University College Dublin, Dept of Agricultural & Food Engineering, Earlsfort Terrace.
Dublin 2, Ireland.
The renewed interest in environmentally compatible fuels has led to the choice of rapeseed oil as the main alternative to diesel fuel in Europe. The objective of this research was to produce and test an economic and high quality non-esterified rapeseed oil suitable for use as a diesel fuel extender. This was achieved by acidified hot water degumming combined with filtration to five microns. This rapeseed oil, designated as a Semi Refined Oil (SRO), has a high viscosity in comparison with diesel. Hence SRO fuel can only be used as a diesel fuel extender, with inclusion rates of up to 25 %.
SRO proved to be a suitable diesel fuel extender, at inclusion rates up to 25 %, when used with direct injection combustion systems (viz. tractor type engines). Power output (at 540 rev/min at the power take off shaft) was reduced by c. 0.06% for every 1% increase in SRO inclusion rate, and brake specific fuel consumption (BSFC) increased by c. 0.14% per 1% increase in SRO inclusion rate (viz. a 25% SRO/diesel blend had a 1.5% decrease in power and a 3.5% increase in BSFC compared with diesel). These values are in accordance with the lower energy density of rapeseed oil fuels compared with diesel. Chemical and viscosity analysis of engine lubrication oil (after c. 170 hours per fuel tested), including metal contamination as an indicator of engine wear occurring, showed that there was no measurable effect on engine lubricating oil due to SRO inclusion in diesel oil. When SRO was used to fuel IDI engines (viz. light duty commercial vehicles), power was considerably reduced mainly due to inadequate air/fuel mixing.
KEYWORDS: Biodiesel, SRO, Injector Fouling, Engine Tests
In 1900 at the Paris Exposition, Dr. Rudolf Diesel ran a prototype of his engine on groundnut oil (Lowry, 1990). In 1911 he was quoted as saying: "The diesel engine can be fed with vegetable oils and would help considerably in the development of agriculture of the countries which will use it". From the very beginning the diesel engine concept has been associated with vegetable oils as well as the original liquid coal-tar and later the petroleum derivatives (Seddon, 1942 and Wiebe et al., 1949). Initially, diesel engines were designed and developed to be of a dual-fuel nature, indeed it is believed that KHD Deutz engine manufacturers, Germany, warranted their original engines for operation with vegetable oils (Harwood, 1984). The practice of developing dual-fuelled engines continued up until the 1940's when two events caused a change in the development of the compression ignition engine. Firstly an abundance of petroleum supplies at a low cost tipped the fuel supply balance in favour of diesel fuel. Secondly, the effects of atmospheric pollution from automobiles were being felt in the Los Angeles basin and so that initiated the development of Clean Air legislation. This began to tighten the levels of emissions allowable from automobiles. The overall result of the two issues was to promote engine manufacturers to develop engines dedicated to run on diesel fuel oil and to tune the engine in order to decrease emissions thereby decreasing the ability of the engine to operate as a dual fuelled engine. Since the 1970's there has been a renewed interest in using vegetable oils in diesel engines for various reasons including: Political considerations, Environmental concerns, Economic aspects and European Union proposals.
The renewed interest in environmentally compatible fuels has led to the choice of rapeseed oil as the main alternative to diesel fuel in Europe. Esterified rapeseed oil (viz. rape methyl ester) has been the predominant vegetable oil fuel used because its characteristics are quite similar to diesel. It is, however, an expensive product (due to high processing costs) leading to renewed interest in an economic and high quality, non-esterified rapeseed oil. It has been demonstrated that the use of crude (gum content c. 2%) or degummed crude (gum content of 1.4%, this study) rapeseed oil leads to performance problems including filter blockages and engine coking. Gums are a major precursor of gel formation which becomes particularly problematic at temperatures below 2 °C. These problems can be ameliorated by using rapeseed oil which has been degummed to food grade standard (gum content < 0.2 %).
The objective of this research was to produce an economic and high quality non-esterified rapeseed oil suitable for use as a diesel fuel extender. This was achieved by acidified hot water degumming combined with filtration to five microns. The resultant degummed and filtered oil, had a gum content, as determined by the Differential Scanning Calorimetry (DSC), of 0.13 %(w/w) compared with c. 1.4% for unfiltered degummed rapeseed oil, 0.4% for non winterised rape methyl ester and c. 2 % for crude rapeseed oil (manufacturers specifications). This rapeseed oil, designated as a Semi Refined Oil (SRO), has a high viscosity in comparison with diesel (589 mPa.s v. 22 mPa.s at -12 °C). Tests on fuel pumping systems have shown that, in order to support adequate fuel flow and atomisation, the maximum acceptable viscosity for a fuel, in order to prevent fuel starvation, is c. 55 mPa.s at - 12 °C. Hence SRO fuel can only be used as a diesel fuel extender, with inclusion rates of up to 25 % as the resultant blend has a viscosity of 55 mPa.s at -12°C. The 25% SRO/diesel blend has a slightly lower energy content than diesel (41 v. 43 MJ/kg) while its density is slightly higher (0.87 v. 0.85 kg/l).
The problems associated with the use of crude vegetable oils in diesel engines have been discussed elsewhere (Barsic and Humke, 1981, Ziejewski and Kaufman, 1983 Goering and Fry, 1984 and Kaufman et al., 1985). The main conclusion from these researchers is that coking is a potentially serious problem with unmodified vegetable oil fuels. A unique method (based on reduced injector needle opening pressure Virk et al., (1991) was used to accelerate fouling combined with 2-dimensional image analysis) for assessing injector fouling was developed which has the advantage of enabling a very rapid engine test cycle to be used. Current methods can require up to a five hour test cycle whereas this new procedure is based on a twenty minute engine cycle, shown to be equivalent to approximately 2 500 hours of normal engine operation. The method used fibre optics and a 2-dimensional image analysis package to assess the extent of injector coking. A Fouling Index (FI) based on the ratio between the fouled injector orifice area compared with a clean injector orifice area was developed which enabled the fouling propensity of various fuel blends to be correlated. This showed that injector orifice blocking increases with increasing SRO inclusion rates. For example, a 25% SRO /diesel blend gave a Fouling Index (FI) of 0.67 compared with 0.40 for diesel. This means that injector fouling would be expected to occur considerably faster when operating on a 25% SRO/diesel blend. Hence injector service intervals would need to be reduced accordingly (viz. an injector with a service interval of c. 1 000 hours would need to be serviced at c. 600 hours, i.e. 1 000 (0.4/0.67)). Further work is required to confirm these preliminary observations.
In the Pennsylvania State University (Braun & Stephenson, 1982), short term tests were carried out on blends of degummed soyabean oil, ethanol and diesel in respective ratios of: 40: 20: 40, and 40: 30: 30. The engine was run for 25 hours on each blend. No problems were reported and no irregularities in the injector spray pattern was observed, however, the engine used for these tests was a 6 cylinder energy cell engine which utilises major and minor combustion chambers as in an indirect injection engine to aid turbulence and hence mixing of the air and fuel charge. This decreases the validity of comparing these results to standard agricultural diesel engines.
International Harvester Science and Technology Laboratory (Fort & Blumberg, 1982) conducted trials using blends of cottonseed oil and diesel oil. The cottonseed oil was refined almost to food grade in order to reduce its "particle content" to as low a value as possible by an inexpensive commercial treatment. The cottonseed oil was mixed with diesel in blends of: 30%, 50%, 65%, and 80% cottonseed oil. The tests consisted of 4 engine cycles at 15 hours per cycle. A 50% cottonseed oil: 50% diesel oil blend was chosen for a two hundred-hour endurance test. The short-term tests showed no significant differences between the fuels. After the endurance test the engine showed scoring on two of the cylinders, the corresponding pistons were also deeply scored with the surfaces torn. All the engines' top rings were heavily filled with a very hard carbonaceous deposit, which was obstructing the rings functions.
Barsic et al., (1981 a and b) evaluated crude soyabean oil, a 50: 50 mixture of crude soyabean oil and diesel, and degummed soyabean oil in a direct injection engine. The vegetable oils were evaluated in short-term tests i.e. 25 hours. Comparison of the engines performance and emissions for diesel versus the vegetable oils resulted in 1-2 g/l kWh lower thermal efficiency, 1-2 g/l kWh lower NOx, 2-20 g/l kWh more carbon monoxide, 1-2 g/l kWh more hydrocarbons and 1-2 g/l kWh more particulates for the vegetable oil. Comparing crude soyabean oil and degummed soyabean oil resulted in a 6% lower thermal efficiency for the crude oil versus a 1% lower thermal efficiency for the degummed oil. The coking of nozzles in both cases increased the emissions, with the crude soyabean giving a greater increase in total emissions than the degummed oil after 25 hours.
Jori and Hanzely (1993) experimented with rapeseed oil and rape methyl ester mixtures i.e. using rape methyl ester instead of diesel to lower the viscosity of the rapeseed oil. This was compared with both diesel and rape methyl ester. The performance of three tractors tested with a power take-off dynamometer was evaluated and concluded that there were no limitations with the use of rape oil fuels. The different fuels decreased the engine power by between 2-4% with a slightly improved energy consumption and combustion efficiencies. The rapeseed oil fuels had lower oxides of nitrogen, hydrocarbons and smoke emissions but slightly higher carbon monoxide emissions compared to diesel.
Worgetter (1981) carried out a series of tests using a research college tractor, which was fuelled with a blend of 50% (v/v) rapeseed oil and diesel. The rapeseed oil used was of food grade quality. These tests showed a power loss after 100 hours. At 350 hours the injector nozzles were visually inspected and although there were carbon residues, the manufacturer deemed that the injectors were suitable for continued use. The tests were stopped due to power losses and to prevent carbon deposits on the upper piston area.
King (1995) examined degummed and filtered rapeseed oil as a diesel fuel extender for direct injection engines. In that research a 15/85 (%v/v) blend of degummed and filtered rapeseed oil/ diesel was used to power a conventional agricultural tractor for c. 400 hours. Dynamometer testing showed that the outputs of power and torque were on average 2.5% lower on the test fuel compared with diesel. Brake specific fuel consumption was 1.5% higher with the test fuel also. An analysis of the engine lubricating oil for both diesel and the test blend showed no abnormal wear or elemental composition changes in the lubricating oil for the test fuel as compared to the diesel lubricating oil sample.
Semi Refined rapeseed Oil (SRO) proved to be a suitable diesel fuel extender, at inclusion rates up to 25 %, when used with direct injection combustion systems (viz. tractor type engines). Power output (at 540 rev/min at the power take off shaft) was reduced by c. 0.06% for every 1% increase in SRO inclusion rate, and brake specific fuel consumption (BSFC) increased by c. 0.14% per 1% increase in SRO inclusion rate (viz. a 25% SRO/diesel blend had a 1.5% decrease in power and a 3.5% increase in BSFC compared with diesel). These values are in accordance with the lower energy density of rapeseed oil fuels compared with diesel. Chemical and viscosity analysis of engine lubrication oil (after c. 170 hours per fuel tested), including metal contamination as an indicator of engine wear occurring, showed that there was no measurable effect on engine lubricating oil due to SRO inclusion in diesel oil. When SRO was used to fuel indirect injection combustion systems (viz. light duty commercial vehicles), power was considerably reduced mainly due to inadequate air/fuel mixing.
It was concluded that SRO can be used as a diesel fuel extender in unmodified direct injection diesel engines. The only practical difference observed in this study is that the injectors require more frequent servicing compared with diesel operation. The technology for producing SRO is relatively simple and hence offers the possibility of small, locally based production units as well as economic mass production units. Rape methyl ester requires major investment in industrial plant. For example, a rape methyl ester plant with a throughput of 36 000 tonnes per annum has an estimated capital cost of $18 million compared with approximately $3 million for an equivalent SRO rapeseed oil plant. Thus at road side diesel station, a 25% SRO/diesel blend would cost approximately $ 0.68/litre as compared to $0.73/litre for a 25% rape methyl ester/ diesel blend. Further work is required to determine if this cost advantage (7%) for a 25% SRO/diesel blend is sufficient to contravene any negative aspects of engine performance.
1. Lowery J.P.A. 1990. Alternative fuels for automotive and stationary engines in developing countries. Institution of Mechanical Engineers Seminar, U.K., November 19-20, pp31-35.
2. Seddon R.H., 1942. Vegetable oils in commercial vehicles. Gas Oil Power, August, pp 136-146.
3. Wiebe R. & Nawakawska J. 1949. USDA Bibliographical US Government printing office, Bulletin no. 10.
4. Harwood H.J. 1984. Oleochemicals as a fuel; Mechanical and economic feasibility. Journal of the American Oil Chemists Society, 61 (2) pp 315-324.
5. Barsic N.J. & Humke A.L. 1981a. Performance and emission characteristics of a naturally aspirated diesel engine with vegetable oil fuels. Society of Automotive Engineers 810262 pp95-109.
6. Ziejewski M. & Kaufmann K.R. 1983. Laboratory endurance test of a sunflower oil blend in a diesel engine. Journal of the American Oil Chemists Society 60 (8) pp1567-1573.
7. Goering C.E. & Fry B. 1984. Engine durability screening test of a diesel oil/soy oil/ alcohol micro emulsion fuel oil. Journal of the American Oil Chemists Society 61 (10) pp. 1627-1632.
8. Kaufmann K.R., Ziejewski M., Pratt G.L. & Goettler H.J. 1985. Fuel injection anomalies observed during long term engine performance tests on alternative fuels. Society of Automotive Engineers 852089 pp 591-600.
9. Virk K., Herbstmann S. & Rawdon R. 1991. Development of direct injection diesel engine injector keep clean and clean up tests. Society of Automotive Engineers 912329 pp 820-830.
10. Braun D.E. & Stephenson K.O. 1982. Alternative fuel blends and diesel engine tests. Proceedings of the International Conference on Plant and Vegetable Oils as Fuels. American Society of Agricultural Engineers, August, pp294-299.
11. Fort E.F. & Blumberg P.N. 1982 Performance and durability of a turbocharged diesel engine fuelled with cottonseed oil blends. Proceedings of the International Conference on Plant and Vegetable Oils as Fuels. American Society of Agricultural Engineers, August, pp374-383.
12. Barsic N.J. & Humke A.L. 1981b. Performance and emission characteristics of a naturally aspirated diesel engine with vegetable oil fuels. Society of Automotive Engineers 810955 pp2925-2935.
13. Jori I.J. & Hanzely G. 1994. Biodiesel research in Hungry. Research Status Report, Institute of Agricultural Engineering, Godollo, Hungry.
14. Worgetter M. 1981. Results of a long term engine test based on rapeseed oil fuel. Contribution to ICEUM III, October, Berlin, Germany, Pergamon Press.
15. King A. 1995. Rapeseed oil as a diesel fuel extender. M.Eng.Sc. Thesis, National University of Ireland.