatropha oil(Jatropha curcas) with high free fatty acids: an optimized process. BiomassBioenerg. 31, 569–575.