The effect of genotype and environment on biodiesel quality prepared from Indian mustard (Brassica juncea) grown in Australia

Industrial Crops and Products 48 (2013) 124–132

Contents lists available at SciVerse ScienceDirect

Industrial Crops and Products journal homepage: www.elsevier.com/locate/indcrop

The effect of genotype and environment on biodiesel quality prepared from Indian mustard (Brassica juncea) grown in Australia M.A. Wilkes a , I. Takei a , R.A. Caldwell a , R.M. Trethowan b,∗ a b

Department of Plant and Food Sciences, Faculty of Agriculture and Environment, The University of Sydney, NSW 2006, Australia Plant Breeding Institute, Faculty of Agriculture and Environment, The University of Sydney, NSW 2006, Australia

a r t i c l e

i n f o

Article history: Received 4 February 2013 Received in revised form 27 March 2013 Accepted 2 April 2013 Keywords: Biodiesel Genotype Environment Mustard Fatty acid

a b s t r a c t Two experiments were conducted in north-western New South Wales, Australia to determine the effect of genotype (G), growing site (S) and year (Y) on the suitability of Indian mustard (Brassica juncea) as a biodiesel feedstock. The first experiment analyzed the effect of growing environment on six mustard genotypes while the second experiment analyzed the effect of sowing on the same genotypes across two seasons. The results demonstrate that late sowing forced maturity of the seed and decreased the yield whilst early sowing resulted in economically viable seed yields (>1.3 t/ha). The oil content of the seed ranged from 34 to 39.8% and the main fatty acids present in the oil were oleic (C18:1) and linoleic acid (C18:2) in both experiments. The main factor that impacted on the fatty acid profile in a single season was the seed genotype while in the second experiment the growing year and interactions between year and the other parameters had a major impact on the fatty acid profile. The main fatty acids affected by the growing year were oleic, linoleic and erucic (C22:1). Oleic and linoleic acids were inversely correlated with erucic acid content which tended to be higher in cooler growing conditions. Two of the genotypes were processed into biodiesel and assessed for quality and the fuel met most requirements except for oxidation stability and kinematic viscosity. The relatively high concentration of polyunsaturated fatty acids was deemed to be responsible for the poor oxidation stability and higher amounts of erucic acid and glycerol would contribute to poor kinematic viscosity values. The mustard genotypes analyzed may prove to be both a viable break crop as well as providing a good feedstock for the establishment of a biodiesel industry in this area. Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved.

1. Introduction Biodiesel is a fuel prepared by transesterification of the longchain fatty acids contained in ‘non-fossilized’ oil feedstocks such as vegetables, seeds, animal fats, algae and used cooking oils. The substitution of traditional petroleum-based fuels with renewables such as biodiesel is seen as a long-term strategy that will provide both a secure and diverse range of energy sources while reducing greenhouse gas emissions. In addition to production from renewable feedstocks, biodiesel possesses other ‘green’ credentials such as low sulfur content, biodegradability, reduced toxicity, and generally lower emissions compared with current petroleum-based fuels (Jham et al., 2009; Moser, 2009). Biodiesel does have some disadvantages when compared to traditional fuels including poor low temperature operability and oxidative stability (Moser, 2009)

∗ Corresponding author. Tel.: +61 2 9351 8860. E-mail addresses: [email protected] (M.A. Wilkes), [email protected] (I. Takei), [email protected] (R.A. Caldwell), [email protected] (R.M. Trethowan).

which are mainly attributed to the fatty acids contained in the oil feedstock. In some cases the use of fuel additives may attenuate some of these problems however the use of appropriate feedstocks that possess better chemical properties may prove to be a sustainable alternative. Even though Australia can currently supply most of its transportation fuels, increased demand will see further pressure to develop alternative fuel sources. Renewable energy sources accounted for just over 5% of energy consumption in Australia in 2008/09 and biofuels accounted for only 8% of the renewable energy consumed (Schultz and Petchey, 2010). To improve the sustainability of energy resources the Australian federal government set an annual production target of 350 million L of biofuels. Production capacity for biodiesel reached 100 million L in 2010/11 and used mainly tallow, canola and recycled cooking oils as the feedstock (ACCC, 2011). Despite the advantages of biodiesel, the high cost of the oil is a constraint to developing a biodiesel industry; around 85–88% of production expenses are attributed to purchasing the feedstocks (Retka-Schill, 2008). It has been reported that high feedstock prices has led to the closure or production stoppages of some established Australian biodiesel plants (ACCC, 2010);alternatives to

0926-6690/$ – see front matter Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.indcrop.2013.04.016

M.A. Wilkes et al. / Industrial Crops and Products 48 (2013) 124–132

expensive commodity feedstocks are critical to the continued survival and expansion of the biodiesel industry. However, sourcing oil for biodiesel from crop land may compete with food and fiber production, potentially driving up the price of essential commodities. Therefore, if a biodiesel industry is to be sustainable a feedstock must be identified that; provides a source of energy equivalent in quality to fossil-based fuels; can be produced in large quantities; and does not directly compete with food and fiber crops for available arable land. Indian mustard (Brassica juncea L.) is an annual herbaceous oilseed-bearing plant belonging to the Brassicaceae family. Conventional breeding methods have focused on improving seed yield and fatty acid profile and mustard is now grown widely in South Asia for cooking oil, and on a small scale in Australia for condiment mustard production (Oram et al., 2005). However, it may prove to be a reliable crop for biodiesel production like other Brassica species that have already gained widespread acceptance as feedstocks for biodiesel production such as B. napus (canola), B. alba (white mustard), and B. carinata (Ethiopian mustard) (Jham et al., 2009; Moser, 2009). Besides its potential as a fuel source, mustard has agronomic benefits and can be used as a break crop in the drier areas of Australia where canola is poorly adapted. Field trials indicate competitive seed yields, a requirement for fewer inputs such as fertilizer, and improved resistance to shattering when compared to canola (Hocking et al., 1997; Angadi et al., 2000; Oram et al., 2005). When grown as a break crop, mustard may improve the yield of crops such as wheat and barley grown in the following winter season. According to Angus et al. (1991), growing mustard as a break crop increased the yield of a subsequent wheat crop by 30% compared to wheat grown after wheat. Break-crop benefits may also extend to a second successive wheat crop resulting in an average increase of 13% in grain yield (Kirkegaard et al., 1997). The rotational benefits of Brassica crops are due to the glucosinolates exuded from root tissues which are hydrolysed to form isothiocyanates that act as biofumigants suppressing the growth of disease-causing microorganisms (Angus et al., 1994; Kirkegaard and Sarwar, 1998). In the north western areas of NSW, B. juncea has been shown to effectively reduce the incidence of crown root rot caused by Fusarium pseudograminearum (Kirkegaard et al., 2003). Thus, the benefits of growing B. juncea are twofold – the root exudates act to break a disease cycle thereby improving cereal crop growth, and the oil can be processed into biodiesel. In the north west of NSW, the potential growing area available for B. juncea oil seed production is around 4 mha, however, the major impediment for establishment of this crop is a lack of suitable mustard varieties that provide both biodiesel quality and an economic return to growers. Hancock (2005) reports that yields would need to reach a minimum of 1.3 t/ha for mustard to be competitive as a feedstock for biodiesel production. However, the influence of genotype and environment on fatty acid profile in B. juncea, and hence biodiesel quality, is not well understood. The main biodiesel properties affected by fatty acid properties are the cetane number (CN), oxidative stability and cold flow properties. The CN is related to the combustion properties and longer, saturated fatty acids result in good cetane numbers (Ramos et al., 2009). Saturated fatty acids also have superior oxidative properties compared with polyunsaturated fatty acids of equivalent chain lengths; however saturated fatty acids result in biodiesel with poor cold flow properties (Moser, 2009). Therefore, good biodiesel quality requires a fine balance of fatty acids of appropriate chain length and saturation to provide sufficient combustion while maintaining oxidative stability and cold flow properties. It appears that a high concentration of oleic acid in the raw feedstock results in biodiesel that satisfies most of these parameters (Ramos et al., 2009). This project examines

125

the effect of environment on the yield and fatty acid profile of six mustard genotypes grown in replicated trials across northwestern NSW, Australia. 2. Materials and methods Fatty acids and biodiesel were prepared from the oilseed of two Indian mustard genotypes (Muscon BM11 and Hermola-805) grown in a preliminary study in 2008. The biodiesel was analyzed in a commercial laboratory and compared to the Australian standards as outlined in the Fuel Standard (Biodiesel) Determination 2003 which uses a combination of both ATSM and EN standards (http://www.comlaw.gov.au/Details/F2009C00146). These fatty acid profiles were also used as a standard for comparison of the samples prepared from the field trial samples. As fatty acid profile in the seed is controlled by the genetics of the plant as well as the growing environment (Hou et al., 2006) two separate experiments were then conducted to analyze the effect of environmental conditions on a range of Indian mustard genotypes and whether variation in the growing environment (location and/or sowing date and year), and/or genotype significantly impacts on the fatty acid profile of the seed and subsequent biodiesel quality prepared from the oils. In the first experiment six B. juncea genotypes were grown in a single season (2009) across five different locations in unique randomized complete block designs with two replications. In the second experiment the same oil seed genotypes were sown at a single location at two different sowing dates in two different seasons (2009 and 2010) in a randomized complete block design with three replications. 2.1. Oilseed genotypes and growth locations To determine the viability of a biodiesel industry six mustard genotypes were grown in experiments at five sites in north western NSW in 2009. The sites were Narrabri (early and late sown), Mungindi, Come-by-Chance and Nowley. Weather data, including rainfall and temperature, and crop growth parameters were recorded at Narrabri (Table 1; Fig. 1). The six mustard genotypes for fatty acid analysis were selected based on high yield and diverse genetic background. The second experiment was sown at Narrabri in 2010 at two planting dates to examine the effect of season on seed and oil yield, and fatty acid profile of the same six genotypes evaluated in the 2009 experiments. The six mustard genotypes were 65-3CSIRO*60–9/Pollen bulk; JM97.3*43B-2/Pollen bulk; JM97.3*5-7/Pollen bulk, Muscon BM11, Hermola-805 and Canadian #2. The canola check cultivar was not assessed for fatty acid profile as the grain yield was significantly lower than the selected mustard genotypes at all sites. Samples of grain were taken from each replicated plot from all genotypes at all locations. Oil content was determined using NIR using the canola calibration (Foss NIRSystems Inc.) and then extracted from the seed by cold pressing and stored at 4 ◦ C until the preparation of fatty acid methyl esters (FAME) for gas chromatography (GC) analysis. Oil from two separate field samples was prepared for each genotype. 2.2. Preparation of FAMEs An esterification reagent was prepared immediately before use by adding 2 g of NH4 Cl to methanol (60 mL) in a 150 mL Quickfit round bottom flask. Concentrated sulfuric acid (98%, 3 mL) was added, and the mixture refluxed for 15 min using a water condenser. The esterification reagent was then allowed to cool before use. Both of the replicate samples of oil were used to prepare duplicate laboratory FAME samples for GC analysis. Oils (0.1–0.2 g)

126

M.A. Wilkes et al. / Industrial Crops and Products 48 (2013) 124–132

Table 1 Planting locations and crop data for B. juncea grown in the 2008–10 seasons. Site and year

Cropping parameters

Narrabri 08 Come-by-Chance Mungindi Nowley Narrabri Early 09 Narrabri Late 09 Narrabri Early 10 Narrabri Late 10

Location

Sowing date

Flowering date

Harvest date

30.33◦ S, 149.77◦ E 30.19◦ S, 148.28◦ E 28.99◦ S, 149.98◦ E 29.98◦ S, 149.18◦ E 30.33◦ S, 149.77◦ E 30.33◦ S, 149.77◦ E 30.33◦ S, 149.77◦ E 30.33◦ S, 149.77◦ E

26/June/2008 15/May/2009 20/June/2009 30/June/2009 8/May/2009 15/June/2009 15/May/2010 20/June/2010

18–22/August/2008 NA NA NA 17–27/July/2009 31/July–7/August/2009 15–27/July/2010 5–10/August/2010

2/October/2008 20/September/2009 25/September/2009 1/October/2009 15/September/2009 20/September/2009 14/September/2010 16/September/2010

NA – not available. Southern hemisphere seasons; autumn, March–May; winter, June–August; spring, September–November.

Temperature (°C)

40

2008

2009

2.3. Statistical analysis Data from the duplicate laboratory FAME preparations were averaged for each separate oil sample and the mean values of fatty acids in the oil from each genotype are presented. Analysis of variance was performed using GENSTAT (Windows Discovery Edition 11) and genotype (G), site (S) and G × S effects estimated for grain yield and fatty acid percentage composition for all six cultivars. 3. Results and discussion 3.1. Effect of location and season on seed yield and oil content Flowering occurred 70 and 61 days after sowing in early sown mustard trials in 2009 and 2010 respectively. In contrast, flowering occurred 46 days after sowing for late sown mustard in both 2009 and 2010. The early sown genotypes reached maturity 130 and 122 days after sowing in 2009 and 2010, respectively, whereas the respective maturity of late sown materials was 97 and 88 days after sowing in 2009 and 2010 (Table 1). These results are similar to other studies on the growth of mustard in Australia where late sowing reduces the cropping cycle by forcing maturity (Hocking et al., 1997; Robertson et al., 2004; Gunasekera et al., 2006). The mean grain yield across the five different locations in 2009 ranged from 0.35 t/ha at Nowley to 1.64 t/ha at the early sown Narrabri site (Table 2). All the B. juncea breeding lines yielded in excess of the canola check variety at all sites demonstrating their superior adaptability in these drier northern growth locations. In comparing the Narrabri sites for difference between seasons (2009/10) and environment (early/late sowing) the grain yields ranged from 0.61 t/ha to 1.641 t/ha (Table 2). When the data were compared between seasons, the average yield was 1.12 t/ha in 2009 160

2010

35

140

30

120

25

100

20

80

15

60

10

40

5

20

0

0

Mean maximum temperature

Mean minimum temperature

Rainfall (mm)

were individually weighed into a Quickfit round bottom flask and methanolic NaOH (2.0 mL, 0.5 M) and small boiling chips added, and the mixture was refluxed for 10 min. Esterification reagent (6 mL) was added and the mixture refluxed for 3 min, followed by the addition of 5 mL hexane and further refluxing for 1 min. The cooled mixture was poured into a 25 mL volumetric flask. The round bottom flask was then washed with saturated NaCl solution and the wash solution transferred to the volumetric flask. More NaCl solution was added to the volumetric flask until the organic phase was wholly within the neck of the flask. The upper organic phase layer was then transferred to a small vial using a Pasteur pipette before being dried with anhydrous sodium sulfate. The dried hexane solution was then transferred to a 10 mL volumetric flask. A small quantity of hexane was added to wash the vial and then transferred into the 10 mL volumetric flask. Hexane was added to a final volume of 10 mL. The resultant FAMEs were stored in tightly capped vials at 4 ◦ C until GC analysis. The FAMEs were analyzed in a gas chromatograph (GC-17A, SHIMADZU, Kyoto, Japan) using a DB-WAX column of 30 m × 0.32 mm id × 0.25 ␮m (J&M Scientific, Folsom, California, USA) with flame ionization detection. The oven temperature was programmed for 18 min at an initial column temperature of 190 ◦ C. The temperature was maintained at 190 ◦ C for 8 min before rising to 210 ◦ C at the rate of 4 ◦ C/min. The temperature was then held at 210 ◦ C for 5 min. The injector and detector were set at 220 ◦ C. Helium was used as the carrier gas. Peaks were identified by comparison of retention times to those from a FAME (C8–C24) standard mixture (Supelco, 18918-1AMP). Absolute amounts were determined using individual standards (methyl palmitate, Sigma P-0750; methyl stearate, Sigma S-5376; methyl oleate, Aldrich 31111; methyl linoleate, Sigma L1876) together with the relative response factors given on the FAME (C8–C24) data sheet.

Total rainfall (mm)

Fig. 1. Mean temperature and rainfall data for Narrabri in the 2008–10 growing seasons (http://www.bom.gov.au/climate/data/).

M.A. Wilkes et al. / Industrial Crops and Products 48 (2013) 124–132

127

Table 2 Effect of different locations and seasons on B. juncea seed and oil yield. B. juncea genotype

Growing location and season Come-by-Chance (09)

Seed yield (t/ha) 65-3CSIRO*60-9 JM97.3*43B-2 JM97.3*5-7 Muscon BM11 Hermola-805 Canadian #2 Mean LSD (5%) Oil yield (%) 65-3CSIRO*60-9 JM97.3*43B-2 JM97.3*5-7 Muscon BM11 Hermola-805 Canadian #2 Mean

Mungindi (09)

Nowley (09)

1.30 1.30 1.19 1.09 1.32 1.13

0.50 0.63 0.64 0.67 0.43 0.46

0.39 0.37 0.36 0.21 0.43 0.34

2.00 1.62 1.56 1.58 1.66 1.44

0.57 0.64 0.65 0.81 0.55 0.43

2.14 1.34 1.29 1.06 0.89 0.64

1.54 1.21 1.16 0.82 1.57 0.69

1.22 0.48

0.56 0.24

0.35 0.28

1.64 1.01

0.61 0.51

1.23 1.20

1.16 0.96

37.7 39.8 38.2 37.6 38.3 38.6 38.4

Narrabri (09 Early)

36.4 37.7 37.0 35.6 37.3 37.2 36.9

Narrabri (09 Late)

35.4 37.4 36.2 34.7 37.0 36.1 36.1

Narrabri (10 Early)

34.7 36.5 35.3 34.9 36.0 37.5 35.8

Narrabri (10 Late)

35.3 36.4 35.4 34.3 36.5 34.8 35.5

LSD, least significant difference of means at 5% level. Oil yields were not available for Nowley and Mungindi.

and 1.18 t/ha in 2010. The early sown experiments produced a mean seed yield of 1.42 t/ha and late sown experiments a significantly reduced yield of 0.89 t/ha (P < 0.001). These results are similar to those of Hocking et al. (1997), Robertson et al. (2004) and Gunasekera et al. (2006) who reported similar grain yield and sowing effects in Australian field trials of Indian mustard. Later sowing reduces the days to flowering and maturity which reduces the amount of dry matter accumulation. Robertson et al. (2004) calculated that yield decreased by 2.75–6.54% for every week sowing was delayed. However, many studies show that individual genotypes may differ in their response to sowing date and other conditions during grain filling. Stress conditions resulting from reduced rainfall and increasing temperatures during crucial developmental stages may also reduce seed yield in Indian mustard (Hocking et al., 1997; Angadi et al., 2000; Robertson et al., 2004; Gunasekera et al., 2006; Adak et al., 2011). This is evident in the late sown mustard experiments in 2009 and 2010 where the mean yield was higher in the cooler and wetter 2010 season compared to the drier and warmer conditions of 2009 (Fig. 1). Of the five individual sites in 2009 only Narrabri (early) and Come-by-Chance produced seed yields (>1.3 t/ha) deemed to be economically viable for sustainable biodiesel production (Hancock, 2005). In the experiments conducted at Narrabri, across sowing dates and seasons, only in the late sown trials of 2009 did any genotype fail to achieve acceptable grain yield. Oil yield ranged from 34.3 to 39.8% (Table 2) and these values are within ranges reported in other Australian studies (Hocking et al., 1997; Robertson et al., 2004; Gunasekera et al., 2006). When oil contents were averaged across sites and seasons, the lowest mean oil content was observed in late sown experiments at Narrabri 2010, and the highest oil content in late sown experiments at Narrabri 2009. Overall there seemed to be little variation due to differences in growing locations or seasons. Robertson et al. (2004) found that oil content ranged from around 32 to 39% and that post-anthesis conditions had a significant impact on oil content which decreased in a linear fashion as the sowing date was delayed. They calculated that every 1 ◦ C increase in temperature during grain filling resulted in a 0.84% loss in oil and this could be further impacted under water deficit. Pritchard et al. (1999) found that canola grown in southern areas of Australia (Victoria) had higher oil contents if the season was cooler and there was more variation in oil content due to season than growing location. They reported a weak correlation between seed yield and spring

rainfall. Oram et al. (2005) have reported that newer mid-season maturing varieties can yield between 1.7 and 1.9 t/ha and contain 42.5 and 43.5% oil in other regions of New South Wales. Clearly there is scope to further improve mustard to consistently meet basic grower and processor requirements. 3.2. Effect of location and season on B. juncea fatty acids GC analysis of the FAMEs indicated that seven fatty acids were consistently present in the oils extracted from B. juncea (Table 3). When the FAMEs were compared across the five different growing locations in the 2009 season it was observed that the main fatty acids were C18:1, oleic acid (30.70%); C18:2, linoleic acid (25.03%); C18:3, linolenic acid (14.27%); C22:1 erucic acid (13.15%); C20, arichidic acid (8.82%); C16, palmitic acid (3.41%); and C18, stearic acid (1.91%). Lignoceric acid (C24, 0.62%) was only a minor constituent of the extracted oils and was not always detected. The ratio of unsaturated to saturated fatty acids was 5.63. The ANOVA (Table 5) revealed that in a single year (2009) at various growing locations (experiment 1) only genotype (G) had a significant impact on the fatty acid profile of the different oils. Neither growing site (S) nor the interaction between site and genotype resulted in any significant sources of variation (Table 5). The fatty acids most affected by the different genetic backgrounds were oleic acid, arichidic acid and erucic acid (P < 0.001) and linoleic acid (P < 0.01). Palmitic acid, oleic acid, linolenic acid and lignoceric acid were significantly influenced by genotype, though to a lesser extent than the other fatty acids detected in the oils (P < 0.05). The temperatures across the sites in 2009 did not vary greatly and the main difference in the growing locations can be attributed to sowing date and some slight variation in rainfall. The order of fatty acids in the B. juncea varieties grown across different seasons at the same location (Narrabri, early/late sowing, 2009/10; experiment 2) was slightly different when compared with the mean contents of the samples grown in 2009 only. The main fatty acids were oleic acid (25.36%), linoleic acid (22.26%), erucic acid (15.89%), linolenic acid (14.43%), arichidic acid (6.68%), palmitic acid (3.61%), stearic acid (2.33%) and lignoceric acid (0.46%) (Table 4). The ratio of unsaturated to saturated fatty acids was 5.96. The change in order of relative abundance in fatty acids between 2009 and 2010 is due to the increased synthesis of erucic acid with a concomitant decrease in linolenic acid in the 2010 season. Despite the slight change in order of abundance of some

128

M.A. Wilkes et al. / Industrial Crops and Products 48 (2013) 124–132

Table 3 Effect of growth location on B. juncea fatty acid composition in a single season (2009). B. juncea variety

Fatty acid (wt%)

Site

C16

C18

C18:1

C18:2

C18:3

65-3CSIRO*60-9 Narrabri early Narrabri late Come-by-Chance Mungindi Nowley

4.14 3.83 3.59 3.59 3.66

2.19 2.00 2.22 1.82 2.16

36.08 32.25 40.97 30.10 39.93

25.43 24.72 27.28 24.84 27.24

13.56 13.70 15.69 14.43 14.65

JM97.3*43B-2 Narrabri early Narrabri late Come-by-Chance Mungindi Nowley

3.12 3.22 3.21 3.73 3.11

1.78 1.63 1.72 1.79 1.68

23.53 25.33 23.95 25.56 25.33

23.28 23.61 22.49 24.17 22.44

JM97.3*5-7 Narrabri early Narrabri late Come-by-Chance Mungindi Nowley

3.70 3.63 3.42 3.57 3.73

1.96 2.22 2.05 1.76 1.84

28.25 26.32 29.94 28.61 37.99

Muscon BM11 Narrabri early Narrabri late Come-by-Chance Mungindi Nowley

3.51 3.24 3.13 4.02 3.21

2.13 1.55 2.14 2.14 2.01

Hermola-805 Narrabri early Narrabri late Come-by-Chance Mungindi Nowley

3.42 3.08 3.12 3.39 1.73

Canadian #2 Narrabri early Narrabri late Come-by-Chance Mungindi Nowley

C20

C22:1

C24

7.00 8.62 4.46 7.51 5.30

7.32 9.93 4.79 13.99 6.02

0.00 0.27 0.32 0.66 0.36

14.41 14.58 15.20 14.65 14.73

11.72 10.11 12.23 10.63 12.16

19.30 19.07 18.35 17.81 17.82

1.02 1.21 1.01 0.00 0.56

27.02 14.99 20.76 24.09 22.93

15.08 12.60 9.50 14.15 12.67

8.28 16.65 11.63 8.76 6.89

15.28 18.90 15.67 16.95 12.02

0.00 1.18 0.48 0.24 0.50

29.62 30.59 30.07 30.81 31.87

27.24 27.16 25.85 27.67 25.82

14.38 15.13 14.77 15.19 14.80

8.36 7.76 8.72 7.88 8.61

11.96 13.88 12.40 10.90 11.39

0.89 0.34 0.94 0.00 0.46

1.88 1.73 1.76 1.87 1.61

26.74 26.99 24.49 29.71 25.33

24.92 25.35 23.73 25.04 22.80

14.26 16.09 14.68 14.93 14.18

10.28 8.93 11.42 5.87 11.95

15.23 14.36 17.92 12.69 17.57

1.12 1.17 0.55 1.14 1.13

3.27 3.90 3.27 3.26 3.41

2.00 1.98 1.90 1.92 1.97

35.43 37.13 36.40 34.88 36.90

28.44 29.68 26.82 27.77 27.24

13.94 15.81 14.42 12.96 12.94

6.75 5.02 6.63 7.32 7.11

8.77 5.75 9.25 9.89 9.26

0.42 0.37 0.72 0.78 0.81

Mean LSD (5%) Muscon BM11a Hermola-805a

3.41 1.1 3.95 4.07

1.91 0.5 2.50 1.39

30.70 9.4 40.98 17.84

25.03 6.5 29.30 13.66

14.27 3.2 15.24 8.97

8.82 5.7 2.15 7.25

13.15 8.2 2.52 41.15

0.62 0.9 0.26 0.52

Sum of saturated FA Sum of MUFA Sum of PUFA

14.76 43.85 39.30

LSD is for genotype × site interaction. FA, fatty acid; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid. a Commercial analysis of 2008 oilseed crop.

fatty acids these profiles and contents are similar to those reported in the literature (Lionneton et al., 2002, 2004; Sinha et al., 2007; Jham et al., 2009). However, some of these studies report much higher erucic acid levels than those reported here. Reducing the level of erucic has been an aim of many breeding programs and it is possible that older and/or wild genotypes contain erucic acid up to 40% of the fatty acids in Indian mustard whilst newer genotypes contain very little, or no, erucic acid (Sinha et al., 2007). When the different genotypes were analyzed over different growing seasons (Y) at a single site (Narrabri, 2009/10 and early/late sowing) there were more complex interactions impacting on the fatty acid profiles detected (Table 6). The fatty acids that were affected by multiple sources of variation were oleic acid (affected by Y, G and Y × G), arichidic acid (Y, S, Y × S, Y × G, Y × S × G), erucic acid (Y, G, Y × S × G) and lignoceric acid (Y, G). Some of these sources of variation were more significant than others. There were fewer sources of variation in the stearic acid (Y) and linoleic acid (Y) content, however, the impact of Y was highly significant (P < 0.001). The amount of palmitic acid was relatively

stable regardless of the genotype, sowing date or year; only Y × G interactions resulted in any significant variation (P < 0.01), and linolenic acid was not significantly affected by variation in any of the crop parameters or their interactions. There are few studies reporting the effects of genotype and environment interactions on fatty acid composition of B. juncea seed oil. The main interactions in the experimental results presented here appear to be growing season and the major fatty acids affected were oleic, linoleic and erucic acids. The amount of oleic and linoleic acid was often lower in the cooler 2010 season compared with the amount of erucic acid which tended to be higher in the 2010 season. Higher amounts of oleic acid in Brassica sp. grown in warmer conditions have been reported by Yaniv et al. (1995) and Pritchard et al. (1999). Correlations between the different fatty acids and yield were calculated and some significant interactions were observed (Table 7). The most significant correlations were between C18:1 and C18:2, C18:1 and C22:1, and C18:2 and C22:1. Erucic acid was significantly negatively correlated with C18:1 and C18:2 reflecting

M.A. Wilkes et al. / Industrial Crops and Products 48 (2013) 124–132

129

Table 4 Effect of different seasons and sowing dates on fatty acid composition of B. juncea grown at Narrabri. Brassica juncea variety

Fatty acid (wt%)

Season and sowing

C16

C18

C18:1

C18:2

C18:3

65-3CSIRO*60-9 09 Early 09 Late 10 Early 10 Late

4.14 3.83 3.31 3.61

2.19 2.00 2.47 2.66

36.08 32.25 15.81 18.79

25.43 24.72 17.34 15.99

13.56 13.70 13.59 11.35

JM97.3*43B-2 09 Early 09 Late 10 Early 10 Late

3.12 3.22 4.05 3.95

1.78 1.63 2.65 3.10

23.53 25.33 23.07 23.38

23.28 23.61 19.88 22.99

JM97.3*5-7 09 Early 09 Late 10 Early 10 Late

3.69 3.63 3.21 3.84

1.96 2.22 2.45 2.91

28.25 26.32 14.14 21.69

Muscon BM11 09 Early 09 Late 10 Early 10 Late

3.51 3.24 4.49 4.23

2.13 1.55 2.77 3.53

Hermola-805 09 Early 09 Late 10 Early 10 Late

3.42 3.08 3.54 3.43

Canadian #2 09 Early 09 Late 10 Early 10 Late

C20

C22:1

C24

7.00 8.62 1.88 4.18

7.32 9.93 21.46 20.34

0.00 0.27 0.30 0.00

14.41 14.58 14.09 15.92

11.72 10.11 2.64 9.89

19.30 19.07 13.86 18.34

1.02 1.21 0.34 0.00

27.02 14.99 16.26 20.38

15.08 12.60 13.77 14.66

8.28 16.65 1.97 2.42

15.28 18.90 29.34 17.31

0.00 1.18 0.00 0.05

29.62 30.59 30.11 30.51

27.24 27.16 21.51 24.34

14.38 15.13 15.24 17.16

8.36 7.76 1.38 1.57

11.96 13.88 9.14 8.53

0.89 0.34 0.30 0.29

1.88 1.73 3.10 2.84

26.74 26.99 19.42 19.11

24.92 25.35 19.38 18.80

14.26 16.09 14.27 13.79

10.28 8.93 2.09 5.96

15.23 14.36 20.44 19.14

1.12 1.17 0.38 0.78

3.27 3.90 3.33 3.52

2.00 1.98 2.57 1.82

35.43 37.13 16.11 18.20

28.44 29.70 16.52 19.08

13.94 15.81 13.46 15.38

6.75 5.02 2.51 14.44

8.80 5.75 20.11 23.64

0.42 0.37 0.62 0.00

Mean LSD (5%)

3.61 0.78

2.33 1.02

25.36 4.82

22.26 7.20

14.43 2.98

6.68 5.22

15.89 6.01

0.46 0.66

Sum of saturated FA Sum of MUFA Sum of PUFA

13.08 41.25 36.69

LSD is for genotype × site × year interaction. FA, fatty acid; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid.

Table 5 Mean squares from analysis of variance demonstrating the impact of location on variation in yield and fatty acids of B. juncea in a single season (2009). Source of variation

df

S G S×G

4 5 20

Yield

C16

C18

C18:1

C18:2

C18:3

C20

C22:1

C24

ms

P

ms

P

ms

P

ms

P

ms

P

ms

P

ms

P

ms

P

ms

P

3.47 0.04 0.03

*** ns ns

0.42 0.79 0.25

ns * ns

0.04 3.20 0.05

ns * ns

20.66 216.98 15.58

ns *** ns

7.05 51.38 8.82

ns ** ns

0.86 6.20 2.80

0.84 * 0.37

0.52 40.59 9.47

ns *** ns

3.72 170.72 10.87

ns *** ns

0.14 0.60 0.27

ns * ns

S, site; G, genotype; df, degrees of freedom; ms, mean squares; *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant.

Table 6 Mean squares from analysis of variance demonstrating the impact of sowing date and season on variation in yield and fatty acids of B. juncea grown at a single location (Narrabri). Source of variation

df

Y S G Y×S Y×G S×G Y×S×G

1 1 5 1 5 5 5

Yield

C16

C18

C18:1

C18:2

C18:3

C20

C22:1

C24

ms

P

ms

P

ms

P

ms

P

ms

P

ms

P

ms

P

ms

P

ms

P

0.02 6.03 0.50 1.27 0.67 0.18 0.32

ns *** ns * * ns ns

0.50 0.01 0.24 0.06 0.72 0.15 0.10

ns ns ns ns ** ns ns

8.03 0.00 0.14 0.23 0.41 0.14 0.24

*** ns ns ns ns ns ns

970.73 11.94 65.15 16.51 124.83 2.90 10.64

*** ns *** ns *** ns ns

400.77 0.00 28.72 38.51 25.27 10.05 20.33

*** ns ns ns ns ns ns

0.06 3.11 5.36 0.21 2.13 2.80 2.61

ns ns ns ns ns ns ns

285.67 78.46 15.33 37.86 35.30 7.87 28.06

*** ** ns * ** ns **

319.20 0.76 82.95 10.23 117.90 9.62 31.02

*** ns *** ns *** ns *

2.07 0.00 0.52 0.31 0.26 0.25 0.21

*** ns ** ns ns ns ns

Y, year; S, site (early/late); G, genotype; df, degrees of freedom; ms, mean squares; *P < 0.05; **P < 0.01; ***P < 0.001; ns, not significant.

130

M.A. Wilkes et al. / Industrial Crops and Products 48 (2013) 124–132

Table 7 Correlations between fatty acids and grain yield for B. juncea grown in 2009/10 at Narrabri. Fatty acid

C16 C18 C20 C24 C18:1 C18:2 C18:3 C22:1 Yield * **

C16

C18

C20

C24

C18:1

C18:2

C18:3

C22:1

0.4608 −0.6491 −0.5638 0.7228 0.3136 0.0669 −0.6215 0.2753

−0.5039 0.185 0.1344 0.0208 0.1707 −0.1245 0.4027

0.3604 −0.712 −0.3339 0.0383 0.7479 −0.2686

−0.2873 0.2782 0.5691 0.1761 −0.2228

0.8096* 0.4371 −0.9648** −0.2071

0.7982 −0.8241** −0.4688

−0.3739 −0.6864

0.0887

P < 0.05. P < 0.01.

the biosynthetic pathway of these fatty acids and the amount and/or activity of the enzymes involved in each step in the pathway. This was also evident in the preliminary 2008 study of the Muscon BM11 and Hermola-805 oils (Table 3). The strong negative relationship between C18:1 and C22:1 has been reported in other studies (Lionneton et al., 2002; Mahmood et al., 2003; Sinha et al., 2007). Mahmood et al. (2003) reported the identification of two QTLs that had a major impact on the amount of these fatty acids in B. juncea and that the genes acted in an opposite manner causing the content of erucic acid to be low while oleic and linoleic acids were high and vice versa. The synthesis of fatty acids in seeds is the result of the interaction of a number of enzymes that act to elongate and desaturate the fatty acid chain before it is esterified with glycerol for deposition as a storage lipid. In Brassicas, the enzyme fatty acyl-ACP thioesterase B (FatB) that releases shorter-medium chain fatty acids such as C16 and C18, is poorly expressed (Sinha et al., 2007), hence the low amounts of these fatty acids in the Indian mustard oils. For medium-to-long chain fatty acids oleic acid appears to act as a semi ‘branch point’ where oleic acid may be released from the fatty acid synthase complex by the action of fatty acyl thioesterase A (FatA) and either further desaturated to form the polyunsaturated fatty acids (PUFAs) linoleic acid and/or linolenic acid through the action of fatty acid desaturases (FAD2 and FAD3);or elongated by fatty acid elongase (FAE) to form gondoic acid (C20:1) and then erucic acid (Sinha et al., 2007; Suresha et al., 2012). Suresha et al. (2012) speculated that the genes fad and fae are inter-regulated and suppressing fad2 can result in up-regulation of fae. Sinha et al. (2007) did find that efforts to decrease erucic acid through transgenic approaches were successful but the effects on other fatty acids in the seed were not always as expected. This demonstrates that the genetics and biochemistry that underpins fatty acid synthesis in Brassica spp. is complex and requires further investigation if this oil is to become a reliable fuel source in the face of predicted variable climates. This complexity is evident in the results reported for Hermola-805 which varied in the amount of erucic acid across the three different seasons indicating the genes involved in this pathway, in this genotype, are strongly affected by the growing environment. It is generally accepted that the PUFA content of many seeds increases as growing temperature decreases due to higher activity of desaturase enzymes at lower temperatures leading to higher amounts of PUFA (Hou et al., 2006). However it is not a strict relationship for all oilseeds (Baud and Lepiniec, 2010) and this relationship was not observed in this study or others on B. juncea (Yaniv et al., 1995; Pritchard et al., 1999). The cooler and wetter growing conditions of the 2010 season did not consistently promote the synthesis of linoleic and/or linolenic acid above the rate recorded for these fatty acids in the warmer 2009 season. In general, the amount of PUFA in Hermola-805 was lower in 2008 compared with the other seasons while the FAME profile of Muscon BM11 was

more similar to the other G analyzed in 2009 and 2010. It may be possible that the minimum temperatures were not low enough in the 2008 and 2010 seed development period to significantly promote the activity of the desaturase enzymes (Hou et al., 2006). Suresha et al. (2012) found that the expression of the fatty acid desaturase fad2 genes isolated from B. juncea were affected by temperature during seed development and were most active 20–30 days after flowering and at lower temperatures. The minimum temperatures recorded at the Narrabri trial sites at these crucial times after flowering were not greatly different which may explain why there was not a significant consistent increase in the PUFA content of all G in 2008 and 2010. 3.3. Effect of fatty acids on biodiesel quality Two mustard genotypes, Muscon BM11 and Hermola805, were processed into biodiesel using a BioMaster (www.bioworks.com.au) small scale processing plant and the biodiesel was analyzed against the Australian ‘Fuel Standard (Biodiesel) Determination 2003’ specifications in a commercial laboratory. For the remaining samples FAMEs were used as an indicator of biodiesel properties as the fatty acids present in the oil feedstock are a good indicator of biodiesel composition (Haagenson and Wiesenborn, 2011; Ramos et al., 2009). The biodiesel parameters were then related to fatty acid composition in these two genotypes as well as the profiles obtained for other varieties in both experiment 1 and experiment 2 (Table 8). The main biodiesel quality parameters affected by fatty acid composition are low temperature operability including properties such as cloud point (CP) and cold filter plugging point (CFPP), oxidative stability (OS), kinematic viscosity (KV), and cetane number (CN) (Ramos et al., 2009, Moser, 2009; Haagenson and Wiesenborn, 2011). The biodiesel met many of the specifications, however, both varieties were out of range for ester content, total glycerol, distillation T90 and OS and Hermola-805 was out of range for KV. Some of these parameters are related to the hydrolysis and esterification of the fatty acids contained in the oil and can be improved with more careful processing whilst the KV and OS are highly dependent on the fatty acid composition of the original oil. 3.3.1. Low temperature operability Parameters such as CP and CFPP are related to the melting point of the fatty acids contained in the biodiesel. The melting point is affected by both the length and the degree of saturation of the fatty acid chain. For fully saturated fatty acid chains the melting point increases as the chain length increases. For example the melting point of C20:0 > C18:0 > C16:0. Therefore biodiesel blends that contain higher proportions of saturated fatty acids, with higher melting points, will start to solidify and form a cloud at more ambient temperatures (Moser, 2009; Haagenson and Wiesenborn, 2011).

M.A. Wilkes et al. / Industrial Crops and Products 48 (2013) 124–132

131

Table 8 Biodiesel quality of B. juncea grown at Narrabri. Biodiesel parameter

Method

Specification

Hermola-805

Muscon BM11

Density (kg/m3 @ 15 ◦ C) Flash point (◦ C) Water and sediment (vol%) Sulfated ash (mass%) Acid number (mg KOH/g) Phosphorus (mg/kg) Sulfur (mg/kg) Free glycerol (mass%) Total glycerol (mass%) Ester content (mass%) Methanol content (mass%) Oxidation stability (h @ 110 ◦ C) Distillation T90 (◦ C vac) Cetane number Kinematic viscosity (mm2 /s @ 40 ◦ C)

ASTM D1298 ASTM D93 ASTM D2709 ASTM D874 ASTM D664 EN14107 ASTM D5453 ASTM D6584 ASTM D6584 EN 14103 EN 14110 EN 14112 ASTM D1160 ASTM D6890 ASTM D445

860.0–890.0 120 (min) 0.05 (max) 0.020 (max) 0.80 (max) 10 (max) 10 (max) 0.020 (max) 0.250 (max) 96.5 (min) 0.20 (max) 6 (min) 360 (max) 51.0 (min) 3.5–5.0

883.8 174 <0.05 0.003 0.15 <4 1.5 0.005 9.508 85.6 0.04 2.4 405.5 >61 7.081

888.1 184 <0.05 0.003 0.25 <4 3.7 0.003 0.497 91.2 <0.01 1.0 396.1 55.1 4.788

Figures in bold are outside specifications.

Biodiesel containing high amounts of long chain saturated fatty acids, particularly C16 and C18, generally have poor CFPP (Ramos et al., 2009). The degree of saturation and orientation of the double bonds in unsaturated fatty acids also impacts on the low temperature operability as the introduction of double bonds reduces the melting point. For example the melting point of C18:0 > C18:1 > C18:2 > C18:3 so high proportions of PUFA will improve the cold flow properties of the biodiesel (Ramos et al., 2009). If the orientation of the double bond is in the cis orientation the melting point is reduced compared with the trans isomer (Moser, 2009). Although low temperature operability parameters (CP, CFPP) were not measured in the commercial laboratory analysis it would be expected that Muscon BM11 biodiesel would provide better performance at lower temperatures based solely on the higher proportions of C18:1, C18:2 and C18:3 compared with the Hermola805 fatty acid profile (Table 8). Muscon BM11 also had less very long chain fatty acids which negatively affects the cloud point. When comparing the seasonal effects on fatty acid and potential biodiesel quality it would be expected that the oils from the seed grown in the warmer and drier 2009 season (early and late sowing) would produce biodiesel with better cold flow properties due to the relatively higher proportions of C18:1 + C18:2 compared with the oil from the 2010 season (early and late sowing). Year, or season, was the major factor contributing to variance in the C18:1 and C18:2 content in experiment 2 whereas variety was the major factor in experiment 1. Growing conditions were not greatly different in experiment 1 indicating that temperature and rainfall have a significant impact on biodiesel quality in regards to cold flow properties. Haagenson and Wiesenborn (2011) found correlations between fatty acid properties and CP but this differed depending on the experiment (bulked sample over a number of years or single variety grown at different locations).

3.3.2. Oxidative stability Oxidative stability is related to the resistance of the oil to oxidation/breakdown caused by heating. Saturated fatty acids are generally much more resistant to oxidation due to the lack of double bonds where oxidative reactions are initiated. Double bonds in the trans orientation are also more stable than the cis-isomers (Moser, 2009). High concentrations of linoleic and linolenic acids result in poor OS and biodiesel prepared from feedstocks rich in these fatty acids may require the addition of antioxidants to ensure storage stability (Ramos et al., 2009). The biodiesel prepared from Hermola-805 and Muscon BM11 failed to meet the OS specifications of 6 min (EN 14112). This is most likely due to the relatively

high proportions of cis isomers of C18:2 and C18:3 especially in Muscon BM11 which was comprised of almost 45% PUFA compared with Hermola-805 which had around 23% PUFA. The relatively high amount of erucic acid in the Hermola-805 sample would also have contributed to the OS as increasing chain length improves the stability (Moser, 2009). In experiment 2 it would be expected that the 2009 season, which resulted in seed richer in PUFA and lower in LCFA, would be less stable than biodiesel from the 2010 season that had generally higher C22:1 and lower C18:2 levels. Haagenson and Wiesenborn (2011) found that canola biodiesel stability was affected by growth year and location and Yaniv et al. (1995) found that cooler and wetter conditions favor the production of PUFA in Brassica (Yaniv et al., 1995). These trends were not evident in this study where PUFA decreased in the cooler, wetter season and the only variable that impacted on PUFA content was growing year (for C18:2 only) in experiment 2. In experiment 1 only the genotype had a significant impact on the PUFA content mainly due to the similar temperatures across the five growing locations in this experiment.

3.3.3. Kinematic viscosity Kinematic viscosity is related to the atomization of the fuel on injection and fuels with high KV result in poor combustion, excess emissions, engine deposits and deterioration of injectors (Refaat, 2009). The KV is affected by similar factors as those that affect the low temperature operability, namely, chain length, degree of saturation and orientation of double bonds in the fatty acid. As the chain length increases the KV increases due to lower fluidity of the fatty acids. The introduction of a double bond reduces the KV as the double bond increases the fluidity of the fatty acid. The cis orientation of double bonds reduces the KV but not to the same extent that cis/trans orientation affects melting point of fatty acid (Moser, 2009). The biodiesel prepared from Muscon BM11 had an acceptable KV however the biodiesel from Hermola-805 oil was above the maximum limit. The Hermola-805 oil had a high proportion of erucic acid which would increase the KV. On the other hand the Muscon BM11 oil consisted of little very long chain fatty acids and had almost 50% PUFA. Another factor to take into account is the very high total glycerol content of Hermola-805 (9.51%) which would adversely affect the KV (Knothe and Steidley, 2005). This is a processing deficiency and can be reduced with more complete hydrolysis and esterification of the fatty acids. The oils that had a similar fatty acid profile to Muscon BM11, mainly around 40% C18:1 and 45% PUFA and low very long chain fatty acids, would most likely meet KV specifications. Most of the varieties grown in the warmer 2009 season meet these profiles.

132

M.A. Wilkes et al. / Industrial Crops and Products 48 (2013) 124–132

3.3.4. Cetane number The CN is the fuel’s ignition delay time and is related to the combustion properties relative to cetane which is assigned a CN of 100. The higher the number the shorter the fuel ignition delay and the better the engine performance, especially cold start properties (Ramos et al., 2009). Higher CN are usually associated with saturation and LCFA (at the same degree of saturation) and poor CN results from the presence of PUFA and branched fatty acids (Ramos et al., 2009; Moser, 2009). Higher proportions of C16, C18, C18:1, C20:1 and C22:1 in the biodiesel result in higher CN (Ramos et al., 2009). In this study both of the varieties produced biodiesel with adequate CNs. Biodiesel produced from Hermola-805 oil had a higher CN than that prepared from Muscon BM11 probably because of the high amount of C16, C18:1, C22:1 in Hermola-805 compared with the high levels of PUFA in Muscon BM11 oil. As both oils produced biodiesel with adequate CN properties it is possible that this parameter is the least affected by variables such as variety, growth season and location as, even though the FAME profiles were quite different for these parameters, they both met the criteria. It also appears that if the level of C18:1 is low in these varieties the level of C22:1 is higher and vice versa enabling the fuel to have sufficient ignition qualities. 4. Conclusions The genotypes grown in the different sites generally produced adequate seed and oil yields demonstrating that B. juncea holds promise as a commercially viable break crop in the warmer and drier areas where canola is not well adapted. The main fatty acid present in the oils was oleic acid and high concentrations of this fatty acid in biodiesel are generally associated with acceptable fuel properties. However, the oleic acid concentration was still low compared with levels reported for canola and further efforts should focus on increasing the amount of this fatty acid in B. juncea seed. In this study, the biodiesel analysis demonstrated that the two genotypes met most criteria for quality although the OS and KV need to be improved. In the short term, improvements in quality could be achieved by better processing and/or the addition of antioxidants. In the long term this study has shown that there are B. juncea genotypes that can be further developed for a successful biodiesel industry in this area. Acknowledgments This authors wish to recognize the financial contribution of Mahnheim Ltd., the Australian Research Council and the Rural Industries Research and Development Corporation, and the technical assistance of Mr. Graeme Rapp and Miss Pascaline Afa in the completion of this work. References ACCC, 2010. Monitoring of the Australian Petroleum Industry 2010. Australian Competition and Consumer Commission. http://www.accc.gov.au/ content/index.phtml/itemId/961783 ACCC, 2011. Monitoring of the Australian Petroleum Industry 2011. Australian Competition and Consumer Commission. http://www.accc.gov.au/content/ index.phtml/itemId/1020827 Adak, T., Chakravarty, N.V.K., Muthukumar, M., Deshmukh, P.S., Joshi, H.C., Katiyar, R.K., 2011. Evaluation of biomass and thermal energy utilization efficiency of oilseed Brassica (Brassica juncea) under altered microenvironments. Biomass Bioenergy 35, 2254–2267.

Angadi, S.V., Cutforth, H.W., Miller, P.R., McConkey, B.G., Entz, M.H., Brandt, S.A., Volkmar, K.M., 2000. Response of three Brassica species to high temperature stress during reproductive growth. Can. J. Plant Sci. 80, 693–701. Angus, J.F., van Herwaarden, A.F., Howe, G.N., 1991. Productivity and break crop effects of winter-growing oilseeds. Aust. J. Exp. Agr. 31, 669–677. Angus, J.F., Gardner, P.A., Kirkegaard, J.A., Desmarchelier, J.M., 1994. Biofumigation: isothiocyanates released from Brassica roots inhibit growth of the take-all fungus. Plant Soil 162, 107–112. Baud, S., Lepiniec, L., 2010. Physiological and developmental regulation of seed oil production. Prog. Lipid Res. 49, 235–249. Gunasekera, C.P., Martin, L.D., Siddique, K.H.M., Walton, G.H., 2006. Genotype by environment interactions of Indian mustard (Brassica juncea L.) and canola (B. napus L.) in Mediterranean-type environments 1. Crop growth and seed yield. Eur. J. Agron. 25, 1–12. Haagenson, D.M., Wiesenborn, D.P., 2011. Impact of North Dakota growing location on canola biodiesel quality. J. Am. Oil Chem. Soc. 88, 1439–1445. Hancock, N., 2005. Biodiesel Overview on Global Production and Policy. Department of Agriculture and Food, Government of Western Australia. http://www. agric.wa.gov.au/objtwr/imported assets/content/sust/biofuel/200511 bdworldoverview.pdf Hocking, P.J., Kirkegaard, J.A., Angus, J.F., Gibson, A.H., Koetz, E.A., 1997. Comparison of canola, Indian mustard and Linola in two contrasting environments. I. Effects of nitrogen fertilizer on dry-matter production, seed yield and seed quality. Field Crop. Res. 49, 107–125. Hou, G., Ablett, G.R., Pauls, K.P., Rajcan, I., 2006. Environmental effects on fatty acid levels in soybean seed oil. J. Am. Oil Chem. Soc. 83, 759–763. 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. Kirkegaard, J.A., Hocking, P.J., Angus, J.F., Howe, G.N., Gardner, P.A., 1997. Comparison of canola, Indian mustard and Linola in two contrasting environments. II. Break-crop and nitrogen effects on subsequent wheat crops. Field Crop. Res. 52, 179–191. Kirkegaard, J.A., Sarwar, M., 1998. Biofumigation potential of brassicas. Plant Soil 201, 71–89. Kirkegaard, J.A., Holland, J., Moore, K., Simpfendorfer, S., Bambach, R., Marcroft, S., Hollaway, G., 2003. Effect of previous crops on crown rot infection and yield of wheat. In: 11th Australian Agronomy Conference: Solutions for a New Environment, Geelong, February 2003. Knothe, G., Steidley, K.R., 2005. Lubricity of components of biodiesel and petrodiesel. The origin of biodiesel lubricity. Energy Fuels 19, 1192–1200. Lionneton, E., Ravera, S., Sanchez, L., Aubert, G., Delourme, R., Ochatt, S., 2002. Development of an AFLP-based linkage map and localization of QTLs for seed fatty acid content in condiment mustard (Brassica juncea). Genome 45, 1203–1215. Lionneton, E., Aubert, G., Ochatt, S., Merah, O., 2004. Genetic analysis of agronomic and quality traits in mustard (Brassica juncea). Theor. Appl. Genet. 109, 792–799. Mahmood, T., Ekuere, U., Yeh, F., Good, A.G., Stringam, G.R., 2003. RFLP linkage analysis and mapping genes controlling the fatty acid profile of Brassica juncea using reciprocal DH populations. Theor. Appl. Genet. 107, 283–290. Moser, B.R., 2009. Biodiesel production, properties, and feedstock. In Vitro Cell. Dev Biol. Plant 45, 229–266. Oram, R.N., Kirk, J.T.O., Veness, P.E., Hurlstone, C.J., Edlington, J.P., Halsall, D.M., 2005. Breeding Indian mustard (Brassica juncea (L.) Czern.) for cold-pressed, edible oil production – a review. Aust. J. Agr. Res. 56, 581–596. Pritchard, F.M., Norton, R.M., Eagles, H.A., Salisbury, P.A., Nicolas, M., 1999. The effect of environment on Victorian canola quality. In: 10th International Rapeseed Congress, Canberra Australia, 1999. Ramos, M.J., Fernández, C.M., Casas, A., Rodríguez, L., Pérez, L., 2009. Influence of fatty acid composition of raw materials on biodiesel properties. Bioresour. Technol. 100, 261–268. Refaat, A.A., 2009. Correlation between the chemical structure of biodiesel and its physical properties. Int. J. Environ. Sci. Technol. 6, 677–694. Retka-Schill, S., 2008. Walking a tightrope. Biodiesel Mag. 5 (3), 64–70. Robertson, M.J., Holland, J.F., Bambach, R., 2004. Response of canola and Indian mustard to sowing date in the grain belt of north-eastern Australia. Aust. J. Exp. Agr. 44, 43–52. Schultz, A., Petchey, R., 2010. Energy Update 2010. Australian Bureau of Agricultural and Resource Economics – Bureau of Rural Sciences. http://143.188.17.20/data/warehouse/pe abarebrs98000001/ energyUpdate2010ProdConsTrade1973to2009.pdf Sinha, S., Jha, J.K., Maiti, M.K., Basu, A., Mukkopadhyay, U.K., Sen, S.K., 2007. Metabolic engineering of fatty acid biosynthesis in Indian mustard (Brassica juncea) improves nutritional quality of seed oil. Plant Biotechnol. J. 1, 185–197. Suresha, G.S., Rai, R.D., Santha, I.M., 2012. Molecular cloning, expression analysis and growth temperature dependent regulation of a novel oleate desaturase gene (fad2) homologue from Brassica juncea. Aust. J. Crop Sci. 6, 296–308. Yaniv, Z., Schafferman, D., Zur, M., 1995. The effect of temperature on oil quality and yield parameters of high- and low-erucic acid Cruciferae seeds (rape and mustard). Ind. Crop. Prod. 3, 247–251.