9934 Energy & Fuels 2009, 23, 4149–4155

4149

Production and Evaluation of Biodiesel from Field Pennycress (Thlaspi arWense L.) Oil† Bryan R. Moser,* Gerhard Knothe, Steven F. Vaughn, and Terry A. Isbell National Center for Agricultural Utilization Research, Agricultural Research SerVice, United States Department of Agriculture, 1815 North UniVersity Street, Peoria, Illinois 61604

Downloaded by USDA NATIONAL AGRICULTURAL LIB on September 9, 2009 | http://pubs.acs.org Publication Date (Web): July 2, 2009 | doi: 10.1021/ef900337g

ReceiVed April 15, 2009. ReVised Manuscript ReceiVed June 1, 2009

Field pennycress (Thlaspi arVense L.) oil is evaluated for the first time as a feedstock for biodiesel production. Biodiesel was obtained in 82 wt % yield by a standard transesterification procedure with methanol and sodium methoxide catalyst at 60 °C and an alcohol to oil molar ratio of 6:1. Acid-catalyzed pretreatment to reduce the acid value of crude field pennycress oil resulted in a yield after methanolysis of 94 wt %. Field pennycress oil had high contents of erucic (13(Z)-docosenoic; 32.8 wt %) and linoleic (9(Z),12(Z)-octadecadienoic; 22.4 wt %) acids with other unsaturated fatty acids comprising most of the remaining fatty acid profile. As a result, the methyl esters (biodiesel) obtained from this oil exhibited a high cetane number of 59.8 and excellent low temperature properties, as evidenced by cloud, pour, and cold filter plugging points of -10, -18, and -17 °C, respectively. The kinematic viscosity and oxidative stability (Rancimat method) of field pennycress oil methyl esters were 5.24 mm2/s (40 °C) and 4.4 h (110 °C), respectively. Other fuel properties such as acid value, lubricity, free and total glycerol content, surface tension, as well as sulfur and phosphorus contents were also determined and are discussed in light of biodiesel standards such as ASTM D6751 and EN 14214. Also reported for the first time are cetane numbers of methyl esters of erucic and gondoic (methyl 11(Z)-eicosenoate) acids, which were found to be 74.2 and 73.2, respectively. In summary, field pennycress oil appears to an acceptable feedstock for biodiesel production.

1. Introduction Biodiesel, defined as an alternative fuel composed of monoalkyl esters of long-chain fatty acids prepared from vegetable oils or animal fats by the American Society for Testing and Materials (ASTM),1 has attracted considerable interest as a substitute or blend component for conventional petroleum diesel fuel (petrodiesel). Technical advantages of biodiesel include derivation from renewable and domestic feedstocks, displacement of imported petroleum, inherent lubricity, essentially no sulfur content, superior flash point and biodegradability, reduced toxicity, and a reduction in most regulated exhaust emissions. Important disadvantages versus petrodiesel are inferior oxidative and storage stability, lower volumetric energy content, inferior low temperature operability, and in most cases higher NOx exhaust emissions.2-5 Biodiesel must be satisfactory according to accepted fuel standards such as ASTM D67511 in the United † Disclaimer: Product names are necessary to report factually on available data; however, the USDA neither guarantees nor warrants the standard of the product, and the use of the name by USDA implies no approval of the product to the exclusion of others that may also be suitable. * To whom correspondence should be addressed. Telephone: (309) 6816511. Fax: (309) 681-6340. E-mail: [email protected]. (1) American Society for Testing and Materials. Standard specification for biodiesel fuel blend stock (B100) for middle distillate fuels, ASTM D6751-08. In ASTM Annual Book of Standards; American Society for Testing and Materials: West Conshohocken, PA, 2008. (2) Mittelbach, M.; Remschmidt, C. BiodieselsThe ComprehensiVe Handbook; M. Mittelbach: Graz, Austria, 2004. (3) Knothe, G.; Krahl, J.; Van Gerpen, J. The Biodiesel Handbook; AOCS Press: Champaign, IL, 2005. (4) McCormick, R. L.; Williams, A.; Ireland, J.; Brimhall, M.; Hayes, R. R. Fiscal year 2006 annual operating plan milestone 10.4; NREL Milestone Report 540-40554, 2006; http://www.nrel.gov/docs/fy07osti/ 40554.pdf. (5) Moser, B. R. In Vitro Cell. DeV. Biol.-Plant 2009, 45, 229-266.

10.1021/ef900337g

States or the Committee for Standardization (CEN) standard EN 142146 in Europe before combustion in compression-ignition (diesel) engines. The high cost of commodity vegetable oils represents a significant challenge to the biodiesel industry.7,8 Presently, feedstock acquisition accounts for 80% or more of the costs associated with biodiesel production.7,8 Feedstock availability varies with geography and climate, as the most abundant lipids in a particular region are the most common biodiesel feedstocks. Thus, rapeseed oil is principally used in Europe, palm oil predominates in tropical countries, and soybean oil and animal fats are primarily used in the United States.2,3,9 However, many commodity vegetable oils are prohibitively expensive and have competing food-related applications. In addition, if all currently available commodity vegetable oils were consumed solely for biodiesel production, the amount of fuel produced would still not suffice to completely displace petrodiesel at current usage levels. Consequently, the development of alternative feedstocks that meet all or most of the following criteria has attracted considerable research attention: low cost, high oil content, low agricultural inputs, favorable fatty acid (FA) composition, compatibility with existing farm equipment and infrastructure, production in off-season from conventional commodity crops or in agriculturally undesirable lands, definable growth seasons, (6) European Committee for Standardization (CEN). AutomotiVe fuelsFatty acid methyl esters (FAME) for diesel engines-Requirement methods, EN 14214:2003; European Committee for Standardization (CEN): Brussels, Belgium, 2003. (7) Paulson, N. D.; Ginder, R. G. Working Paper 07-WP 448, Center for Agricultural and Rural DeVelopment; Iowa State University: Ames, IA, May 2007. (8) Retka-Schill, S. Biodiesel Mag. 2008, 5, 64–70. (9) Demirbas, A. Energy ConVers, Manage. 2006, 47, 2271–2282.

This article not subject to U.S. Copyright. Published 2009 by the American Chemical Society Published on Web 07/02/2009

Downloaded by USDA NATIONAL AGRICULTURAL LIB on September 9, 2009 | http://pubs.acs.org Publication Date (Web): July 2, 2009 | doi: 10.1021/ef900337g

4150

Energy & Fuels, Vol. 23, 2009

uniform rates of seed maturation, and viable markets for coproducts such as seed meal. Selected recent examples of alternative feedstocks reported in the scientific literature include desert date (Balanites aegyptica L.),10 moringa (Moringa oleifera L.),11 karanja (Pongamia pinnata L.),12 and pumpkin (Cucurbita pepo L.)13 seed oils, among numerous others.5 Field pennycress (Thlaspi arVense L.), also known as stinkweed or French-weed, is a winter annual belonging to the Brassicaceae family. Native to Eurasia but with an extensive distribution throughout temperate North America, field pennycress is highly adapted to a wide variety of climatic conditions.14 The Brassicaceae family is a prolific source of biodiesel feedstocks, as evidenced by canola (or rapeseed, Brassica napus L.) and recent reports on B. alba L.,15 B. carinata L.,16 Camelina satiVa L.,17 and Raphanus satiVus L.18 oils, among others. Generally considered to be an agricultural pest (weed), field pennycress has potential to serve in a summer/winter rotational cycle with conventional commodity crops (such as corn or soybean), thus not displacing existing agricultural production. Field pennycress is tolerant of fallow lands, requires minimal agricultural inputs (fertilizer, pesticides, water), is not part of the food chain, is compatible with existing farm infrastructure, and has high oil content (20-36 wt %).13,19-21 In addition, each plant may produce up to 15 000 seeds, and fields heavily infested with field pennycress are reported to yield up to 1345 kg of seed/ha.22 More recent results indicate that the yield from wild populations is in the range of 1120-2240 kg of seed/ha, which equates to around 600-1200 L of oil/ha versus 450 and 420-640 L/ha in the cases of soybean (SBO)21 and camelina (C. satiVa L.)23 oils, respectively. Defatted field pennycress seed meal cannot be used as an animal feed as a result of its high glucosinolate content, but other applications such as biofumigation have been reported.14 The objective of the present study was to prepare field pennycress oil methyl esters (FPME) and evaluate their properties, such as low temperature operability, cetane number, oxidative stability, kinematic viscosity, lubricity, and surface tension. Compositional characteristics such as fatty acid (FA) profile, free and total glycerol content, acid value, and tocopherol, phytosterol, sulfur, and phosphorus contents were of additional interest, along with a comparison with biodiesel fuel standards such as ASTM D6751 and EN 14214. Furthermore, as a member of the Brassicaceae family, field pennycress is (10) Chapagain, B. P.; Yehoshua, Y.; Wiesman, Z. Bioresour. Technol. 2009, 100, 1221–1226. (11) Rashid, U.; Anwar, F.; Moser, B. R.; Knothe, G. Bioresour. Technol. 2008, 99, 8175–8179. (12) Naik, M.; Meher, L. C.; Naik, S. N.; Das, L. M. Biomass Bioenerg. 2008, 32, 354–357. (13) Schinas, P.; Karavalakis, G.; Davaris, C.; Anastopoulos, G.; Karonis, D.; Zannikos, F.; Stournas, S.; Lois, E. Biomass Bioenerg. 2009, 33, 44– 49. (14) Vaughn, S. F.; Isbell, T. A.; Weisleder, D.; Berhow, M. A. J. Chem. Ecol. 2005, 31, 167–177. (15) Ahmad, M. Asian J. Chem. 2008, 20, 6402–6403. (16) Bouaid, A.; Martinez, M.; Aracil, J. Bioresour. Technol. 2009, 100, 2234–2239. (17) Frohlich, A.; Rice, B. Ind. Crops Prod. 2005, 21, 25–31. (18) Domingos, A. K.; Saad, E. B.; Wilhelm, H. M.; Ramos, L. P. Bioresour. Technol. 2008, 99, 1837–1845. (19) Moser, B. R.; Shah, S. N.; Winkler-Moser, J. K.; Vaughn, S. F.; Evangelista, R. L. Ind. Crops Prod. DOI: 10.1016/j.indcrop.2009.03.007. (20) Dolya, V. S.; Koreshchuk, K. E.; Shkurupii, E. N.; Kaminskii, N. A. Chem. Nat. Compd. 1976, 10, 447–449. (21) Marek, L. F.; Bingaman, B.; Gardner, C. A. C.; Isbell, T. 20th Annual Meeting of the Association for the AdVancement of Industrial Crops Book of Abstracts, American Association for the Advancement of Industrial Crops: Maricopa, Arizona, 2008; p 49. (22) Best, K. F.; McIntyre, G. I. Can. J. Plant Sci. 1975, 55, 279–292. (23) Sawyer, K. Biodiesel Mag. 2008, 5 (7), 82–87.

Moser et al.

expected to possess a high level of unsaturated FA with a significant amount contributed by erucic (13(Z)-docosenoic) acid and therefore may serve as a model biodiesel feedstock for oils with similar FA profiles. 2. Experimental Section 2.1. Materials. Field pennycress seeds were collected from a wild population in Peoria County, IL. Methyl erucate (>99%; methyl 13(Z)-docosenoate), methyl gondoate (>99%; methyl 11(Z)-eicosenoate), and fatty acid methyl ester standards were purchased from Nu-Chek Prep, Inc. (Elysian, MN). Tocopherol standards (g97% purity), as well as stigmasterol and 5R-cholestane, were obtained from Matreya, LLC (Pleasant Gap, PA). Campesterol and brassicasterol were purchased from Steraloids (Newport, RI). N,OBis(trimethylsilyl)fluoroacetamide with 1% trimethylchlorosilane (BSTFA + 1% TMCS) was purchased from Regis, Inc. (Morton Grove, IL). Each phytosterol standard was g97% purity. All other chemicals and reagents were obtained from Sigma-Aldrich Corp. (St. Louis, MO) and used as received. 2.2. Extraction of Field Pennycress Oil. Field pennycress seeds were cold pressed using a heavy-duty laboratory screw press (Model L250, French Oil Mill Machinery Co., Piqua, OH). The details of this expeller are available elsewhere.24 Once extracted, the crude oil was filtered to remove solid material. Quantification of oil content by hexane extraction of field pennycress seeds is described elsewhere.19 2.3. Pretreatment of Field Pennycress Oil. Acid-catalyzed pretreatment of field pennycress oil (FPO) with an initial acid value (AV) of 0.61 mg of KOH/g was accomplished in a 500 mL threenecked round-bottom flask connected to a reflux condenser and a mechanical magnetic stirrer set to 1200 rpm. Initially, FPO (200 g, 220 mL, 0.207 mol) and methanol (78 mL, 1.91 mol, 35 vol %) were added to the flask, followed by dropwise addition of sulfuric acid (conccentrated, 2.20 mL, 0.04 mol, 1.0 vol %). The contents were heated at reflux for 2 h. Upon cooling to room temperature (RT), the alcoholic phase was removed utilizing a separatory funnel. The oil phase was washed with distilled water (3 × 20 mL), which was followed by rotary evaporation under reduced pressure (20 mbar; 30 °C) to remove residual methanol. Finally, treatment with magnesium sulfate (MgSO4) afforded dried FPO (185.2 g, 93 wt %) with a final AV of 0.09 mg of KOH/g. 2.4. Methanolysis of Field Pennycress Oil. Transesterification of FPO was performed on both crude and acid-pretreated oils. Methanolysis was carried out in a 500 mL three-necked roundbottom flask connected to a reflux condenser and a mechanical magnetic stirrer set to 1200 rpm. Initially, FPO (180 g, 200 mL, 0.187 mol) and methanol (46 mL, 1.12 mol; 6:1 mol ratio) were added to the flask and heated to 60 °C (internal reaction temperature monitored by digital temperature probe), followed by addition of 0.50 wt % sodium methoxide (25 wt % in methanol). After 1.5 h, the reaction mixture was equilibrated to RT and transferred to a separatory funnel. The lower glycerol phase was removed by gravity separation (>2 h settling time) followed by removal of residual methanol by rotary evaporation under reduced pressure (20 mbar; 30 °C). The crude methyl esters were washed until a neutral pH was obtained with distilled water (4 × 20 mL) and dried with MgSO4 to provide FPME. The yields of FPME were 82 and 94 wt % from crude and acid-pretreated FPO, respectively. 1H NMR (500 MHz, CDCl3): δ 5.35 (m, 2H, vinyl), 3.67 (s, 3H, -OCH3), 2.79 (m, 2H, bis-allylic), 2.31 (t, 2H, allylic), 2.03 (m, 2H, R to ester), 1.62 (m, 2H, β to ester), 1.28 (m, 28H, methylene), 0.90 (t, 3H, methyl). 13C NMR (125 MHz, CDCl3): δ 174.30, 174.27, 131.95, 130.26, 130.20, 130.05, 130.00, 129.92, 129.90, 129.87, 129.83, 129.74, 128.27, 128.24, 128.04, 127.90, 127.73, 127.12, 51.40, 34.11, 34.10, 31.90, 31.52, 29.77-29.09 (multiplet of 15 peaks), 27.20, 27.16, 25.63, 25.52, 24.96, 22.67, 22.57, 20.54, 14.25, 14.09, 14.05. FT-IR (neat): 3008, 2923, 2853, 1742, 1462, 1435, 1362, 1245, 1195, 1169, 1117, 1100, 1015, 880, 843, 722 cm-1. (24) Evangelista, R. L. Ind. Crops Prod. 2009, 29, 189–196.

EValuation of Biodiesel from Field Pennycress Oil

Energy & Fuels, Vol. 23, 2009 4151

Downloaded by USDA NATIONAL AGRICULTURAL LIB on September 9, 2009 | http://pubs.acs.org Publication Date (Web): July 2, 2009 | doi: 10.1021/ef900337g

Table 1. Fatty Acid Composition (wt %) of Field Pennycress Oil (FPO) fatty acida

FPO

C14:0 C16:0 C16:1 9c C18:0 C18:1 9c C18:1 11c C18:2 9c, 12c C18:3 9c, 12c, 15c C20:0 C20:1 11c C20:2 11c, 14c C22:0 C22:1 13c C22:2 13c, 16c C22:3 13c, 16c, 19c C24:1 15c unknown (sum) ∑saturatedb ∑monounsaturatedc ∑polyunsaturatedd ∑C20+e

0.1 3.1 0.2 0.5 11.1 1.5 22.4 11.8 0.3 8.6 1.6 0.6 32.8 0.7 0.3 2.9 1.5 4.6 55.6 38.3 47.8

a

For example, C18:1 9c signifies an 18 carbon fatty acid chain with one cis (c) double bond located at carbon 9 (methyl 9Z-octadecenoate; methyl oleate). b ∑saturated ) C14:0 + C16:0 + C18:0 + C20:0 + C22:0. c ∑monounsaturated ) C16:1 + C18:1 + C20:1 + C22:1 + C24:1. d ∑polyunsaturated ) C18:2 + C18:3 + C20:2 + C22:2 + C22:3. e ∑C20+ ) C20:0 + C20:1 + C20:2 + C22:0 + C22:1 + C22:2 + C22:3 + C24:1.

2.5. Fatty Acid Profile by Gas Chromatography (GC). Fatty acid methyl esters (FAME) were separated (triplicates, means reported) using a Varian (Walnut Creek, CA) 8400 GC equipped with an FID detector and a SP2380 (Supelco, Bellefonte, PA) column (30 m × 0.25 mm i.d., 0.20 µm film thickness). The carrier gas was He at 1 mL/min. The oven temperature was initially held at 150 °C for 15 min, increased to 210 at 2 °C/ min, increased to 220 at 50 °C/min, and then held for 10 min. The injector and detector temperatures were 240 and 270 °C, respectively. FAME peaks were identified by comparison to the retention times of reference standards. The FA profile of FPME is reported in Table 1. 2.6. Free and Total Glycerol Determination by GC. Free and total glycerol determinations (Table 2) were made according to ASTM standard D658425 employing an Agilent (Santa Clara, CA) Model 7890A GC-FID equipped with a Model 7683B series injector and an Agilent D8-5HT (15 m × 0.32 mm i.d., 0.10 µm film thickness) column. 2.7. NMR and Fourier Transform Infrared (FT-IR) Spectroscopy. 1H and 13C NMR data were recorded using a Bruker AV500 spectrometer (Billerica, MA) operating at 500 MHz (125 MHz in the case of 13C NMR) using a 5-mm broadband inverse Z-gradient probe in CDCl3 (Cambridge Isotope Laboratories, Andover, MA) as solvent. FT-IR spectra were obtained on a Thermo-Nicolet Nexus 470 FTIR spectrometer (Madison, WI) with a Smart ARK accessory containing a 45 ZeSe trough in a scanning range of 650-4000 cm-1 for 64 scans at a spectral resolution of 4 cm-1. The 1H NMR spectrum of FPME is depicted in Figure 1. 2.8. Tocopherol Content by HPLC. Tocopherols were quantified (triplicate determinations, means reported) according to AOCS official method Ce 8-8926 using a hexane/2-propanol mobile phase on an Inertsil (Varian) silica column (5 µm, 150 Å, 250 mm × 4.6 (25) American Society for Testing and Materials. Standard test method for determination of free and total glycerin in B-100 biodiesel methyl esters by gas chromatography, ASTM D6584-08. In ASTM Annual Book of Standards; American Society for Testing and Materials: West Conshohocken, PA, 2008. (26) American Oil Chemists’ Society. Official Methods and Recommended Practices of the American Oil Chemists’ Society (Method AOCS Ce 8-89), 5th ed.; Firestone, D., Ed.; American Oil Chemists’ Society: Champaign, IL, 1999.

Table 2. Properties of Field Pennycress Methyl Esters (FPME) with Comparison to Standards ASTM D6751 EN 14214

FPME

acid value, mg of KOH/g

0.50 max

free glycerol, mass % total glycerol, mass % cloud point, °C pour point, °C cold filter plugging point, °C oxidative stability (110 °C), h kinematic viscosity, mm2/s -10 °C 0 °C 20 °C 40 °C lubricity (HFRR, 60 °C), µm sulfur, ppm phosphorus, mass % surface tension, mN/m 24 °C 40 °C cetane number

0.020 max 0.240 max report 3 min

0.50 max 0.04 (0.03)/ 0.02 (0.02)a,b 0.020 max 0.005 0.250 max 0.041 -10 (1) -18 (1) c -17 (1) 6 min 4.4 (0.1)

1.9-6.0 d 15 max 0.001 max

3.5-5.0 d 10 max 0.001 max

35.52 16.70 8.65 5.24 (0.01) 125 (3) 7 0.0000

47 min

51 min

31.0 (0.1) 29.6 (0.1) 59.8e

a Values in parentheses are standard deviations from the reported means (n ) 3; n ) 2 for lubricity; n ) 5 for surface tension). b Values represent AV of FPME prepared from crude (0.04) and pretreated (0.02) FPO. All subsequent data are for FPME prepared from crude FPO. c Variable by location and time of year. d Maximum wear scar values of 460 and 520 µm are specified in petrodiesel standards EN 590 and ASTM D975. e Derived cetane number.

mm i.d.), Varian HPLC Pro-Star Model 230 pump, Model 410 auto sampler, and Model 363 fluorescence detector using excitation and emission wavelengths of 290 and 330 nm, respectively. Peaks were identified by comparison to the retention times of reference standards. 2.9. Phytosterol Content by GC. Samples were saponified and phytosterols were extracted as previously described.27 After saponification, phytosterols were manually injected onto a Varian 3400 GC equipped with an FID and a Supelco SPB-1701 (30 m × 0.25 mm × 0.25 µm) capillary column. Helium was used as the carrier gas with a 1:50 injector split. The injector and detector temperatures were 270 and 290 °C, respectively. The column oven initial temperature was 250 °C for 0.5 min, increased at 10 °C/min to 270 °C and held for 27 min, and then increased at 10 °C/min to 280 °C and held for 3.5 min. Phytosterols were identified by comparison to the retention times (relative to the internal standard) of reference standards. Phytosterols without commercially available standards, such as δ5-avenasterol, were identified by their relative retention times compared to the literature,27 and by comparison with samples known to contain those phytosterols. Quantification (triplicates, means reported) was carried out by the internal standard method developed with available standards. For phytosterols with no available commercial standard, the response factor for β-sitosterol was used for quantification. 2.10. Fuel Properties of Methyl Esters. Cloud (CP, °C) and pour point (PP, °C) determinations were made following ASTM standards D577328 and D5949,29 respectively, using a Model PSA70S Phase Technology Analyzer (Richmond, BC, Canada). Cloud and pour points were rounded to the nearest whole degree (°C). For a greater degree of accuracy, PP measurements were done with a resolution of 1 °C instead of the specified 3 °C increment. Cold filter plugging point (CFPP, °C) was measured in accordance with (27) Dutta, P. C.; Norme´n, L. J. Chromatogr., A 1998, 816, 177–184. (28) American Society for Testing and Materials. Standard test method for cloud point of petroleum products (constant cooling rate method), ASTM D5773-07. In ASTM Annual Book of Standards; American Society for Testing and Materials: West Conshohocken, PA, 2007. (29) American Society for Testing and Materials. Standard test method for pour point of petroleum products (automatic pressure pulsing method), ASTM D5949-01. In ASTM Annual Book of Standards; American Society for Testing and Materials: West Conshohocken, PA, 2001.

Downloaded by USDA NATIONAL AGRICULTURAL LIB on September 9, 2009 | http://pubs.acs.org Publication Date (Web): July 2, 2009 | doi: 10.1021/ef900337g

4152

Energy & Fuels, Vol. 23, 2009

Moser et al.

Figure 1. 1H NMR spectrum of field pennycress oil methyl esters. Table 3. Physical Properties of Crude Field Pennycress Oil crude FPOa oil content (wt %) Gardner color cloud point, °C pour point, °C oxidative stability (110 °C), h kinematic viscosity, mm2/s 25 °C 40 °C 100 °C viscosity index specific gravity 25 °C 40 °C lubricity (60 °C), µm acid value, mg of KOH/g sulfur, ppm phosphorus, mass %

29.0 10 -25 (1)b -28 (1) 5.0 (0.1) 70.01 (0.04) 40.97 (0.03) 9.39 (0.01) 224 0.913 (0.001) 0.904 (0.001) 125 (5) 0.61 (0.03)/0.09 (0.03)c 2 0.0002

a From ref 19 with the exception of Gardner color, sulfur, and phosphorus data. b Values in parentheses are standard deviations from the reported means (n ) 3; n ) 2 in the case of lubricity). c Values represent AV before (0.61) and after (0.09) acid-catalyzed pretreatment.

ASTM standard D637130 utilizing a Model FPP 5Gs ISL Automatic CFPP Analyzer provided by PAC, L.P. (Houston, TX). All experiments were run in triplicate, and mean values are reported (Tables 2 and 3). Kinematic viscosity (υ, mm2/s) was measured with a CannonFenske viscometer (Cannon Instrument Co., State College, PA) following ASTM standard D445 (Tables 2 and 3).31 Lubricity (duplicates, means reported; Tables 2 and 3) was measured at 60 °C ((1 °C) according to ASTM standard D607932 using a high-frequency reciprocating rig (HFRR) lubricity tester (PCS Instruments, London, England) via Lazar Scientific (Granger, IN). Reported wear scars (µm) were the result of measuring the maximum lengths of the x- and y-axes of each wear scar using a Prior Scientific (Rockland, MA) Epimat Model M4000 microscope, followed by calculating the average of these maximum values. (30) American Society for Testing and Materials. Standard test method for cold filter plugging point of diesel and heating fuels, ASTM D6371-05. In ASTM Annual Book of Standards; American Society for Testing and Materials: West Conshohocken, PA, 2005. (31) American Society for Testing and Materials. Standard test method for kinematic viscosity of transparent and opaque liquids (and calculation of dynamic viscosity), ASTM D445-06. In ASTM Annual Book of Standards; American Society for Testing and Materials: West Conshohocken, PA, 2006. (32) American Society for Testing and Materials. Standard test method for evaluating lubricity of diesel fuels by high frequency reciprocating rig (HFRR), ASTM D6079-04. In ASTM Annual Book of Standards; American Society for Testing and Materials: West Conshohocken, PA, 2004.

Oxidative stability (induction period, IP, h) was determined (triplicates, means reported; Tables 2 and 3) at 110 °C with a correction factor of 1.5 °C utilizing a Metrohm USA, Inc. (Riverview, FL) Model 743 Rancimat instrument according to the standard EN 14112.33 Acid value (AV, mg of KOH/g) was measured (triplicates, means reported; Tables 2 and 3) as described in AOCS official method Cd 3d-6334 using a Metrohm 836 Titrando (Westbury, NY) autotitrator equipped with a Model 801 stirrer and a Solvotrode electrode. The official method was modified for scale to use 2 g of sample and 0.02 M KOH. The titration end point was automatically determined and visually verified using a phenolphthalein indicator. Surface tension (γ, mN/m) was determined (five times; means reported; Table 2) at 24 ( 1 and 40 ( 1 °C with a Sita t60 bubble pressure tensiometer (Dresden, Germany). Dilutions and temperature control were handled by a Cat Ingenieurbu¨ro M. Zipperer GmbH (Staufen, Germany) M26 stir plate and a Model µ10MC buret. A bubble lifetime of at least 4 s was used so that dynamic effects were not a factor in the measurements. The instrument was calibrated using pure water. Additionally, the surface tensions of several organic solutions were measured and found to agree with literature values. Cetane numbers (CN) were determined as derived cetane numbers (DCN) (Table 2) by Southwest Research Institute (San Antonio, TX) utilizing an Ignition Quality Tester (IQT) following ASTM standard D6890.35 Results generated by ASTM D6890 generally correlate with CN determination by ATSM D613. Sulfur (S, ppm) and phosphorus (P, mass %) were measured (Tables 2 and 3) by Magellan Midstream Partners, L.P. (Kansas City, KS) according to ASTM standards D545336 and D4951,37 respectively. Gardner color (Table 3) was measured on a Lovibond 3-Field Comparator from Tintometer, Ltd. (Salisbury, England) using (33) European Committee for Standardization (CEN). Fat and oil deriVatiVes. Fatty acid methyl esters (FAME). Determination of oxidatiVe stability (accelerated oxidation test), EN 14112:2003; European Committee for Standardization (CEN): Brussels, Belgium, 2003. (34) American Oil Chemists’ Society. Official Methods and Recommended Practices of the American Oil Chemists’ Society (Method AOCS Cd 3d-63), 5th ed.; Firestone, D., Ed.; American Oil Chemists’ Society: Champaign, IL, 1999. (35) American Society for Testing and Materials. Standard test method for determination of ignition delay and derived cetane number (DCN) of diesel fuel oils by combustion in a constant volume chamber, ASTM 689008. In ASTM Annual Book of Standards; American Society for Testing and Materials: West Conshohocken, PA, 2008. (36) American Society for Testing and Materials. Standard test method for determination of total sulfur in light hydrocarbons, spark ignition engine fuel, diesel engine fuel, and engine oil by ultraviolet fluorescence, ASTM D5453-08b. In ASTM Annual Book of Standards; American Society for Testing and Materials: West Conshohocken, PA, 2008.

EValuation of Biodiesel from Field Pennycress Oil

Energy & Fuels, Vol. 23, 2009 4153

3.1. Composition and Physical Properties of Field Pennycress Oil. The oil content of dried field pennycress seeds was 29.0 wt %, which was in agreement with the range reported previously.19-21 The primary FA detected in FPO was erucic acid (C22:1 13Z; 32.8 wt %; Table 1), which is typical among members of the Brassicaceae family.16,19,20,39,40 Other FA of significance included linoleic (C18:2 9Z,12Z; 22.4 wt %), linolenic (C18:3 9Z,12Z,15Z; 11.8 wt %), oleic (C18:1 9Z; 11.1 wt %), gondoic (C20:1 11Z; 8.6 wt %), palmitic (C16:0; 3.1 wt %), and nervonic (C24:1 15Z; 2.9 wt %) acids. The overall level of saturated FA was low (4.6 wt % total), with palmitic acid comprising the majority of the saturated constituents. Data derived from the 1H NMR spectrum of the methyl esters synthesized here (Figure 1) according to procedure in the literature41 showing 12.7% FA with ω-3 unsaturation, 25.0% FA with two double bonds, and 59.0% FA with one double bond confirmed this result. The percentage of free fatty acids (FFA) in crude FPO was relatively low, as evidenced by an AV of 0.61 mg of KOH/g (Table 3). The color of the crude oil was 10 as measured by the Gardner scale (1 is lightest, 18 is darkest). Field pennycress oil primarily contained R- (714 ppm) and γ-tocopherols (126 ppm), with the β- (6 ppm) and δ-homologues (5 ppm) present at significantly lower levels (Table 4). The total tocopherol content was 851 ppm, which was higher than that reported for common crude commodity vegetable oils such as palm (642 ppm, combined), sunflower (546 ppm), and safflower

(413 ppm) oils, with the notable exception of soybean.42 The combined tocopherol contents of refined, bleached, and deodorized (RBD) SBO and soybean oil methyl esters (SME) prepared from crude SBO were 75719 and 130143 ppm, respectively. The principle phytosterols in FPO were sitosterol (3.88 mg/g) and campesterol (3.00 mg/g), with brassicasterol (0.76 mg/g), avenasterol (0.44 mg/g), cholesterol (0.27 mg/g), and stigmasterol (0.21 mg/g) accounting for the remaining phytosterol content (Table 4). The combined phytosterol concentration was 8.55 mg/g, which was higher than that previously found for RBD SBO (4.29 mg/g).19 As was the case here, brassicasterol normally comprises 5-20% of total phytosterols in oils from members of the Brassicaceae family (9% for FPO).44,45 Although seed meals from members of the Brassicaceae family are known to contain significant quantities of sulfurcontaining glucosinolates, crude FPO was largely free of sulfur content (2 ppm; Table 3) as a result of the polar nature of glucosinolates and their decomposition products.14 Crude FPO contained a very low amount of phosphorus (0.0002 mass %; Table 3). As summarized in a previous report, crude FPO exhibited kinematic viscosities at 25, 40, and 100 °C of 70.01, 40.97, and 9.39 mm2/s, resulting in a viscosity index of 224 (Table 3).19 The CP and PP values of FPO were -25 and -28 °C, respectively, indicating excellent low temperature operability versus SBO19 as a result of the low saturated FA content.46 The oxidative stability of crude FPO, as determined by the Rancimat method (EN 14112), was 5.0 h. The specific gravities (25 and 40 °C) and lubricity values (Table 3) were in the typical range reported for vegetable oils.19,20 3.2. Preparation of Methyl Esters from Field Pennycress Oil. Field pennycress oil was subjected to homogeneous base-catalyzed transesterification to afford FPME employing classic reaction conditions described previously.5,11,47,48 The yield of FPME (82 wt %) prepared from crude FPO was relatively low. Under ideal conditions, base-catalyzed transesterifications normally proceed to near quantitative yield.5,47-49 The cause was the FFA content of crude FPO, as measured by AV (0.61 mg KOH/g). Free fatty acids react with homogeneous base catalysts such as sodium methoxide to form soap (sodium salt of FA) and methanol (or water in the case of sodium hydroxide), thus irreversibly quenching the catalyst and reducing product yield.49 To improve the yield of FPME, sulfuric acid catalyzed pretreatment of crude FPO with methanol was conducted prior to base-catalyzed transesterification to reduce the AV of FPO to 0.09 mg of KOH/g (Table 3) following reaction conditions described previously.11,12,48,50,51 Subsequent transesterification of acid-pretreated FPO afforded FPME in excellent (94 wt %) yield. The product prepared from crude FPO also easily met the specifications for free and total glycerol in both the ASTM D6751 and EN 14214 biodiesel standards with values of 0.005 and 0.041 mass %, respectively (Table 2).

(37) American Society for Testing and Materials. Standard test method for determination of additive elements in lubricating oils by inductively coupled plasma atomic emission spectroscopy, ASTM D4951-06. In ASTM Annual Book of Standards; American Society for Testing and Materials: West Conshohocken, PA, 2006. (38) American Oil Chemists’ Society. Official Methods and Recommended Practices of the American Oil Chemists’ Society (Method AOCS Td 1a-64), 5th ed.; Firestone, D., Ed.; American Oil Chemists’ Society: Champaign, IL, 1999. (39) Matthaus, B.; Vosmann, K.; Pham, L. Q.; Aitzetmuller, K. J. Am. Oil Chem. Soc. 2003, 80, 1013–1020. (40) Dorado, M. P.; Ballesteros, E.; Lopez, F. J.; Mittelbach, M. Energy Fuels 2004, 18, 77–83. (41) Knothe, G.; Kenar, J. A. Eur. J. Lipid Sci. Technol. 2004, 106, 88–96.

(42) Frankel, E. N. Lipid Oxidation; The Oily Press: Bridgewater, 2004; p 225. (43) Moser, B. R. Eur. J. Lipid Sci. Technol. 2008, 110, 1167–1174. (44) Kochhar, S. P. Prog. Lipid Res. 1983, 22, 161–188. (45) Phillips, K. M.; Ruggio, D. M.; Toivo, J. I.; Swank, M. A.; Simpkins, A. H. J. Food Compos. Anal. 2002, 15, 123–142. (46) Moser, B. R. Energy Fuels 2008, 22, 4301–4306. (47) Moser, B. R.; Haas, M. J.; Winkler, J. K.; Jackson, M. A.; Erhan, S. Z.; List, G. R. Eur. J. Lipid Sci. Technol. 2007, 109, 17–24. (48) Freedman, B.; Pryde, E. H.; Mounts, T. L. J. Am. Oil Chem. Soc. 1984, 61, 1638–1643. (49) Tiwari, A. K.; Kumar, A.; Raheman, H. Biomass Bioenerg. 2007, 31, 569–575. (50) Lotero, E.; Liu, Y.; Lopez, D. E.; Suwannakarn, K.; Bruce, D. A.; Goodwin, J. G., Jr. Ind. Eng. Chem. Res. 2005, 44, 5353–5363.

Table 4. Tocopherol (ppm) and Phytosterol Contents (mg/g) of Field Pennycress Oil and the Corresponding Methyl Esters

Downloaded by USDA NATIONAL AGRICULTURAL LIB on September 9, 2009 | http://pubs.acs.org Publication Date (Web): July 2, 2009 | doi: 10.1021/ef900337g

FPOa

FPME

R-tocopherol β-tocopherol δ-tocopherol γ-tocopherol ∑tococ

Tocopherols 714 (7)b 6 (0) 126 (3) 5 (0) 851 (6)

527 (3) 6 (1) 105 (3) 5 (0) 644 (2)

cholesterol brassicasterol campesterol stigmasterol sitosterol avenasterol ∑phytod

Phytosterols 0.27 (0.01) 0.76 (0.01) 3.00 (0.07) 0.21 (0.01) 3.88 (0.07) 0.44 (0.02) 8.55 (0.17)

0.24 (0.01) 0.68 (0.03) 2.76 (0.06) 0.19 (0.01) 3.57 (0.09) 0.42 (0.01) 7.87 (0.17)

a Data for field pennycress oil is from ref 19. b Values in parentheses are standard deviations from the reported means. c ∑toco ) sum of tocopherols. d ∑phyto ) sum of phytosterols.

AOCS official method Td 1a-64.38 Specific gravity and viscosity index (Table 3) were collected according to AOCS official method Cc 10a-25 and ASTM standard D2270, respectively, as described previously.19

3. Results and Discussion

Downloaded by USDA NATIONAL AGRICULTURAL LIB on September 9, 2009 | http://pubs.acs.org Publication Date (Web): July 2, 2009 | doi: 10.1021/ef900337g

4154

Energy & Fuels, Vol. 23, 2009

The 1H NMR and FT-IR spectra of FPME were qualitatively similar to spectra of FAME reported elsewhere.11,41,47 For example, FPME contained a methyl ester moiety that was prominently indicated in the 1H NMR spectrum (Figure 1) by a strong singlet at around 3.67 ppm and in the FT-IR spectrum by a strong carbonyl signal at 1742 cm-1. Furthermore, the carbonyl and methyl ester carbons of the ester moiety are indicated in the 13C NMR spectrum by characteristic signals at 174.30 and 51.40 ppm, respectively. Methanolysis of crude FPO to afford FPME resulted in reductions in tocopherol and phytosterol contents of 24 and 8%, respectively (Table 4). The amount of total tocopherols was reduced from 851 to 644 ppm, and the combined phytosterol content was reduced from 8.55 to 7.87 mg/g (Table 4). These results are not surprising, as tocopherols and phytosterols are nonpolar and should remain soluble in hydrophobic materials such as biodiesel as opposed to partitioning into polar glycerolic or aqueous phases during purification. Retention of tocopherol content is beneficial, as these minor constituents serve as inhibitors of oxidation.43 3.3. Properties of Field Pennycress Oil Methyl Esters. The properties of FPME are summarized in Table 2 along with relevant fuel specifications from the biodiesel standards ASTM D6751 and EN 14214. For the sake of consistency, all properties listed in Table 2 and discussed below are from FPME prepared from crude as opposed to acid-pretreated FPO. The CN determined as DCN of FPME was 59.8 (Table 2), which well exceeded the minimum limits of 47 and 51 prescribed in ASTM D6751 and EN 14214, respectively. The relatively high DCN of FPME can be explained by the presence of methyl esters of erucic and gondoic acids. In the course of this work, the DCN of neat methyl erucate and methyl 11(Z)eicosenoate (methyl gondoate) were determined for the first time and were found to be 74.2 and 73.2, respectively, complementing data on other fatty acid alkyl esters.52 The DCN of FPME is then well-explained taking into consideration the amounts and DCN of the other major FAME, with the DCN of methyl oleate being in the range 56-59, that of methyl linoleate being 38.2, and that of methyl linolenate being even lower at 22.7.53 The kinematic viscosity of FPME was 5.24 mm2/s at 40 °C (Table 2), which was within the range specified in ASTM D6751 but was higher than the maximum specification in EN 14214 (Table 2). The high kinematic viscosity of FPME was largely a result of the presence of longer-chain (C20+) FAME. For example, the kinematic viscosity of methyl erucate has been reported as 7.33 mm2/s and that of methyl gondoate as 5.77 mm2/s.54 As these two species, together with smaller amounts of other FAME in the C20-C24 range, comprised approximately 48% of the FA profile (Table 1) of FPO, the relatively high kinematic viscosity is well-explained. The kinematic viscosities of the methyl esters of other major components of FPME are 4.38 mm2/s for methyl palmitate, 4.51 mm2/s for methyl oleate, 3.65 mm2/s for methyl linoleate, and 3.14 mm2/s for methyl linolenate.54 Kinematic viscosity data of FPME at temperatures below 40 °C (20, 0, and -10 °C) are also given in Table 2. Blending FPME with less viscous biodiesel fuels represents a potential strategy to satisfy the EN 14214 kinematic viscosity specification. (51) Naik, M.; Meher, L. C.; Naik, S. N.; Das, L. M. Biomass Bioenerg. 2008, 32, 354–357. (52) Knothe, G.; Matheaus, A. C.; Ryan, T. W., III. Fuel 2003, 82, 971– 975. (53) Knothe, G.; Bagby, M. O.; Ryan, T. W., III. SAE Tech. Pap. Ser. 1997, 971681. (54) Knothe, G.; Steidley, K. R. Fuel 2005, 84, 1059–1065.

Moser et al.

The low temperature properties of FPME were determined by CP, PP, and CFPP. As seen in Table 2, FPME provided CP, PP, and CFPP values of -10, -18, and -17 °C, respectively. The relatively low level of saturated FAME contained in FPME (4.6 wt %; Table 1) was in part attributed to the low temperature operability of FPME.46 In addition, unsaturated FAME have markedly lower melting points (mp) than the corresponding saturated analogues. For instance, the melting points of methyl esters of stearic (C18:0), oleic, linoleic, and linolenic acids are 39, -20, -35, and -57 °C, respectively.55 Furthermore, the melting points of methyl gondoate and erucate are -45 and -34 °C.55,56 The oxidative stability of FPME was 4.4 h, as measured by the Rancimat method (IP; EN 14112). Addition of antioxidants or blending with more oxidatively stable feedstocks would be necessary to satisfy the oxidative stability requirement (IP > 6 h) in EN 14214. FPME was acceptable according to the less stringent specification (IP > 3 h) in ASTM D6751. It should be noted that the IP of FPME was in excess of IP reported for individual unsaturated FAME,57 suggesting that native tocopherols (Table 4) were in part responsible for the oxidative stability of FAME.43 Both ASTM D6751 and EN 14214 restrict AV to a maximum value of 0.50 mg of KOH/g. The AV of FPME was easily within the specified limit with a value of 0.04 mg of KOH/g. FPME prepared from acid-pretreated FPO displayed an AV of 0.02 mg of KOH/g. The wear scar generated by FPME by the high-frequency reciprocating rig (HFRR) lubricity method ASTM D6079 (60 °C) was 125 µm (Table 2). Lubricity (ASTM D6079) is not specified in ASTM D6751 or EN 14214, but it is included in the petrodiesel standards ASTM D975 and EN 590 with maximum wear scars of 520 and 460 µm, respectively. Fuels with poor lubricity can cause failure of diesel engine parts that rely on lubrication from fuels, such as fuel pumps and injectors.2,3,58 As expected, the lubricity of FPME was considerably below the maximum limits set forth in the aforementioned petrodiesel standards, which was in agreement with several previous studies indicating that biodiesel possessed inherent lubricity.58-60 Sulfur content is limited in ASTM D6751 and EN 14214 to maximum values of 15 and 10 ppm, respectively. The sulfur content of FPME was 7 ppm (Table 2), which was below the specified maximum limits but higher than that for FPO (2 ppm). Anthropogenic introduction of sulfur to FPME by the use of a sulfur-containing drying agent (MgSO4) during purification (see section 2.4) was attributed to the small increase in sulfur content observed here. Phosphorus content is also limited in ASTM D6751 and EN 14214 to a maximum value of 0.001 mass %. FPME contained no phosphorus. Although surface tension is not specified in either ASTM D6751 or EN 14214, it is nevertheless an important property that affects fuel atomization in combustion chambers in diesel engines.61 The surface tensions (at 24 and 40 °C) of FPME were nearly identical (Table 2) to the values reported in the literature (55) Anonymous. Dictionary Section. In The Lipid Handbook, 3rd ed.; Gunstone, F. D., Harwood, J. L., Dijkstra, A. J., Eds.; CRC Press: Boca Raton, 2007; pp 444-445. (56) Chang, S. P.; Rothfus, J. A. J. Am. Oil Chem. Soc. 1996, 73, 403– 410. (57) Moser, B. R. J. Am. Oil Chem. Soc. 2009, 86, 699-706. (58) Knothe, G.; Steidley, K. R. Energy Fuels 2005, 19, 1192–1200. (59) Moser, B. R.; Cermak, S. C.; Isbell, T. A. Energy Fuels 2008, 22, 1349–1352. (60) Suarez, P. A. Z.; Moser, B. R.; Sharma, B. K.; Erhan, S. Z. Fuel 2009, 88, 1143–1147. (61) Ejim, C. E.; Fleck, B. A.; Amirfazli, A. Fuel 2007, 86, 1534–1544.

EValuation of Biodiesel from Field Pennycress Oil

for SME, indicating that adequate fuel atomization should not be an issue with FPME.62

Downloaded by USDA NATIONAL AGRICULTURAL LIB on September 9, 2009 | http://pubs.acs.org Publication Date (Web): July 2, 2009 | doi: 10.1021/ef900337g

4. Conclusions Biodiesel was prepared in high yield from field pennycress oil by alkali-catalyzed transesterification with methanol both before and after acid pretreatment. Fuel properties such as low temperature operability, cetane number, kinematic viscosity, oxidative stability, lubricity, surface tension, and others were determined. Excellent low temperature properties were observed, as indicated by low CP, PP, and CFPP values. The methyl esters provided a high derived cetane number as a result of the high contents of methyl esters of erucic and gondoic acids (32.8 and 8.6 wt %). Also reported for the first time are the derived cetane numbers of methyl esters of erucic and gondoic acids, which were 74.2 and 73.2, respectively. As a result of the high content of methyl esters of erucic, gondoic, and other acids with 20 or (62) Doll, K. M.; Moser, B. R.; Erhan, S. Z. Energy Fuels 2007, 21, 3044–3048.

Energy & Fuels, Vol. 23, 2009 4155

more carbons, FPME exhibited a kinematic viscosity in excess of the EN 14214 limit but within the ASTM D6751 specification. Blending with less viscous feedstocks would ameliorate the high viscosity of FPME. The oxidative stability of FPME was acceptable according to the limit contained in ASTM D6751, but not EN 14214. Consequently, the addition of antioxidant additives or blending with more stable feedstocks would be necessary to provide acceptable oxidative stability values according to the more stringent EN 14214 limit. The lubricity, surface tension, AV, and sulfur as well as phosphorus contents of FPME were satisfactory and within accepted limits, where applicable. Thus, FPME represents an acceptable substitute for petrodiesel, as FPME compares favorably to most biodiesel fuel specifications. Acknowledgment. The authors acknowledge Dr. Jill K. WinklerMoser for collection of phytosterol data as well as Benetria N. Banks and Kevin R. Steidley for excellent technical assistance. Dr. Karl Vermillion is acknowledged for acquisition of NMR spectra. EF900337G

Supplied by the United States Department of Agriculture, National Center for Agricultural Utilizaton Research, Peoria, Illinois