Journal of Oil Palm & The Environment - JOPEH

25

Journal of Oil Palm & The Environment An official publication of the Malaysian Palm Oil Council (MPOC)

Journal of Oil Palm & The Environment 2011, 2:25-37 doi:10.5366/jope.2011.04

LCI comparisons of five vegetable oils as feedstock for biodiesel J.M. van Zutphen 1*, R.A. Wijbrans 1, Foo-Yuen Ng 2

Abstract A LCI study is carried out on rape seed, canola, palm oil, soy bean and sunflower for their potential to be used as feedstocks for biodiesel. Biodiesels made from them are shown to be produced with lower CO2eq emissions relative to fossil diesel. As such, the CO2eq emission reduction savings, using the carbon balance approach, ranges from 67% for soy to 53% for sunflower biodiesel. Using the energy balance approach, the energy use savings ranges from 79% for palm oil to 62% for sunflower. In all cases, they are above EU RED’s threshold value of 35% savings, needed for a feedstock to qualify as an acceptable biofuel feedstock in EU.

Keywords Carbon emission, energy balance, biodiesel, palm oil, rape seed, soy bean, sunflower, canola JOPE 2011, 2:25-37 1. Introduction

Vegetable oils have been used for decades in many food and oleochemical applications

1. As

such, demand for vegetable oils, both food grade and inedible, has risen steadily over time

2. In recent years, vegetable oils have also

become major feedstocks for biofuels3. The

demand for biofuels is the result of the growing concern about climate change and a global drive to reduce greenhouse gas emissions. Although the anthropogenic emissions at the tailpipe of a motor vehicle are clearly reduced by using biodiesel instead of fossil diesel, there is scanty information on the greenhouse gas emissions incurred in producing biofuels. Performing a Life Cycle Inventory (LCI) over the entire supply chain in the production of the vegetable oils up to the biodiesel stage, can provide such useful information.

The results would give a good indication to assist European biofuel producers and decision makers who are looking for feedstocks that are able to meet the minimum requirement of the European Union’s Renewable Energy Directive (EU RED)

4. This

study would also provide an opportunity to look into the carbon emission sources and potential areas of improvement to reduce the carbon emissions during the production of these biofuel crops.

2. Methodology

A LCI study is carried out on five vegetable oils, namely, palm oil, rape seed, soy bean, sunflower and canola. Consideration is given to the environmental performance (net carbon reduction and energy balance). The LCI inventories of these vegetable oils are then compiled and its net carbon and energy reduction compared against fossil diesel.

Address: 1CarbonCapital Solutions, Barbiershoeve 105, 7

326 DC Apeldoorn, The Netherlands 2Malaysian Palm Oil Council, 2nd Floor, Wisma Sawit, Lot

6, SS 6, Jalan Perbandaran, 47301 Kelana Jaya, Selangor, Malaysia. Email: R.A. Wijbrans * ([email protected]), Foo Yuen Ng ([email protected]) *Corresponding Author Published: 10 June 2011 Received: 17 March 2010 Accepted: 1 June 2011 This article is available from: http://www.jope.com.my

© 2011 CarbonCapital Solutions This is an Open Access article which permits unrestricted use, distribution and reproduction in any medium, provided

the original work is properly cited.

Review Open Access

mailto:[email protected]

mailto:[email protected]

http://www.jope.com.my/

26

Through the comparison, the environmental performance of each vegetable oil is determined by its net reduction percentage of carbon and energy.

2.1 Calculation and allocation methodology

The basic model for determining greenhouse gas reductions (or primary energy savings) by using biomass as a substitute is shown in Figure 1 below. The model is derived from standard LCA methods as described in ISO 14040:2006

5

.

Figure 1 - Model for determining carbon balance of biomass applications

27

In essence, the model entails the comparison between the emissions from the supply chain of the biomass and those from the supply chain (and use) of fossil fuels. The emissions and energy uses taken into account for both supply chains are the fossil fuel related or anthropogenic emissions and energy uses. In this study, the CO2eq emissions and energy use of the supply chain are expressed in tonnes CO2eq/tonne biodiesel and GJ/tonne biodiesel respectively.

2.2 Major Assumptions

2.2.1 Land use change

Emission sources related to land use from the general model have been left out of the calculations for specific supply chains. Although these sources are relevant for the LCI of a biodiesel supply chain, they are too diverse for any type of general characterization. For example, the prior use and vegetation of land on which rape seed is now being cultivated is very diverse and therefore the impact on the carbon balance as well.

2.2.2 Definition of CO2eq emission and energy use reductions

CO2 reduction = CO2eq emissions fossil fuel supply chain – CO2eq emissions biomass supply chain

Energy use reduction = Energy fossil fuel supply chain – Energy biomass supply chain

2.2.3 Allocation of emissions

In many supply chains of biomass, several other useful products are made besides the main product. This inevitably leads to the question of how the emissions and energy use of the biomass supply chain should be

allocated to these different products. The study decided that allocation based on market values is the best possible approach, as all production of oils is economically driven and this allocation method best reflects the impact of market factors on oil production.

3. Results and discussions

3.1 The supply chain for vegetables oils as a feedstock for biofuel

Only the biodiesel supply chain is taken into account in this study, as these oils are primarily used as feedstocks for biodiesel and to a much lesser extent as direct combustion fuels for electricity generation.

3.2 The supply chain for rape seed oil-based biodiesel

The supply chain for rape seed based biodiesel is shown in Figure 2. The parameters of the rape seed oil supply chain is shown in Table 1. Rape is grown on soils in temperate climates and harvested on an annual basis. After harvesting, new seeds must be planted to grow new rape. Crop rotation is practised in order to minimize disease incidence. The seeds are subsequently separated from the rape straw. Rape oil is extracted in a crushing facility and further extracted using a solvent, usually hexane. After further refining (i.e. free fatty acid removal), the rape oil (consisting mainly of triglycerides, with a few percent free fatty acids) is synthesized to biodiesel, using methanol. As a by-product, crude glycerine is formed.

28

Figure 2 - Rape seed oil based biodiesel supply chain

Table 1 - Main Parameters for rape seed oil supply chain analysis

Source: 6 N.D. Mortimer, P. Cormack, M.A. Elsayed, R.E. Home (2003), “Evaluation of the Comparative

Energy, Global Warming and Socio-Economic Costs and Benefits of Biodiesel”, Sheffield Hallam University

Notes and assumptions on rape seed calculations:

The rape seed is assumed to be produced on Western European soil. After harvesting, the seed is transported over a distance of 250 km (round-trip) to a storage facility. At this location, a crusher and refining plant and biodiesel production facility are also present in the Rotterdam area.

In this study, it is assumed that ammonium nitrate is used as the fertilizer and applied at a rate of 196 kg/ha/annum, corresponding to a rape seed yield of 2.88 t/ha/a.

6

Emissions due to seed production and minor chemical additives in the oil extraction and esterification process have been left out of the results, as they have a negligible impact on the overall result.

Parameter Quantity

Average rape seed yield 2.88 t/ha/a

Rape seed/straw ratio 1.02

Meal fraction 59.1%

Oil fraction 40.5%

Refining efficiency 97.5 %

Rapeseed oil required for biodiesel 1.052 t/t biodiesel

29

Fertilizer manufacturing data used is based on average European ammonium nitrate production plant data, as ammonium nitrate is the dominant fertilizer type used in Europe.

The emission factor used for nitrous oxide (N2O) emissions from the application of synthetic fertilizer (0.0125 kg N2O/kg N applied) is taken from the 1996 IPCC reference manual.

7

3.3 The Supply chain for soybean oil based biodiesel

The supply chain for soybean oil based biodiesel is shown in Figure 3. By and large, the supply chain is similar to that of rape, although the ratios of the different products generated, as shown in Table 2, differ from rape seed oil. In the supply chain for soy bean oil, it is assumed the soybeans are transported from the point of origin (e.g. North America) to Europe and further crushed and refined there.

Figure 3 - Soybean oil based biodiesel supply chain

Table 2 - Main parameters for soybean oil supply chain analysis

Source: 9 J. Sheehan, V. Camobreco, J. Duffield, M. Graboski, H. Shapouri (1998), “Life Cycle

Inventory of Biodiesel and Petroleum Diesel, for use in an Urban Bus”, National Renewable Energy Laboratory

Parameter Quantity

Average soy bean yield 2.55 t/ha/a

Meal fraction 76%

Oil fraction 17%

Refining efficiency 96 %

Soy bean oil required for biodiesel 1.037 t/t biodiesel

30

Notes and assumptions on soybean calculations:

The beans are assumed to be produced on North American soil. After harvesting, the beans are transported over a distance of 750 km (round-trip) to a storage facility at a North American harbour (east coast) and to Rotterdam harbour via sea vessel (6000 km one-way). In Rotterdam, the beans are transported to a facility in the direct area over a distance of 100 km (round-trip). At this location, the beans are crushed. The oil is then extracted, refined and converted to biodiesel.

The N2O emissions arising from nitrogen fixation by crops is a topic of current scientific debate. For the moment, the IPCC has assumed zero N2O emissions due to nitrogen fixation, due to uncertainties about which values to recommend. In this study, we have followed the IPCC guideline on this issue.

Just as with rape seed, the nitrogen fertilizer taken into account is ammonium nitrate.

It is assumed the esterification step takes place at the same physical location as the oil extraction and refining.

3.4 The Supply chain for sunflower oil based biodiesel

The supply chain for sunflower oil based biodiesel is shown in Figure 4. The main parameters for sunflower oil supply chain analysis are shown in Table 3. Sunflower is grown on soils in temperate climates and harvested on an annual basis. After harvesting, new seeds must be planted to grow new sunflowers. Sunflower oil is extracted in a crushing facility and extracted using a solvent, usually hexane. As a co-product, sunflower meal is produced, which is generally used as cattle feed. After further refining (i.e. free fatty acid removal), the sunflower oil can be synthesized to biodiesel, using methanol. As a by-product, crude glycerine is formed.

Figure 4 - Sunflower oil based biodiesel supply chain

31

Table 3 - Main parameters for sunflower oil supply chain analysis

Source: 9 F. Rosa (2007), “The Cogeneration Farm”, University of Udine, Italy

Notes and assumptions on sunflower calculations:

The sunflowers are assumed to be grown on Southern European soil. After harvesting, the seeds are transported over a distance of 250 km (round-trip) to a storage facility, where a crusher and refining plant are also present. Following these steps, the refined sunflower oil is transported over another distance of 500 km (round-trip) to a biodiesel production facility in the Rotterdam area.

Sunflower meal, the co-product generated in the supply chain, has a relatively low market value. Hence, a larger amount of the total supply chain emissions are allocated to sunflower oil than for other oils.

Emissions due to seed production and minor chemical additives in the oil extraction and esterification process are not included in the results, as their impact is negligible.

The calculations in this study are based on high oleic sunflower seeds.

3.5 The supply chain for canola based biodiesel

The main parameters for canola seed oil supply chain analysis is shown in Table 4. The supply chain for canola oil based biodiesel is shown in Figure 5. Canola is primarily grown on Canadian and Australian soils in temperate climates and harvested on an annual basis. After harvesting, new seeds must be planted to grow new canola. Canola oil is extracted in a crushing facility and extracted using a solvent, usually hexane. After further refining, the canola oil is synthesized to biodiesel, using methanol. As a by-product, crude glycerine is formed.

Table 4 - Main Parameters for canola seed oil supply chain analysis

Source: 10

J. Rollefson, G. Fu, A. Chan (2004), “Assessment of the Environmental Performance and Sustainability of Biodiesel in Canada”, National Research Council Canada

Parameter Quantity

Average sunflower seed yield 2.5 t/ha/a

Oil fraction 50%

Meal fraction 44%

Refining efficiency 98%

Sunflower oil required for biodiesel 1.052 t/t biodiesel

Parameter Quantity

Average canola seed yield 1.64 t/ha/a

Canola seed/straw ratio 1.00

Meal fraction 59%

Oil fraction 41%

Refining efficiency 96% %

Rapeseed oil required for biodiesel 1.035 t/t biodiesel

32

Figure 5 - Canola seed oil based biodiesel supply chain

Notes and assumptions on canola calculations:

The canola seed is assumed to be produced in Canada. After harvesting, the seed is transported over a distance of 250 km (round-trip) to a storage facility where the seed is also crushed; the oil is extracted and refined. Following these steps, the refined canola oil is transported over a further 500 km (round-trip) to a Canadian harbour. (east coast). From there, the canola oil is transported by sea vessel over 6000 km (one-way) to the Rotterdam harbour. In the harbour, the oil is transported a further 100 km (round-trip) to a biodiesel production facility in the Rotterdam area.

3.6 The supply chain for palm oil based biodiesel

The LCI model for palm oil based biodiesel is shown in Figure 6. The main parameters for palm oil supply chain analysis are shown in Table 5, based on typical Malaysian data and common practices in the sector. Palm oil is extracted during the milling stage. After further refining and esterification, palm oil is converted into biodiesel. As a by-product, crude glycerine is formed.

.

33

Figure 6 - Palm Oil specific LCI Model

Table 5 - Main parameters for Palm oil supply chain analysis

Source: 11

J. van Zutphen, R.A. Wijbrans (2007, the CO2 and Energy Balance of Malaysian Palm Oil, Current Status and Potential for Future Improvements”, CarbonCapital Solutions

Notes and assumptions on palm oil calculations:

N-fertilizer used for palm oil is assumed to be 50% urea and 50% ammonium sulphate

12

A refining step prior to esterification is introduced

Transportation distances of palm oil from Rotterdam harbour to a biodiesel facility have been brought to the same level as for other oils

Esterification data for palm oil based biodiesel has been harmonized with that of other vegetable oils

Parameter Quantity

Average palm tree density 143 palms/ha

Average FFB production 19.6 t/ha/a

Products generated from 1t FFB

- CPO 21%

- Palm kernels 6.5%

- EFB 23%

- Fibres 14.5%

- Shells 6.5%

- POME 0.7m3

Average oil extraction rate 20%

Average CPO yield 3.9 t/ha/a

Calorific value CPO 39.36 kJ/kg

34

3.7 Carbon balance for vegetable oils based biodiesel versus fossil diesel

The results for the carbon balance for biodiesel production based on typical rape seed, soy bean, sunflower, canola and palm oil supply chain are shown in Table 6. The total carbon emissions are for rapeseed 1,663, soy 1,387, sunflower 2,008, canola 1,562 and palm oil 1,711 kg CO2/tonne biodiesel produced. The

largest amount of GHG emissions to produce biodiesel is found for sunflower, followed by palm oil, rape seed, canola and soy in this order. As such, the carbon emission reduction of the five biofuels relative to fossil fuel, which has a total emission of 4,228 kg CO2/tonne, are as follows: soy 67%, canola 63%, rapeseed 61%, palm oil 60% and sunflower 53%

Table 6 - Results for carbon balance for biodiesel based on typical rapeseed, soybean, canola, sunflower and oil palm supply chain

Rape seed Soy bean Sunflower Canola Palm Oil (PO)

Biodiesel supply chain

Total emissions (A)

1,663 kg CO2/t biodiesel





1. Fertilizer use 648 kg/t 103 kg/t 688 kg/t 369 kg/t 112 kg/t

2. Fertilizer application (N2O)

338 kg/t 44 kg/t 294 kg/t 322 kg/t 125 kg/t

3. Pesticide use 19 kg/t 73 kg/t 32 kg/t 39 kg/t 32 kg/t

4. Cultivation & harvesting


5. Transport to crusher / mill (for PO)


6. Waste from Milling (for PO)

- - - - 876 kg/t

7. Oil extraction process / Milling process (for PO)


8. Refining

26 kg/t - 26 kg/t - 26 kg/t

9. Oil transport

9 kg/t - 18 kg/t 51 kg/t 45 kg/t

10. Esterification


Fossil diesel supply chain

Total emissions (B)

4,228 kg CO2/t diesel





1. Pre-treatment fossil fuels


2. Diesel combustion

3,809 kg/t 3,809 kg/t 3,809 kg/t 3,809 kg/t 3,809 kg/t

Net reduction (B) – (A)





2,537kg CO2/t biodiesel

Net reduction %

61%

67%

53%

63%

60%

35

3.8 Energy balance for vegetable oils based biodiesel versus fossil diesel

The results for the energy balance for biodiesel based on typical rape seed, soy bean, sunflower, canola and palm oil supply chain are shown in Table 7. The total emissions, expressed as GJ/tonne biodiesel are rape seed 14.46, soy 20.77, sunflower 21.14,

canola 18.66 and palm oil 11.57. The energy emission reduction of the five biofuels relative to fossil fuel are as follows: palm oil 79%, rape seed 74%, canola 66% and both soya and sunflower 62%.

Table 7 - Results for energy balance for biodiesel based on typical rape seed, soy bean, canola,

sunflower and oil palm supply chain

Rape seed Soy bean Sunflower Canola Palm Oil (PO)

Biodiesel supply chain

Total energy use (A)

14.46 GJ/t biodiesel





1. Fertilizer use 4.80 GJ/t 2.04 GJ/t 6.63 GJ/t 5.85 GJ/t 2.23 GJ/t

2. Fertilizer application (N2O)

0.00 GJ/t 0.00 GJ/t 0.00 GJ/t 0.00 GJ/t 0.00 GJ/t

3. Pesticide use 0.34 GJ/t 1.30 GJ/t 0.57 GJ/t 0.69 GJ/t 0.44 GJ/t

4. Cultivation & harvesting


5. Transport to crusher


6. Waste from Milling (for PO)

- - - - -0.47 GJ/t

7. Oil extraction process / Milling process (for PO)


8. Refining 0.36 GJ/t - 0.36 GJ/t - 0.37 GJ/t

9. Oil transport 0.12 GJ/t - 0.24 GJ/t 0.98 GJ/t 2.41 GJ/t

10. Esterification 5.47 GJ/t 5.47 GJ/t 5.47 GJ/t 5.47 GJ/t 5.47 GJ/t

Fossil diesel supply chain

Total energy use (B)

54.98 GJ/t diesel

54.98 GJ/t diesel

54.98 GJ/t diesel

54.98 GJ/t diesel

54.98 GJ/t diesel

1. Pre-treatment fossil fuels


2. Diesel combustion


Net reduction (B) – (A)






Net reduction %

74%

62%

62%

66%

79%

36

Figure 7- Emissions of vegetable oil supply chains for rape, soy, canola, sunflower and oil palm

3.9 Comparing the carbon balance of vegetable oils

Figure 7 shows the carbon balances grouped in the following categories: cultivation, transport, crushing/milling and biodiesel production. When compared to the other oil crops, palm oil has an additional emission source, which is the emission from waste disposed in the milling process. This source of emission is principally methane emission from the effluent ponds Referring to Figure 7 above, emissions associated with cultivation have the highest contribution to total emissions for all vegetable oil types, except palm oil. CO2 emissions for rape and canola are higher than for soy. The main causes are the much higher nitrogen fertilizer requirements for rape and canola, and the N2O emissions from the land application of artificial N-fertilizers. Fertilizers account for almost 60% of the CO2 emissions associated with the cultivation of rape, and 40% for canola. The higher emissions for rape seed compared to canola are the result of the use of a more “carbon-intensive” fertilizer for rape (i.e. ammonium nitrate) vs canola (a mixture of ammonia and urea). For soy this is less than 20%. N2O emissions account for 30% of the cultivation emissions for rape and 35% for canola.

For soy this is less than 10%. For sunflower, a higher total emission level is attributed to higher fertilizer requirements, especially for P and K. The yield of soy per ha of land is quite low compared to other oils, due to the low oil content of soy beans. Consequentially, fuel use for cultivation of the crop has the highest contribution (55%). Also, the emissions incurred in transport and crushing are relatively high compared to other oils. When looking at the overall emissions of palm oil, no major differences in CO2 emissions from other oils are apparent. However, the different components out of which the total emissions are composed are very different from the other oils. In the case of palm oil, fertilizer requirements have a much smaller impact on the overall result. This is because the oil yield per ha of land (or per kg fertilizer) is much higher for palm oil. The major component contributing to palm oil’s greenhouse gas emissions is the waste in its milling process. This component does not occur in the other supply chains. Figure 7 also shows that among the five biofuels, palm oil has the greatest potential to provide the means to lower greenhouse gas emissions by elimination or utilization of the wastes. Several options are available to achieve this, such as trapping methane for

37

renewable power generation and composting by combining empty fruit bunches with palm oil mill effluent. Application of these techniques will lead to a marked improvement of the carbon balance for palm oil from 60% net reduction to 80% net reduction. 4. Conclusions

Rape seed, palm oil, soy, sunflower and canola biodiesels are produced with lower GHG emissions relative to fossil diesel. Among them, sunflower biodiesel is produced with the highest GHG emissions, expressed as CO2eq, with 2,008 kg CO2/tonne biodiesel and soy with the least, at 1,387 kg CO2/tonne biodiesel. The carbon emission reduction of the five biofuels relative to fossil fuel, are soy 67%, canola 63%, rape seed 61%, palm oil 60% and sunflower 53%. LCI study for the biodiesels, using the energy balance approach, shows that energy use reductions, relative to fossil fuel are palm oil 79%, rapeseed 74%, canola 66% and both soya and sunflower 62%. In all cases, the GHG emission reductions exceed the critical value of 35% savings needed in European Union’s Renewable Energy Directive (EU RED) for a feedstock to qualify as a biofuel feedstock.

Fertilizer use has a significant impact on the overall emissions of rape, soy, sunflower and canola. The carbon balance for palm oil is much less sensitive to fertilizer use, due to its much higher oil yield per ha. Among the feedstocks studied, palm oil has the greatest potential for further GHG emission reduction. If palm oil can be produced without incurring emissions from the disposal of milling waste, the carbon balance will become significantly better than those of the other vegetable feedstocks. By utilizing its waste products and evading GHG emissions, the emission reduction increases to 80%. It then becomes the feedstock with the highest GHG reduction capability. 5. Acknowledgements This paper is condensed from a 2008 report titled “The CO2 and Energy Balance of Biodiesel derived from Vegetable Oils - LCI Comparisons of vegetable oils serving as feedstock for biodiesel” published by J.M. van Zutphen and R.A. Wijbrans of Carboncapital Solutions, Netherlands.

References

1. Wikipedia - Vegetable fats and oils,

http://en.wikipedia.org/wiki/Vegetable_fats_and_oils. Accessed on 24 November 2010

2. Food inflation rising as cooking oil to catch grain, Bloomberg Businessweek, http://www.businessweek.com/news/2010-11-01/food-inflation-rising-as-cooking-oil-to-catch-grain.html. Accessed on 24 November 2010

3. OECD-FAO Agricultural Outlook (2010) – Biofuel production 2010-19,http://www.agri-outlook.org/document/9/0,3343,en_36774715_36775671_45438665_1_1_1_1,00.html Accessed on 24 November 2010

4. Europa (2010). Directive 2009/28/EC (on the promotion of the use of energy from renewable sources, http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:32009L0028:EN:NOT Accessed on 24 November, 2010

5. ISO 14040:2006, Environmental Management – Life Cycle Assessment – Principles and Framework

6. N.D. Mortimer, P. Cormack, M.A. Elsayed, R.E. Home (2003), “Evaluation of the Comparative Energy, Global Warming and Socio-Economic Costs and Benefits of Biodiesel”, Sheffield Hallam University

7. IPCC Reference Manual, (1996) 8. J. Sheehan, V. Camobreco, J. Duffield,

M. Graboski, H. Shapouri (1998), “Life Cycle Inventory of Biodiesel and Petroleum Diesel, for use in an Urban Bus”, National Renewable Energy Laboratory

9. F. Rosa (2007), “The Cogeneration Farm”, University of Udine, Italy

10. J. Rollefson, G. Fu, A. Chan (2004), “Assessment of the Environmental Performance and Sustainability of Biodiesel in Canada”, National Research Council Canada

11. J. van Zutphen, R.A. Wijbrans (2007), The CO2 and Energy Balance of Malaysian Palm Oil, Current Status and Potential for Future Improvements”, CarbonCapital Solutions

12. G. Kongshaug (1998), “Energy Consumption and Greenhouse Gas Emissions in Fertilizer Production”, Hydro Agri Europe

http://en.wikipedia.org/wiki/Vegetable_fats_and_oils

http://en.wikipedia.org/wiki/Vegetable_fats_and_oils

http://www.businessweek.com/news/2010-11-01/food-inflation-rising-as-cooking-oil-to-catch-grain.html



http://www.agri-outlook.org/document/9/0,3343,en_36774715_36775671_45438665_1_1_1_1,00.html




http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:32009L0028:EN:NOT



Documents

Journal of Oil Palm & The Environment - JOPEH