- Email: [email protected]
Experimental Study on the Potential Application of Cottonseed Oil–Diesel Blends as Fuels for Automotive Diesel Engines
EXPERIMENTAL STUDY ON THE POTENTIAL APPLICATION OF COTTONSEED OIL -- DIESEL BLENDS AS FUELS FOR AUTOMOTIVE DIESEL ENGINES G. Fontaras1, T. Tzamkiozis1, E. Hatziemmanouil2 and Z. Samaras1, 1
Laboratory of Applied Thermodynamics, Aristotle University Thessaloniki, Thessaloniki, Greece. Karagiorgos Bros Cotton Industry, Thessaloniki, Greece.
2
Abstract: This paper presents the results of an experimental study on the direct application of cottonseed oil–diesel blends as fuel for diesel engine vehicles without using additional retrofit mechanical systems. The use of biofuels is one of the main actions promoted by the European Union and member states in an effort to tackle global warming, enhance energy security and contribute to regional development. Here, the possibility to blend cottonseed oil directly with fossil diesel as a fuel for diesel engines is examined. This option has lower cost and larger well-to-wheel greenhouse gas benefits than fatty acid methylesters. The paper presents measurements of important fuel properties, density, viscosity, cetane number and cold flow characteristics. In addition, a common rail Euro 3 compliant diesel car is tested using 10% v/v cottonseed oil-diesel blends in order to examine the effects on performance and emissions of regulated pollutants and CO2. Furthermore, particle emission characteristics are studied, including total and solid particle number concentrations and particle size distributions over driving cycles and steady state modes. The results indicate that the test fuel presents good operating characteristics and limited effects on regulated emissions and vehicle performance. These results would justify further research on the direct use of vegetable oils as automotive fuels. Keywords: biofuels; cottonseed oil; greenhouse gas emissions; exhaust emissions.
Correspondence to: Professor Z. Samaras, Laboratory of Applied Thermodynamics, Aristotle University Thessaloniki, P.O. Box 458, GR 54124 Thessaloniki, Greece. E-mail: [email protected]
DOI: 10.1205/psep07017 0957–5820/07/ $30.00 þ 0.00 Process Safety and Environmental Protection Trans IChemE, Part B, September 2007 # 2007 Institution of Chemical Engineers
INTRODUCTION
(He and Bao, 2005), cooked vegetable oils (Zaher et al., 2003) and even tomato seed oil (Giannelos et al., 2005). These studies reveal the potential vegetable oils have as fuels either directly or through transesterification. However they are usually limited to the analysis of the physical properties of the oils and fuels. In the preliminary work conducted in this study (Fontaras et al., 2006), results from an extensive (20 000 km) test fuel application on a Euro 2, 1.9 l turbo diesel vehicle indicated that using cottonseed oil–diesel blends is feasible for older diesel engines without a high pressure injection system. Nevertheless because of the need to limit emissions, improve fuel efficiency and optimise driveability, high pressure injection and common rail technology are currently the state of the art in diesel passenger cars and light trucks. These sophisticated systems are very sensitive to fuel quality and characteristics and thus any diesel fuel substitute should comply with the existing standards and not affect durability and performance. In the following the potential of vegetable oil application and the methodological
The use of vegetable oil blends with diesel fuel is foreseen in the European Union (EU) with the Directive 2003/30/EC (EC, 2003), which recognises pure vegetable oils as biofuels. These fuels are associated with interesting benefits such as low cost, fewer CO2 emissions over lifecycle, direct use without expensive installations and equipment requirements and no by-products. The application of such fuels may act as a precursor for future more sophisticated bio-fuels, providing the necessary time to the local economies and consumers for adapting in the new conditions. Furthermore vegetable oil–diesel blends present significant interest for all developing countries which lack energy sources and have agriculture based economies. So far, several types of vegetable oils have been investigated for direct application as diesel engine fuel such as rapeseed oil, jatropha oil (Forson et al., 2004), coconut oil, rubber oil (Ramadhas et al., 2005), cottonseed oil
396
Vol 85 (B5) 396–403
STUDY ON AUTOMOTIVE DIESEL ENGINES approach adopted for the evaluation of the test fuels are described. Additionally, results are presented of measurements conducted both on the fuel physical properties and the performance of a common rail vehicle operated under laboratory and real world conditions.
397
collaboration with local stakeholders and supported by the Greek state, have initiated a continuous experimental effort in order to investigate the possibility of pure cottonseed oil, as fuel for automotive applications.
METHODOLOGY AND RESULTS MOTIVATION Driven by the Kyoto protocol commitment and the need to reduce their dependency on fossil fuel imports, EU member states are in need of new renewable and domestic energy sources. With directive 2003/30/EC, the EU recognises the important role bio-fuels will play in a new sustainable and competitive economy. The directive proposes that bio-fuels should replace by the end of 2010 the 5.75% of the total fuel energy content used in transport (EC, 2003). Important efforts have been made towards creating the necessary background and infrastructure for supporting such a transition, but it is still doubtful that member states will achieve this target. Although the EU is currently the world leader in vegetable oil methylesters production (IEA, 2004), commonly known as bio-diesel, the production of gasoline bio-fuel substitutes remains very low, a fact which impends the achievement of the 5.75% goal. Furthermore since diesel consumption in EU road transport has surpassed that of gasoline (IFP, 2005) it is more sensible to target towards the development of diesel type biofuels. Second generation biofuels such as Biomass to Liquid products and Fischer Tropsch diesel bare important potential, but the necessary industrial facilities are still under development and will not be available before 2010 (BIOFRAC, 2006). It is evident that diesel-like biofuels should be exploited in the most efficient way. For southern European economies such as Greece, biofuels can play an important role by stimulating their relatively grown agricultural sector and reducing oil dependency. Local agriculture can provide the raw material for various biofuels and ensure the necessary supply chain. Nevertheless, the raw material production differentiation, the level of industrialization and the ability to implement various production technologies and strategies may require different solutions for different countries. Direct use of vegetable oils such as cottonseed oil as fuels offers several advantages especially in countries that lack industrial infrastructure and know-how to produce biofuels, but have a significant vegetable oil production. In these cases the application of vegetable oils can help accelerating the incorporation of biofuels in the country’s economy and gradually help building-up the necessary framework to support advanced biofuels. It can also provide the time in order for agriculture to adapt to the cultivation of new energy plants by starting with crops and practises producers are already familiar with. Greece in particular, according to 2003/30/EC should replace at least 150kTOE of automotive diesel by the end of 2010. Today approximately 80 000 people and 57 corporations are involved in cotton cultivation and production in Greece resulting in the production of 90 –100 ktons of cottonseed oil, which however is not an efficient raw material for biodiesel production. It becomes clear that taking advantage of the energy potential of cottonseed oil bares significant interest for the Greek economy. Based on these ideas and the fact that Greece has, amongst other vegetable oils, a significant cottonseed oil production, the authors, in
The approach adopted in the present study for the evaluation of cottonseed oil–diesel fuel blends was based mainly on the current EU legislation. According to Directive 2003/30/EC ‘pure vegetable oils produced from oil plants through pressing, extraction or comparable procedures, crude or refined but chemically unmodified are recognised as biofuels—plain or blended with diesel—when compatible with the type of engines involved and the corresponding emission requirements’. With regard to the legislative provisions the measurements conducted had two distinct targets: . Examine if and at what proportion cottonseed oil - diesel blends comply with the existing diesel fuel quality standards (diesel engine compatibility). . Run vehicle exhaust emission tests (emission standard requirements). A preliminary set of measurements of some key fuel properties (density, viscosity, cetane number) was conducted for blends of various oil concentrations in order to limit the number of possible fuel– oil concentrations that fulfil the basic fuel standard requirements. The test fuels were then applied on a VW Golf 1.9 TDi passenger car Euro 2 compliant for a mileage of 20 000 km in order to test the effectiveness and applicability of the test fuels. Regular measurements of the vehicle emissions and performance where conducted in the laboratory during this phase. The ratio of the average value measured using the test fuels over the average baseline measurement values for each driving cycle and pollutant is presented in Table 1. Results showed that the presence of cottonseed oil did not significantly affect the vehicle emission and consumption levels. After 20 000 km no important problems appeared due to the use of the test fuels with the exception of engine start-up difficulties under cold weather conditions (Fontaras et al., 2006) which highly depend on the oil concentration. It was concluded that the concentration of cottonseed oil that complied better with existing legislation and presented better operational characteristics was 10% v/v. In view of the findings of the initial test phase new measurements were conducted focusing on the standardized properties set for diesel fuel in Europe and some additional non-legislated cold flow characteristics. Moreover extensive application and detailed laboratory measurements were decided for evaluating the compatibility of the test fuel with modern common rail technology vehicles.
Cottonseed Oil -- Diesel Blends Properties Measurements The fuel properties investigated were density, viscosity at 408C, higher heating value, cetane number, cetane index, cold filter plugging point, cloud point, pour point, flash point and copper strip corrosion test. The fuels tested were cottonseed oil–diesel fuel blends of low oil concentration (10 and 20% v/v) and standard diesel fuel for reference. As vegetable oil properties can be affected by several factors
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396 –403
398
FONTARAS et al.
Table 1. Average value of the test fuel measurements over baseline measurements average value for each driving cycle-pollutant for the VW Golf.
Emission
Cottonseed oil concentration in fuel v/v
UDC cold
EUDC
NEDC
Artem Urban
Artem Road
Artem Mtw
10% 20% 10% 20% 10% 20% 10% 20% 10% 20% 10% 20%
0.98 1.05 0.98 1.03 1.19 1.35 1.14 1.47 0.97 1.07 1.34 1.68
0.90 0.97 0.90 0.95 1.62 1.52 1.27 1.41 0.89 1.10 1.07 1.14
0.94 1.01 0.93 0.99 1.26 1.38 1.17 1.45 0.93 1.09 1.17 1.34
0.94 0.97 0.93 0.95 1.15 1.06 1.21 1.14 0.91 1.04 1.05 0.96
0.93 0.97 0.92 0.96 0.95 0.86 1.04 1.01 0.88 1.06 1.19 1.00
0.95 0.97 0.94 0.95 0.95 0.88 0.95 0.95 0.85 1.05 1.00 0.87
CO2 FC CO HC NOx PM
such as the plant itself, weather and geographical conditions and the extraction procedure, some measurements were also conducted for plain cottonseed oil in order to obtain a picture of the quality of the cottonseed oil used. The results of the measurements are presented in Table 2.
Vehicle Measurements Following the fuel properties measurements, cottonseed oil–diesel blends of 10% v/v oil concentration were applied on a Euro 3 compliant Renault Laguna 1.9 dCi Common Rail passenger car. The main target was to investigate whether the vehicle remained within its Euro 3 emissions specification when the test fuel is used. It was important to examine the operation of the vehicle not only under type approval but also under real world driving conditions as well. This approach offers a more complete picture of vehicle operation and can support a thorough analysis. For these reasons the Artemis driving cycles were used in combination with the standard European type approval procedure. The Artemis driving cycles were developed in the framework of the project ARTEMIS (Assessment and Reliability of Transport Emission Models and Inventory Systems) a scientific programme funded by the European Commission, which aimed at the development of a harmonized emission model. The Artemis cycles (Andre´, 2004) are distinguished into three different sub-cycles that simulate different on road operating conditions: Artemis urban cycle (URBAN) for urban driving conditions, a semi-urban cycle (ROAD)
simulating the operation of the vehicle in a regular medium speed road, and the extra urban cycle (MOTORWAY) replicating the operation in high speed freeway. The speed versus time profile of the aforementioned cycles is presented in Figure 1. The protocol adopted for the measurements included one cold New European Driving Cycle—NEDC (the combined legislated driving cycle), one hot Urban Driving Cycle—UDC (urban sub-cycle of NEDC) and then the Artemis driving cycles. The timeline and the protocols followed for each measurement are presented in Table 3. Considering the vehicle mileage at the baseline as zero, a first thorough measurement was conducted after 2000 km of biofuel application, which included two repetitions of the Artemis protocol. In order to check the condition of the vehicle a second set of measurements was conducted following the type approval protocol after 6000 km of biofuel use. An additional set of measurements under the Artemis protocol was conducted after 12 000 km in order to examine the effect of the long term cottonseed oil use. Finally a second baseline measurement was conducted after 14 000 km for bracketing. In all cases for the emissions measurement, legislated constant volume sampling (CVS) was applied. The results of the measurements are summarised in Figures 2 –7. As regards particle emissions, measurements were also extended to non legislated particle properties such as number concentration and size distribution. Figure 8 shows the setup employed for these measurements. Samples were taken from the CVS with a Dekati Fine Sampler (FPS4000) operating at a nominal dilution ratio of 12 : 1.
Table 2. Fuel properties measurements and the respective fuel standards Property
Diesel 21
832 Density @ 208C (g l ) Viscosity @ 408C (cSt) 2.72 Higher heating value (kJ kg21) 44 963 Cetane number 52.5 55.1 Cetane index (mg kg21) CFPP (8C) 210 Cloud point (8C) 1 Pour point (8C) 216 Flash point (8C) 67.7 Copper strip corrosion 1A
Cottonseed Cottonseed Cottonseed oil 10% oil 20% oil 100% 841 3 44 475 54 53.27 26 1.7 217 68.4 1A
850 3.79 43 988 55 51.87 26 1.4 214 70.7 1A
920 32.75 40 086 41.2 — — — — — —
EN590
EN14214
Test method applied
820– 845 860 –900 2.0– 4.6 3.5– 5 — — 51 (min) 51 (min) 46 (min) — 25 (max Greece) 25 — — — — 55 (min) 120 (min) Class 1 Class 1
ASTM 287 ASTM D445 CEN/TS 14918:2005 DIN 51773 ASTM D976 IP 309 ASTM D2500 ASTM D97 ASTM D93 ASTM D130
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396– 403
STUDY ON AUTOMOTIVE DIESEL ENGINES
399
During each of these tests the vehicle was accelerated on the chassis dynamometer under full throttle and using the fourth gear. The time for accelerating from 60 to 110 km h21 was recorded. A set of five accelerations was conducted for each measurement. The results are summarized in Figure 12.
DISCUSSION AND CONCLUSIONS
Figure 1. Driving cycles of the measurement protocol.
Additionally, two calibrated ejector-type dilutors (Giechaskiel et al., 2004) were employed in order to bring particle emissions levels within the measuring range of the instruments. A Condensation Particle Counter (TSI’s 3010 CPC) was used to monitor the total particle number concentration. A Dekati’s Electrical Low Pressure Impactor (ELPI), sampling downstream of a Dekati’s Thermodenuder, provided the aerodynamic size distribution of ‘dry’ particles in real time. The ELPI operated with wet (oil-soaked) sintered plates and a filter stage that extended the lower cutpoint to 7 nm (Marjama¨ki et al., 2002). Over steady-state tests (50–90 – 120 km h21), a scanning mobility particle sizer (Model 3936L10 SMPS) was used instead of the CPC, monitoring the number weighted mobility size distribution. The SMPS operated on a sheath over sample flow ratio of 10 : 1l pm and the scan time was 90 s. In accordance with the gravimetric results, the number emission rates have been expressed per kilometre driven. It should be noticed that the ELPI data reduction requires the knowledge of the effective particle density. As this information is not available, a unit effective density was assumed (a common assumption in studies where an ELPI is employed). Moreover the ELPI results have been corrected for diffusion and space charge losses (Virtanen et al., 2001) as well as thermophoretic losses (Dekati, 2001) inside the thermodenuder. The results of the non regulated particle measurements are presented in Figures 9–11. Finally in order to quantify the effects of the test fuel on vehicle performance acceleration tests were performed.
Fuel properties measurements presented in Table 2, clearly show that a 10% v/v cottonseed oil– diesel fuel blend is in line with all EN590 specifications. In most cases the blends also comply with the EN14214, which is also provided for comparison. At this point it is important to stress the need of a broader biofuels standardization. EN14214 refers specifically to fatty acid methylesters taking into account several of their characteristics. As a result this norm becomes unsuitable for evaluating other biofuels such as vegetable oil –diesel blends. From Table 2 it is clear that the presence of cottonseed oil increases flash point, a fact that may influence the combustion evolution by interfering with the fuel evaporation process. In addition the fact that vegetable oil molecules include oxygen may result in different combustion characteristics. Cetane number, which is generally used as an indicator of combustion quality of the fuel, appears improved in the blends a fact which was not expected as vegetable oils tend to have low cetane numbers. The indirect calculation of cetane index on the other hand shows slight decreases, but cetane index is not the best indicator for vegetable oil based fuels that contain no aromatics (Graboski and McCormick, 1998). Considering the uncertainty of the measurements the only conclusion that can be drawn without a more detailed analysis is that the test fuels’ cetane number remains within the standardized limits for diesel fuel and thus combustion quality should be similar to that of plain diesel. Although all cold flow properties of the test fuel were found close to those of standard diesel, experience has shown that under cold weather conditions (ambient temperature at or below 08C), the engine was difficult to start when cold. At the same period the vehicle presented fuel atrophy symptoms and eventually the engine stopped operating after some days. The problem was fixed as soon as the fuel filter was replaced. Such problems are important but do not cancel the possibility of using cottonseed oil–diesel blends as automotive fuel. Possible solutions to this problem could be the use of a cold flow improver additive, removing the heavier molecules of the oil which cause filter plugging during the extraction process or retrofitting a fuel heater in the fuelling system. In any case additional research is necessary for dealing with this issue. After fixing the aforementioned problem the entire fuelling system of the vehicle was examined
Table 3. Measurements history and protocols followed. Measurement Baseline (diesel 50 ppm S) Cottonseed oil 10% v/v 1# (90% diesel 50 ppm S) Cottonseed oil 10% v/v 2# (90% diesel 50 ppm S) Cottonseed oil 10% v/v 3# (90% diesel 50 ppm S) Baseline #2 (diesel 50 ppm S)
Relative mileage (km) 0 (56 000 km real vehicle mileage) 2000 6000 12 000 14 000
Measurements conducted 2NEDC þ 1 Artemis þ Acceleration Tests 2(NEDC þ Artemis) þ Acceleration Tests 2NEDC 2(NEDC þ Artemis) þ Acceleration Tests 2(NEDC þ Artemis) þ Acceleration Tests
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396 –403
400
FONTARAS et al.
Figure 2. CO2 measurements results.
Figure 4. CO emissions measurements results.
in search of other possible malfunctions. At the time of this test the vehicle had almost completed the 12 000 km using test fuels. A certified electronic diagnostic tool was used in order to check the condition of the fuelling system and the measurements revealed no deviations from the values specified by the manufacturer. Furthermore since vegetable oils are also accused of causing injector coking (Pundir et al., 1994; Bauer et al., 2004) after long periods of use an optical inspection of the fuel injectors revealed no mechanical failures. Regarding CO2 emissions and fuel consumption several variations are observed (Figures 2 and 3) when using cottonseed oil–diesel blends. Most of these variations lay within the
accepted +5% uncertainty which is foreseen by legislation for such measurements. The use of vegetable oil– diesel fuel blends is also reported (Rakopoulos et al., 2006) to cause such differentiations that are also affected by engine load conditions. Results from previous measurements conducted using standard diesel indicate that the CO2 measurements values fall within the average range experienced for this vehicle. Thus any apparent increase in CO2 emissions time cannot be confirmed. Finally, since the average value of the measurements falls very close to the baseline it can be concluded that the overall energy performance and CO2 emissions of the vehicle are not affected by cottonseed oil.
Figure 3. Fuel consumption measurements results.
Figure 5. HC emissions measurements results.
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396– 403
STUDY ON AUTOMOTIVE DIESEL ENGINES
401
Figure 8. Schematic of the setup employed for the measurement of the unregulated particle properties.
Figure 6. NOx emissions measurements results.
Regarding regulated gaseous pollutants the situation is similar to that of CO2. In all cases emissions remain below the Euro 3 emission standard limits for the legislated test cycle (NEDC) as it is also presented in Table 4. For CO, HC and NOx the emission levels are close to those of the baseline. For PM the presence of vegetable oil appears to increase the scatter of the measurements but not the average value. It is important to note that the emissions of the vehicle under real world conditions remain also unaffected by the fuel. The situation is slightly different in non regulated particle emissions. The total particle number population for each fuelcycle combination is shown in Figure 9(a). Within the experimental uncertainty, excluding the Artemis Motorway cycle and the steady–state test at 120 km h21, the use of 10%
Figure 7. PM emissions measurements results.
v/v cotton oil, does not seem to change the total number of particles emitted. Under high load conditions though, addition of 10% v/v oil, results in significantly lower emissions. The reduction is in the range of 55%. Oil seems to prevent the formation of nanoparticles, which are usually generated under motorway driving conditions. Figure 9(b) shows the solid particle number emissions for all test cycles and fuels employed. The addition of 10% v/v cotton oil seems to cause a slight increase, in the range of
Figure 9. (a) Total particle number emission rate for the different car setups. The dots correspond to average values while the error-bars depict the maximum– minimum obtained. (b) Solid particle number emission rate for the different car setups. The dots correspond to average values while the error-bars depict the maximum– minimum obtained.
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396 –403
402
FONTARAS et al.
Figure 10. Number weighted size distributions at 90 kph steady speed cruising. The error bars enclose the minimum –maximum concentrations measured over different test repetitions.
14% on average, of the solid particle number emission rate. This apparent increase appears due to the high scatter of the measurement data, rather than the use of cotton oil itself. This is also verified by a statistical analysis (hypothesis testing). At the 95% level of confidence, there is insufficient evidence to reject the null hypothesis that the mean values of the solid particle emission rate are equal. Overall, within the experimental uncertainty, the cotton oil does not seem to affect the solid particle number emission rate. A nearly lognormal distribution was obtained under 50 kph and 90 kph steady state cruising. As an example, Figure 10 shows the number weighted size distributions at 90 kph. The presence of 10% v/v cotton oil, does not affect the shape of the distribution. The geometric mean diameter for the tests with oil agreed within +1 nm to those determined when diesel fuel was used. Figure 11 shows the number weighted size distributions at 120 kph steady state speed. Doping the fuel with 10% v/v cotton oil seems to lead in total suppression of the nucleation mode. This is consistent with the significant reduction (55%) of the total particle number emissions, which is observed under motorway driving conditions. The geometric mean diameter of the accumulation mode of all distributions is not affected. As regards the power output of the vehicle
Figure 11. Number weighted size distributions at 120 kph steady speed cruising. The error bars enclose the minimum –maximum concentrations measured over different test repetitions.
Figure 12. Acceleration test results.
expressed in acceleration times, a minor increase between the initial and the final baseline measurement may be noticed. This apparent 2% decrease of the vehicle power output is close to the accuracy of the method and can be caused by a number of other factors apart from the fuel like tire condition and so on. In order to reach a solid conclusion regarding the effect of the test fuels on the power output of the engine more accurate measurements should be conducted on engine test bed. Finally it is important to say that the picture obtained by these measurements was similar to that of the initial phase of the study when another test vehicle was used. The experience gained from these two case studies indicates that a 10% v/v cottonseed oil diesel blend does neither affect the performance of the vehicle nor the gaseous emissions. It should also be stressed that in both cases fuel filter plugging was experienced at low ambient temperatures, a problem which was more intense in the case of the vehicle presented here. Such problems can be tackled while in the oil refining process—remove stearines and other heavy molecules—or through special fuel additives. Both approaches need further detailed investigation. Optical examination and electronic diagnosis of the fueling system indicated no other mechanical failures in sensitive parts of the engine such as fuel injectors and the high pressure pump. Summarizing the analysis and the results discussed above it can be concluded that:
. Cottonseed oil–diesel blends may under certain conditions be applied directly as automotive fuel. . A 10% cottonseed oil –diesel blend fulfils all the requirements of the diesel EN590 standard. More specifically the values of fuel properties such as cetane number, viscosity, density, flash point and CFPP were found being within the acceptable range. Non-regulated cold flow properties were also found being close to those of the reference diesel fuel. . Cottonseed oil–diesel blends of 10% v/v concentration were applied successfully on a Euro 3 compliant diesel vehicle for a mileage of about 12 000 km
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396– 403
STUDY ON AUTOMOTIVE DIESEL ENGINES
403
Table 4. Euro 3 emission limits and legislated emissions measurements results. Pollutant CO NOx PM HC þ NOx
Euro 3 limits
Baseline
Cottonseed oil measurement #1
Cottonseed oil measurement #2
Cottonseed oil measurement #3
Baseline #2
0.64 0.50 0.05 0.56
0.39 0.40 0.03 0.44
0.50 0.35 0.03 0.39
0.27 0.44 0.03 0.47
0.25 0.40 0.04 0.42
0.23 0.41 0.04 0.43
. Fuel consumption and CO2 emissions fluctuate when applying the test fuel. The results scatter falls within the accuracy limits of the measurement and it is difficult to conclude whether the test fuel has a positive effect on vehicle fuel consumption or not. For this purpose targeted tests are necessary. . Gaseous pollutants emissions are not affected when operating with the oil-diesel blends. In some cases NOx appear to increase slightly, but the overall emissions over NEDC never exceeded the Euro 3 emission limits for which the vehicle is certified. . Particle number and size distribution measurements showed that the presence of cottonseed oil may suppress the production of nucleation mode particles. . The acceleration tests showed that the test fuel did not affect the power output of the vehicle. . Problems were experienced at low ambient temperatures as the engine suffered from misfueling. These problems were tackled by replacing the fuel filter. The overall conclusion drawn is that the potential of vegetable oils is important and should not be neglected. Further studies are necessary in order to fully understand the mechanisms that affect the properties of cottonseed oil-diesel blends as well as their effects on vehicle operation.
REFERENCES Andre´, M., 2004, The ARTEMIS European driving cycles for measuring car pollutant emissions, Science of the Total Environment, 334– 335: 73–84. Bauer, H., Dietsche, K.H. and Jager, T., 2004, Diesel Engine Management (Robert Bosch GmbH, Automotive Aftermarket Division, Germany). Biofuels Research Advisory Council (BIOFRAC), 2006, Biofuels in the European Union, a vision for 2030 and beyond, available at: http://ec.europa.eu/research/energy/nn/nn_rt/nn_rt_bm/ article_4012_en.htm. Dekati Ltd, 2001, Sampling automotive exhaust with a thermodenuder, Dekati technical note. European Commission (EC), 2003, Directive 2003/30/EC of the European Parliament and of the Council of 8 May 2003 on the promotion of the use of biofuels or other renewable fuels for transport, available at: http://ec.europa.eu/energy/res/legislation/ biofuels_en.htm. Fontaras, G., Mamakos, A., Ntziachristos, L., Miltsios, G. and Samaras, Z., 2006, Evaluation of cottonseed oil –diesel fuel blends as fuel for automotive diesel engines, Fisita World Automotive Congress Yokohama Japan, Fisita technical paper number F2006P380.
Forson, F., Oduro, E. and Donkoh, E., 2004, Performance of jatropha oil blends in a diesel engine, Renewable Energy, 29: 1135 –1145. Giannelos, P.N., Sxizas, S., Lois, E., Zannikos, F. and Anastopoulos, G., 2005, Physical, chemical and fuel related properties of tomato seed oil for evaluating its direct use in diesel engines, Industrial Crops and Products, 22(3): 193– 199. Giechaskiel, B., Ntziachristos, L. and Samaras, Z., 2004, Calibration and modelling of ejector dilutors for automotive exhaust sampling, Measurement Science and Technology, 15: 1– 8. Graboski, M. and McCormick, R.L., 1998, Combustion of fat and vegetable oil derived fuels in diesel engines, Progress in Energy and Combustion Science, 24(2): 125 –164. He, Y. and Bao, Y., 2005, Study on cottonseed oil as a partial substitute for diesel oil in fuel for single cylinder diesel engine, Technical note, Renewable Energy, 30: 805 –813. International Energy Agency (IEA), 2004, Biofuels for transport: an international perspective, Paris, downloaded from: http:// www.iea.org/textbase/nppdf/free/2004/biofuels2004.pdf. Institut Francais du Petrole (IFP), 2005, Road transport fuels in Europe: the explosion of demand for diesel fuel, downloaded from: http://www.ifp.fr/IFP/en/files/cinfo/IFP-Panorama05_10CarburantsRoutiersVA.pdf. Marjama¨ki, M., Ntziachristos, L., Virtanen, A., Ristima¨ki, J., Keskinen, J., Moiso, M., Palonen, M. and Lappi, M., 2002. Electrical filter stage for the ELPI. SAE paper 2002-01-0055. Pundir, B.P., Singal, S.K. and Gondal, A.K., 1994, Diesel fuel quality: Engine performance and emissions, SAE Technical Paper Series, Paper No 942020. Rakopoulos, C.D., Antonopoulos, K.A., Rakopoulos, D.C., Hountalas, D.T. and Giakoumis, E.G., 2006, Comparative performance and emissions study of a direct injection diesel engine using blends of diesel fuel with vegetable oils or bio-diesels of various origins, Energy Conversion and Management, 47(18–19): 3272– 3287. Ramadhas, A., Jayaraj, S. and Muraleedharan, C., 2005, Characterization and effect of using rubber seed oil as fuel in the compression ignition engines, Technical note, Renewable Energy, 30:795–803. Virtanen, A., Marjama¨ki, M., Ristima¨ki, J. and Keskinen, J., 2001, Fine particle losses in electrical low-pressure impactor, Journal of Aerosol Science, 32: 389 –401. Zaher, F.A., Megahed, O.A. and El Kinawy, O.S., 2003, Utilization of used frying oil as diesel engine fuel, Energy Sources, 25: 819– 826.
ACKNOWLEDGEMENT Part of this experimental work was supported by the Greek Scholarships Foundation (IKY). The manuscript was received 5 February 2007 and accepted for publication after revision 13 May 2007.
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396 –403
Laboratory of Applied Thermodynamics, Aristotle University Thessaloniki, Thessaloniki, Greece. Karagiorgos Bros Cotton Industry, Thessaloniki, Greece.
2
Abstract: This paper presents the results of an experimental study on the direct application of cottonseed oil–diesel blends as fuel for diesel engine vehicles without using additional retrofit mechanical systems. The use of biofuels is one of the main actions promoted by the European Union and member states in an effort to tackle global warming, enhance energy security and contribute to regional development. Here, the possibility to blend cottonseed oil directly with fossil diesel as a fuel for diesel engines is examined. This option has lower cost and larger well-to-wheel greenhouse gas benefits than fatty acid methylesters. The paper presents measurements of important fuel properties, density, viscosity, cetane number and cold flow characteristics. In addition, a common rail Euro 3 compliant diesel car is tested using 10% v/v cottonseed oil-diesel blends in order to examine the effects on performance and emissions of regulated pollutants and CO2. Furthermore, particle emission characteristics are studied, including total and solid particle number concentrations and particle size distributions over driving cycles and steady state modes. The results indicate that the test fuel presents good operating characteristics and limited effects on regulated emissions and vehicle performance. These results would justify further research on the direct use of vegetable oils as automotive fuels. Keywords: biofuels; cottonseed oil; greenhouse gas emissions; exhaust emissions.
Correspondence to: Professor Z. Samaras, Laboratory of Applied Thermodynamics, Aristotle University Thessaloniki, P.O. Box 458, GR 54124 Thessaloniki, Greece. E-mail: [email protected]
DOI: 10.1205/psep07017 0957–5820/07/ $30.00 þ 0.00 Process Safety and Environmental Protection Trans IChemE, Part B, September 2007 # 2007 Institution of Chemical Engineers
INTRODUCTION
(He and Bao, 2005), cooked vegetable oils (Zaher et al., 2003) and even tomato seed oil (Giannelos et al., 2005). These studies reveal the potential vegetable oils have as fuels either directly or through transesterification. However they are usually limited to the analysis of the physical properties of the oils and fuels. In the preliminary work conducted in this study (Fontaras et al., 2006), results from an extensive (20 000 km) test fuel application on a Euro 2, 1.9 l turbo diesel vehicle indicated that using cottonseed oil–diesel blends is feasible for older diesel engines without a high pressure injection system. Nevertheless because of the need to limit emissions, improve fuel efficiency and optimise driveability, high pressure injection and common rail technology are currently the state of the art in diesel passenger cars and light trucks. These sophisticated systems are very sensitive to fuel quality and characteristics and thus any diesel fuel substitute should comply with the existing standards and not affect durability and performance. In the following the potential of vegetable oil application and the methodological
The use of vegetable oil blends with diesel fuel is foreseen in the European Union (EU) with the Directive 2003/30/EC (EC, 2003), which recognises pure vegetable oils as biofuels. These fuels are associated with interesting benefits such as low cost, fewer CO2 emissions over lifecycle, direct use without expensive installations and equipment requirements and no by-products. The application of such fuels may act as a precursor for future more sophisticated bio-fuels, providing the necessary time to the local economies and consumers for adapting in the new conditions. Furthermore vegetable oil–diesel blends present significant interest for all developing countries which lack energy sources and have agriculture based economies. So far, several types of vegetable oils have been investigated for direct application as diesel engine fuel such as rapeseed oil, jatropha oil (Forson et al., 2004), coconut oil, rubber oil (Ramadhas et al., 2005), cottonseed oil
396
Vol 85 (B5) 396–403
STUDY ON AUTOMOTIVE DIESEL ENGINES approach adopted for the evaluation of the test fuels are described. Additionally, results are presented of measurements conducted both on the fuel physical properties and the performance of a common rail vehicle operated under laboratory and real world conditions.
397
collaboration with local stakeholders and supported by the Greek state, have initiated a continuous experimental effort in order to investigate the possibility of pure cottonseed oil, as fuel for automotive applications.
METHODOLOGY AND RESULTS MOTIVATION Driven by the Kyoto protocol commitment and the need to reduce their dependency on fossil fuel imports, EU member states are in need of new renewable and domestic energy sources. With directive 2003/30/EC, the EU recognises the important role bio-fuels will play in a new sustainable and competitive economy. The directive proposes that bio-fuels should replace by the end of 2010 the 5.75% of the total fuel energy content used in transport (EC, 2003). Important efforts have been made towards creating the necessary background and infrastructure for supporting such a transition, but it is still doubtful that member states will achieve this target. Although the EU is currently the world leader in vegetable oil methylesters production (IEA, 2004), commonly known as bio-diesel, the production of gasoline bio-fuel substitutes remains very low, a fact which impends the achievement of the 5.75% goal. Furthermore since diesel consumption in EU road transport has surpassed that of gasoline (IFP, 2005) it is more sensible to target towards the development of diesel type biofuels. Second generation biofuels such as Biomass to Liquid products and Fischer Tropsch diesel bare important potential, but the necessary industrial facilities are still under development and will not be available before 2010 (BIOFRAC, 2006). It is evident that diesel-like biofuels should be exploited in the most efficient way. For southern European economies such as Greece, biofuels can play an important role by stimulating their relatively grown agricultural sector and reducing oil dependency. Local agriculture can provide the raw material for various biofuels and ensure the necessary supply chain. Nevertheless, the raw material production differentiation, the level of industrialization and the ability to implement various production technologies and strategies may require different solutions for different countries. Direct use of vegetable oils such as cottonseed oil as fuels offers several advantages especially in countries that lack industrial infrastructure and know-how to produce biofuels, but have a significant vegetable oil production. In these cases the application of vegetable oils can help accelerating the incorporation of biofuels in the country’s economy and gradually help building-up the necessary framework to support advanced biofuels. It can also provide the time in order for agriculture to adapt to the cultivation of new energy plants by starting with crops and practises producers are already familiar with. Greece in particular, according to 2003/30/EC should replace at least 150kTOE of automotive diesel by the end of 2010. Today approximately 80 000 people and 57 corporations are involved in cotton cultivation and production in Greece resulting in the production of 90 –100 ktons of cottonseed oil, which however is not an efficient raw material for biodiesel production. It becomes clear that taking advantage of the energy potential of cottonseed oil bares significant interest for the Greek economy. Based on these ideas and the fact that Greece has, amongst other vegetable oils, a significant cottonseed oil production, the authors, in
The approach adopted in the present study for the evaluation of cottonseed oil–diesel fuel blends was based mainly on the current EU legislation. According to Directive 2003/30/EC ‘pure vegetable oils produced from oil plants through pressing, extraction or comparable procedures, crude or refined but chemically unmodified are recognised as biofuels—plain or blended with diesel—when compatible with the type of engines involved and the corresponding emission requirements’. With regard to the legislative provisions the measurements conducted had two distinct targets: . Examine if and at what proportion cottonseed oil - diesel blends comply with the existing diesel fuel quality standards (diesel engine compatibility). . Run vehicle exhaust emission tests (emission standard requirements). A preliminary set of measurements of some key fuel properties (density, viscosity, cetane number) was conducted for blends of various oil concentrations in order to limit the number of possible fuel– oil concentrations that fulfil the basic fuel standard requirements. The test fuels were then applied on a VW Golf 1.9 TDi passenger car Euro 2 compliant for a mileage of 20 000 km in order to test the effectiveness and applicability of the test fuels. Regular measurements of the vehicle emissions and performance where conducted in the laboratory during this phase. The ratio of the average value measured using the test fuels over the average baseline measurement values for each driving cycle and pollutant is presented in Table 1. Results showed that the presence of cottonseed oil did not significantly affect the vehicle emission and consumption levels. After 20 000 km no important problems appeared due to the use of the test fuels with the exception of engine start-up difficulties under cold weather conditions (Fontaras et al., 2006) which highly depend on the oil concentration. It was concluded that the concentration of cottonseed oil that complied better with existing legislation and presented better operational characteristics was 10% v/v. In view of the findings of the initial test phase new measurements were conducted focusing on the standardized properties set for diesel fuel in Europe and some additional non-legislated cold flow characteristics. Moreover extensive application and detailed laboratory measurements were decided for evaluating the compatibility of the test fuel with modern common rail technology vehicles.
Cottonseed Oil -- Diesel Blends Properties Measurements The fuel properties investigated were density, viscosity at 408C, higher heating value, cetane number, cetane index, cold filter plugging point, cloud point, pour point, flash point and copper strip corrosion test. The fuels tested were cottonseed oil–diesel fuel blends of low oil concentration (10 and 20% v/v) and standard diesel fuel for reference. As vegetable oil properties can be affected by several factors
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396 –403
398
FONTARAS et al.
Table 1. Average value of the test fuel measurements over baseline measurements average value for each driving cycle-pollutant for the VW Golf.
Emission
Cottonseed oil concentration in fuel v/v
UDC cold
EUDC
NEDC
Artem Urban
Artem Road
Artem Mtw
10% 20% 10% 20% 10% 20% 10% 20% 10% 20% 10% 20%
0.98 1.05 0.98 1.03 1.19 1.35 1.14 1.47 0.97 1.07 1.34 1.68
0.90 0.97 0.90 0.95 1.62 1.52 1.27 1.41 0.89 1.10 1.07 1.14
0.94 1.01 0.93 0.99 1.26 1.38 1.17 1.45 0.93 1.09 1.17 1.34
0.94 0.97 0.93 0.95 1.15 1.06 1.21 1.14 0.91 1.04 1.05 0.96
0.93 0.97 0.92 0.96 0.95 0.86 1.04 1.01 0.88 1.06 1.19 1.00
0.95 0.97 0.94 0.95 0.95 0.88 0.95 0.95 0.85 1.05 1.00 0.87
CO2 FC CO HC NOx PM
such as the plant itself, weather and geographical conditions and the extraction procedure, some measurements were also conducted for plain cottonseed oil in order to obtain a picture of the quality of the cottonseed oil used. The results of the measurements are presented in Table 2.
Vehicle Measurements Following the fuel properties measurements, cottonseed oil–diesel blends of 10% v/v oil concentration were applied on a Euro 3 compliant Renault Laguna 1.9 dCi Common Rail passenger car. The main target was to investigate whether the vehicle remained within its Euro 3 emissions specification when the test fuel is used. It was important to examine the operation of the vehicle not only under type approval but also under real world driving conditions as well. This approach offers a more complete picture of vehicle operation and can support a thorough analysis. For these reasons the Artemis driving cycles were used in combination with the standard European type approval procedure. The Artemis driving cycles were developed in the framework of the project ARTEMIS (Assessment and Reliability of Transport Emission Models and Inventory Systems) a scientific programme funded by the European Commission, which aimed at the development of a harmonized emission model. The Artemis cycles (Andre´, 2004) are distinguished into three different sub-cycles that simulate different on road operating conditions: Artemis urban cycle (URBAN) for urban driving conditions, a semi-urban cycle (ROAD)
simulating the operation of the vehicle in a regular medium speed road, and the extra urban cycle (MOTORWAY) replicating the operation in high speed freeway. The speed versus time profile of the aforementioned cycles is presented in Figure 1. The protocol adopted for the measurements included one cold New European Driving Cycle—NEDC (the combined legislated driving cycle), one hot Urban Driving Cycle—UDC (urban sub-cycle of NEDC) and then the Artemis driving cycles. The timeline and the protocols followed for each measurement are presented in Table 3. Considering the vehicle mileage at the baseline as zero, a first thorough measurement was conducted after 2000 km of biofuel application, which included two repetitions of the Artemis protocol. In order to check the condition of the vehicle a second set of measurements was conducted following the type approval protocol after 6000 km of biofuel use. An additional set of measurements under the Artemis protocol was conducted after 12 000 km in order to examine the effect of the long term cottonseed oil use. Finally a second baseline measurement was conducted after 14 000 km for bracketing. In all cases for the emissions measurement, legislated constant volume sampling (CVS) was applied. The results of the measurements are summarised in Figures 2 –7. As regards particle emissions, measurements were also extended to non legislated particle properties such as number concentration and size distribution. Figure 8 shows the setup employed for these measurements. Samples were taken from the CVS with a Dekati Fine Sampler (FPS4000) operating at a nominal dilution ratio of 12 : 1.
Table 2. Fuel properties measurements and the respective fuel standards Property
Diesel 21
832 Density @ 208C (g l ) Viscosity @ 408C (cSt) 2.72 Higher heating value (kJ kg21) 44 963 Cetane number 52.5 55.1 Cetane index (mg kg21) CFPP (8C) 210 Cloud point (8C) 1 Pour point (8C) 216 Flash point (8C) 67.7 Copper strip corrosion 1A
Cottonseed Cottonseed Cottonseed oil 10% oil 20% oil 100% 841 3 44 475 54 53.27 26 1.7 217 68.4 1A
850 3.79 43 988 55 51.87 26 1.4 214 70.7 1A
920 32.75 40 086 41.2 — — — — — —
EN590
EN14214
Test method applied
820– 845 860 –900 2.0– 4.6 3.5– 5 — — 51 (min) 51 (min) 46 (min) — 25 (max Greece) 25 — — — — 55 (min) 120 (min) Class 1 Class 1
ASTM 287 ASTM D445 CEN/TS 14918:2005 DIN 51773 ASTM D976 IP 309 ASTM D2500 ASTM D97 ASTM D93 ASTM D130
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396– 403
STUDY ON AUTOMOTIVE DIESEL ENGINES
399
During each of these tests the vehicle was accelerated on the chassis dynamometer under full throttle and using the fourth gear. The time for accelerating from 60 to 110 km h21 was recorded. A set of five accelerations was conducted for each measurement. The results are summarized in Figure 12.
DISCUSSION AND CONCLUSIONS
Figure 1. Driving cycles of the measurement protocol.
Additionally, two calibrated ejector-type dilutors (Giechaskiel et al., 2004) were employed in order to bring particle emissions levels within the measuring range of the instruments. A Condensation Particle Counter (TSI’s 3010 CPC) was used to monitor the total particle number concentration. A Dekati’s Electrical Low Pressure Impactor (ELPI), sampling downstream of a Dekati’s Thermodenuder, provided the aerodynamic size distribution of ‘dry’ particles in real time. The ELPI operated with wet (oil-soaked) sintered plates and a filter stage that extended the lower cutpoint to 7 nm (Marjama¨ki et al., 2002). Over steady-state tests (50–90 – 120 km h21), a scanning mobility particle sizer (Model 3936L10 SMPS) was used instead of the CPC, monitoring the number weighted mobility size distribution. The SMPS operated on a sheath over sample flow ratio of 10 : 1l pm and the scan time was 90 s. In accordance with the gravimetric results, the number emission rates have been expressed per kilometre driven. It should be noticed that the ELPI data reduction requires the knowledge of the effective particle density. As this information is not available, a unit effective density was assumed (a common assumption in studies where an ELPI is employed). Moreover the ELPI results have been corrected for diffusion and space charge losses (Virtanen et al., 2001) as well as thermophoretic losses (Dekati, 2001) inside the thermodenuder. The results of the non regulated particle measurements are presented in Figures 9–11. Finally in order to quantify the effects of the test fuel on vehicle performance acceleration tests were performed.
Fuel properties measurements presented in Table 2, clearly show that a 10% v/v cottonseed oil– diesel fuel blend is in line with all EN590 specifications. In most cases the blends also comply with the EN14214, which is also provided for comparison. At this point it is important to stress the need of a broader biofuels standardization. EN14214 refers specifically to fatty acid methylesters taking into account several of their characteristics. As a result this norm becomes unsuitable for evaluating other biofuels such as vegetable oil –diesel blends. From Table 2 it is clear that the presence of cottonseed oil increases flash point, a fact that may influence the combustion evolution by interfering with the fuel evaporation process. In addition the fact that vegetable oil molecules include oxygen may result in different combustion characteristics. Cetane number, which is generally used as an indicator of combustion quality of the fuel, appears improved in the blends a fact which was not expected as vegetable oils tend to have low cetane numbers. The indirect calculation of cetane index on the other hand shows slight decreases, but cetane index is not the best indicator for vegetable oil based fuels that contain no aromatics (Graboski and McCormick, 1998). Considering the uncertainty of the measurements the only conclusion that can be drawn without a more detailed analysis is that the test fuels’ cetane number remains within the standardized limits for diesel fuel and thus combustion quality should be similar to that of plain diesel. Although all cold flow properties of the test fuel were found close to those of standard diesel, experience has shown that under cold weather conditions (ambient temperature at or below 08C), the engine was difficult to start when cold. At the same period the vehicle presented fuel atrophy symptoms and eventually the engine stopped operating after some days. The problem was fixed as soon as the fuel filter was replaced. Such problems are important but do not cancel the possibility of using cottonseed oil–diesel blends as automotive fuel. Possible solutions to this problem could be the use of a cold flow improver additive, removing the heavier molecules of the oil which cause filter plugging during the extraction process or retrofitting a fuel heater in the fuelling system. In any case additional research is necessary for dealing with this issue. After fixing the aforementioned problem the entire fuelling system of the vehicle was examined
Table 3. Measurements history and protocols followed. Measurement Baseline (diesel 50 ppm S) Cottonseed oil 10% v/v 1# (90% diesel 50 ppm S) Cottonseed oil 10% v/v 2# (90% diesel 50 ppm S) Cottonseed oil 10% v/v 3# (90% diesel 50 ppm S) Baseline #2 (diesel 50 ppm S)
Relative mileage (km) 0 (56 000 km real vehicle mileage) 2000 6000 12 000 14 000
Measurements conducted 2NEDC þ 1 Artemis þ Acceleration Tests 2(NEDC þ Artemis) þ Acceleration Tests 2NEDC 2(NEDC þ Artemis) þ Acceleration Tests 2(NEDC þ Artemis) þ Acceleration Tests
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396 –403
400
FONTARAS et al.
Figure 2. CO2 measurements results.
Figure 4. CO emissions measurements results.
in search of other possible malfunctions. At the time of this test the vehicle had almost completed the 12 000 km using test fuels. A certified electronic diagnostic tool was used in order to check the condition of the fuelling system and the measurements revealed no deviations from the values specified by the manufacturer. Furthermore since vegetable oils are also accused of causing injector coking (Pundir et al., 1994; Bauer et al., 2004) after long periods of use an optical inspection of the fuel injectors revealed no mechanical failures. Regarding CO2 emissions and fuel consumption several variations are observed (Figures 2 and 3) when using cottonseed oil–diesel blends. Most of these variations lay within the
accepted +5% uncertainty which is foreseen by legislation for such measurements. The use of vegetable oil– diesel fuel blends is also reported (Rakopoulos et al., 2006) to cause such differentiations that are also affected by engine load conditions. Results from previous measurements conducted using standard diesel indicate that the CO2 measurements values fall within the average range experienced for this vehicle. Thus any apparent increase in CO2 emissions time cannot be confirmed. Finally, since the average value of the measurements falls very close to the baseline it can be concluded that the overall energy performance and CO2 emissions of the vehicle are not affected by cottonseed oil.
Figure 3. Fuel consumption measurements results.
Figure 5. HC emissions measurements results.
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396– 403
STUDY ON AUTOMOTIVE DIESEL ENGINES
401
Figure 8. Schematic of the setup employed for the measurement of the unregulated particle properties.
Figure 6. NOx emissions measurements results.
Regarding regulated gaseous pollutants the situation is similar to that of CO2. In all cases emissions remain below the Euro 3 emission standard limits for the legislated test cycle (NEDC) as it is also presented in Table 4. For CO, HC and NOx the emission levels are close to those of the baseline. For PM the presence of vegetable oil appears to increase the scatter of the measurements but not the average value. It is important to note that the emissions of the vehicle under real world conditions remain also unaffected by the fuel. The situation is slightly different in non regulated particle emissions. The total particle number population for each fuelcycle combination is shown in Figure 9(a). Within the experimental uncertainty, excluding the Artemis Motorway cycle and the steady–state test at 120 km h21, the use of 10%
Figure 7. PM emissions measurements results.
v/v cotton oil, does not seem to change the total number of particles emitted. Under high load conditions though, addition of 10% v/v oil, results in significantly lower emissions. The reduction is in the range of 55%. Oil seems to prevent the formation of nanoparticles, which are usually generated under motorway driving conditions. Figure 9(b) shows the solid particle number emissions for all test cycles and fuels employed. The addition of 10% v/v cotton oil seems to cause a slight increase, in the range of
Figure 9. (a) Total particle number emission rate for the different car setups. The dots correspond to average values while the error-bars depict the maximum– minimum obtained. (b) Solid particle number emission rate for the different car setups. The dots correspond to average values while the error-bars depict the maximum– minimum obtained.
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396 –403
402
FONTARAS et al.
Figure 10. Number weighted size distributions at 90 kph steady speed cruising. The error bars enclose the minimum –maximum concentrations measured over different test repetitions.
14% on average, of the solid particle number emission rate. This apparent increase appears due to the high scatter of the measurement data, rather than the use of cotton oil itself. This is also verified by a statistical analysis (hypothesis testing). At the 95% level of confidence, there is insufficient evidence to reject the null hypothesis that the mean values of the solid particle emission rate are equal. Overall, within the experimental uncertainty, the cotton oil does not seem to affect the solid particle number emission rate. A nearly lognormal distribution was obtained under 50 kph and 90 kph steady state cruising. As an example, Figure 10 shows the number weighted size distributions at 90 kph. The presence of 10% v/v cotton oil, does not affect the shape of the distribution. The geometric mean diameter for the tests with oil agreed within +1 nm to those determined when diesel fuel was used. Figure 11 shows the number weighted size distributions at 120 kph steady state speed. Doping the fuel with 10% v/v cotton oil seems to lead in total suppression of the nucleation mode. This is consistent with the significant reduction (55%) of the total particle number emissions, which is observed under motorway driving conditions. The geometric mean diameter of the accumulation mode of all distributions is not affected. As regards the power output of the vehicle
Figure 11. Number weighted size distributions at 120 kph steady speed cruising. The error bars enclose the minimum –maximum concentrations measured over different test repetitions.
Figure 12. Acceleration test results.
expressed in acceleration times, a minor increase between the initial and the final baseline measurement may be noticed. This apparent 2% decrease of the vehicle power output is close to the accuracy of the method and can be caused by a number of other factors apart from the fuel like tire condition and so on. In order to reach a solid conclusion regarding the effect of the test fuels on the power output of the engine more accurate measurements should be conducted on engine test bed. Finally it is important to say that the picture obtained by these measurements was similar to that of the initial phase of the study when another test vehicle was used. The experience gained from these two case studies indicates that a 10% v/v cottonseed oil diesel blend does neither affect the performance of the vehicle nor the gaseous emissions. It should also be stressed that in both cases fuel filter plugging was experienced at low ambient temperatures, a problem which was more intense in the case of the vehicle presented here. Such problems can be tackled while in the oil refining process—remove stearines and other heavy molecules—or through special fuel additives. Both approaches need further detailed investigation. Optical examination and electronic diagnosis of the fueling system indicated no other mechanical failures in sensitive parts of the engine such as fuel injectors and the high pressure pump. Summarizing the analysis and the results discussed above it can be concluded that:
. Cottonseed oil–diesel blends may under certain conditions be applied directly as automotive fuel. . A 10% cottonseed oil –diesel blend fulfils all the requirements of the diesel EN590 standard. More specifically the values of fuel properties such as cetane number, viscosity, density, flash point and CFPP were found being within the acceptable range. Non-regulated cold flow properties were also found being close to those of the reference diesel fuel. . Cottonseed oil–diesel blends of 10% v/v concentration were applied successfully on a Euro 3 compliant diesel vehicle for a mileage of about 12 000 km
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396– 403
STUDY ON AUTOMOTIVE DIESEL ENGINES
403
Table 4. Euro 3 emission limits and legislated emissions measurements results. Pollutant CO NOx PM HC þ NOx
Euro 3 limits
Baseline
Cottonseed oil measurement #1
Cottonseed oil measurement #2
Cottonseed oil measurement #3
Baseline #2
0.64 0.50 0.05 0.56
0.39 0.40 0.03 0.44
0.50 0.35 0.03 0.39
0.27 0.44 0.03 0.47
0.25 0.40 0.04 0.42
0.23 0.41 0.04 0.43
. Fuel consumption and CO2 emissions fluctuate when applying the test fuel. The results scatter falls within the accuracy limits of the measurement and it is difficult to conclude whether the test fuel has a positive effect on vehicle fuel consumption or not. For this purpose targeted tests are necessary. . Gaseous pollutants emissions are not affected when operating with the oil-diesel blends. In some cases NOx appear to increase slightly, but the overall emissions over NEDC never exceeded the Euro 3 emission limits for which the vehicle is certified. . Particle number and size distribution measurements showed that the presence of cottonseed oil may suppress the production of nucleation mode particles. . The acceleration tests showed that the test fuel did not affect the power output of the vehicle. . Problems were experienced at low ambient temperatures as the engine suffered from misfueling. These problems were tackled by replacing the fuel filter. The overall conclusion drawn is that the potential of vegetable oils is important and should not be neglected. Further studies are necessary in order to fully understand the mechanisms that affect the properties of cottonseed oil-diesel blends as well as their effects on vehicle operation.
REFERENCES Andre´, M., 2004, The ARTEMIS European driving cycles for measuring car pollutant emissions, Science of the Total Environment, 334– 335: 73–84. Bauer, H., Dietsche, K.H. and Jager, T., 2004, Diesel Engine Management (Robert Bosch GmbH, Automotive Aftermarket Division, Germany). Biofuels Research Advisory Council (BIOFRAC), 2006, Biofuels in the European Union, a vision for 2030 and beyond, available at: http://ec.europa.eu/research/energy/nn/nn_rt/nn_rt_bm/ article_4012_en.htm. Dekati Ltd, 2001, Sampling automotive exhaust with a thermodenuder, Dekati technical note. European Commission (EC), 2003, Directive 2003/30/EC of the European Parliament and of the Council of 8 May 2003 on the promotion of the use of biofuels or other renewable fuels for transport, available at: http://ec.europa.eu/energy/res/legislation/ biofuels_en.htm. Fontaras, G., Mamakos, A., Ntziachristos, L., Miltsios, G. and Samaras, Z., 2006, Evaluation of cottonseed oil –diesel fuel blends as fuel for automotive diesel engines, Fisita World Automotive Congress Yokohama Japan, Fisita technical paper number F2006P380.
Forson, F., Oduro, E. and Donkoh, E., 2004, Performance of jatropha oil blends in a diesel engine, Renewable Energy, 29: 1135 –1145. Giannelos, P.N., Sxizas, S., Lois, E., Zannikos, F. and Anastopoulos, G., 2005, Physical, chemical and fuel related properties of tomato seed oil for evaluating its direct use in diesel engines, Industrial Crops and Products, 22(3): 193– 199. Giechaskiel, B., Ntziachristos, L. and Samaras, Z., 2004, Calibration and modelling of ejector dilutors for automotive exhaust sampling, Measurement Science and Technology, 15: 1– 8. Graboski, M. and McCormick, R.L., 1998, Combustion of fat and vegetable oil derived fuels in diesel engines, Progress in Energy and Combustion Science, 24(2): 125 –164. He, Y. and Bao, Y., 2005, Study on cottonseed oil as a partial substitute for diesel oil in fuel for single cylinder diesel engine, Technical note, Renewable Energy, 30: 805 –813. International Energy Agency (IEA), 2004, Biofuels for transport: an international perspective, Paris, downloaded from: http:// www.iea.org/textbase/nppdf/free/2004/biofuels2004.pdf. Institut Francais du Petrole (IFP), 2005, Road transport fuels in Europe: the explosion of demand for diesel fuel, downloaded from: http://www.ifp.fr/IFP/en/files/cinfo/IFP-Panorama05_10CarburantsRoutiersVA.pdf. Marjama¨ki, M., Ntziachristos, L., Virtanen, A., Ristima¨ki, J., Keskinen, J., Moiso, M., Palonen, M. and Lappi, M., 2002. Electrical filter stage for the ELPI. SAE paper 2002-01-0055. Pundir, B.P., Singal, S.K. and Gondal, A.K., 1994, Diesel fuel quality: Engine performance and emissions, SAE Technical Paper Series, Paper No 942020. Rakopoulos, C.D., Antonopoulos, K.A., Rakopoulos, D.C., Hountalas, D.T. and Giakoumis, E.G., 2006, Comparative performance and emissions study of a direct injection diesel engine using blends of diesel fuel with vegetable oils or bio-diesels of various origins, Energy Conversion and Management, 47(18–19): 3272– 3287. Ramadhas, A., Jayaraj, S. and Muraleedharan, C., 2005, Characterization and effect of using rubber seed oil as fuel in the compression ignition engines, Technical note, Renewable Energy, 30:795–803. Virtanen, A., Marjama¨ki, M., Ristima¨ki, J. and Keskinen, J., 2001, Fine particle losses in electrical low-pressure impactor, Journal of Aerosol Science, 32: 389 –401. Zaher, F.A., Megahed, O.A. and El Kinawy, O.S., 2003, Utilization of used frying oil as diesel engine fuel, Energy Sources, 25: 819– 826.
ACKNOWLEDGEMENT Part of this experimental work was supported by the Greek Scholarships Foundation (IKY). The manuscript was received 5 February 2007 and accepted for publication after revision 13 May 2007.
Trans IChemE, Part B, Process Safety and Environmental Protection, 2007, 85(B5): 396 –403