Low Temperature Operability Test: Phase 2 - Impact of Saturated Monoglycerides on Heavy Duty Diesel Truck Operation

Transcription

1 Low Temperature Operability Test: Phase 2 - Impact of Saturated Monoglycerides on Heavy Duty Diesel Truck Operation Funded by Imperial Oil, Canadian Petroleum Products Institute and Natural Resources Canada under National Renewable Diesel Demonstration Initiative (NRDDI) Research conducted by Imperial Oil, Products and Chemicals Division Research Department Sarnia, Ontario, Canada R December, 29

2 TABLE OF CONTENTS SUMMARY (i) 1. INTRODUCTION EXPERIMENTAL Properties of Materials Used Definition of the SMG Spiked Fuels Preparation of the AWCD Fuels Test Vehicles Used in the AWCD Fuel System and Fuel Filter Configurations All Weather Chassis Dynamometer (AWCD) AWCD Test Protocol Testing of the SMG Spiked Bio-Diesel Fuels in the AWCD RESULTS AND DISCUSSION Test with 25 mg/l SMG in the Fuel Test with 2 mg/l SMG in the Fuel Test with 15 mg/l SMG in the Fuel CONCLUSION and RECOMMENDATION REFERENCES TABLES APPENDICES Page Disclaimer This report is produced with financial support from Natural Resources Canada. Its content does not necessarily reflect the opinions of the Government of Canada.

3 SUMMARY (i) Low Temperature Operability Test Phase 2: Impact of Saturated Monoglycerides on HDD Truck Operation The use of renewable fuels, such as biodiesel, in motor vehicle fuels is expected to grow rapidly in North America as a result of government mandates, both federal and state/provincial. Biodiesel is a fuel component made from plant or animal feedstocks through an esterification process. The resulting fatty acid methyl esters (FAME) have a large variation in cloud point from -5 C to +15 C, depending upon the source. In Canada where a large geographic area experiences a cold climate, the poor low temperature properties of FAME and blends containing same (wax, gelling, and phase separation above the Cloud Point) must be well understood to avoid operability issues. The need for this understanding is underscored by reports of field issues that have occurred in the United States and Europe. Investigations of these issues have implicated saturated mono-glycerides (SMG) as the cause of filter plugging. To gain a better understanding of these concerns, the low temperature operability, above the cloud point, of a B5 CME bio-diesel fuel spiked with three levels of saturated monoglycerides (SMG) was examined in 24/25 model year heavy-duty trucks in an All Weather Chassis Dynamometer. The levels of SMG were selected to give a Filter Blocking Tendency (ASTM D268) range of ~1 to 6 based on the results of a previous low temperature storage study. The study showed that the performance of the fuel delivery systems associated with the heavy duty diesel trucks tested was negatively impacted by the phase separation of spiked SMG from the renewable diesel fuel at -16 C well above its -26 C cloud point. A spiked SMG content of 25 mg/l resulted in an operability failure; 2 mg/l resulted in a high fuel filter pressure drop and restricted fuel re-circulation; 15 mg/l caused high fuel filter pressure drop without causing fuel re-circulation problems. The SMG on the fuel filter did not re-dissolve in the fuel as it warmed up, either via fuel re-circulation or by letting the truck warm up to ambient conditions. This is due to the high melting point (71 to 81 C) of individual SMG components. The performance of the fuel systems associated with the heavy duty diesel trucks tested correlated reasonably well with the FBT of the spiked fuel post a low temperature pre-soak. The data suggests that the renewable diesel blend should have a FBT in the 1 to 1.4 range similar to conventional diesel fuels. Future work to determine whether the SMG spiking procedure/level adversely affected the way precipitates form in renewable diesel fuels would be valuable.

4 The results of this study further confirm the previous reports in the literature regarding the deleterious impact of saturated monoglycerides in FAME on the low temperature operability of filters in fuel handling systems. This is a very important consideration when formulating renewable diesel fuels for a Canadian climate. The information generated by this study may be used to direct future renewable diesel blending formulations (e.g. seasonal bio-diesel for winter per conventional practice) and by standard-setting bodies to set specifications that will ensure "fit for service" fuel products.

5 1 1. INTRODUCTION The use of renewable fuels, such as biodiesel, in motor vehicle fuels is expected to grow rapidly in North America as a result of government mandates, both federal and state/provincial. Biodiesel is a fuel component made from plant or animal feedstocks through a trans-esterification process. The resulting fatty acid methyl esters (FAME) have a large variation in cloud point from - 5 C to +15 C, depending upon the source. In Canada where a large geographic area experiences a cold climate, the poor low temperature properties of FAME and blends containing same (wax, gelling, and phase separation above the cloud point) must be well understood to avoid operability issues. The need for this understanding is underscored by reports of field issues that have occurred in the United States and Europe. Unexpected filter plugging issues occurred in vehicles and dispensing filters during the winter of in Minnesota (1-3). The filter plugging was attributed by several sources to biodiesel that exceeded the <.24 wt% total glycerin limit of the ASTM D6571 standard (4). Analysis of the plugged filters identified "...a preponderance of saturated monoglycerides in the organic component."(2). In response to these problems, a cold soak filtration test consisting of chilling the B1 bio-diesel to 4.4 C for 16 hours and then allowing warming up to room temperature prior to filtering was developed and incorporated into the ASTM D Standard. Biodiesel produced to the new standard achieved a dramatic reduction in filter plugging issues during the winter of However, Flint Hill Resources reported the plugging of dispensing filters at -18 C caused by B2.5 soybean methyl ester (SME) in December 26 (1). Analysis of the material on the filters showed a substantial elevation of the saturated monoglycerides monopalmatin and monostearin. During the winter of 27 Sweden experiences cold temperatures for long periods of time which resulted in filter blocking in vehicles and precipitates in customer above ground tanks at temperatures above the cloud point of the fuel (5). Analyses of the precipitated material showed it to be saturated monoglycerides. B5 blends had been introduced into the market in the summer of 26. B2 had been successfully used in the Swedish market. Lab tests showed that the solubility of saturated monoglycerides in Swedish diesel to be low (i.e < 5 4 C). The monoglyceride specification was reduced to.3 wt% in Sweden versus.8 wt% in EN Filter plugging operability issues (3/day/service station) were reported in France during the winter of (6). The first episode was attributed to the fuel not meeting the EN59 specification. During the second episode all fuels tested met EN59 specs. The fuel filters were found to contain high levels of saturated monoglycerides. The authors pointed out that once formed, the saturated monoglyceride crystals do not re-dissolve in the fuel. They also pointed out that tests such as CFPP will not detect the problem due to the fast cooling rate used and that a cold filterability or other test methods need to be developed to address the issue. There have also been several research studies in this area. Infineum has also reported precipitation of saturated monoglycerides, above the cloud point, in Bxx biodiesel fuels. This was demonstrated by saturated monoglycerides (SMG) add-back experiments using IP 387 filter blocking tendency test method (7). Cosmo Oil Ltd examined the low temperature storage [laboratory as well as All Weather Chassis Dynamometer (AWCD) testing] of biodiesel fuel

6 2 blends (8). Their study showed that B5 Palm Methyl Ester (PME) stored at 1 C produced crystals of C 14 to C 18 saturated monoglycerides. To gain a better understanding of these concerns, a low temperature storage stability study (9) was performed on fifty seven bio-diesel fuels comprising of B, B2, B5 and B2 involving six commercial low cloud point base fuels and seven FAME comprising Canola Methyl Ester (CME), Soybean Methyl Ester (SME), Tallow Methyl Ester (TME) and PME. The bio-diesel fuels were stored at two different temperatures above the fuel cloud point for a period of 1 days. The low temperature stability was then assessed by the filter blocking tendency (FBT) test method ASTM D268 (1). This study further confirmed that filter plugging at low temperature can be the result of the phase separation of saturated monoglycerides from the fuel. It was also shown that the FBT of the fuel correlated reasonably well with its saturated monoglyceride (SMG) content. In this study, the low temperature operability, above the cloud point, of a B5 CME biodiesel fuel spiked with three levels of SMG was examined in 24/25 model year heavy-duty trucks in an All Weather Chassis Dynamometer (AWCD). The levels of SMG were selected to give a FBT range of ~1 to 6 based on the results of the low temperature storage stability study (9). The objective of this study was to define an acceptable level of saturated monoglycerides in the FAME that would not cause operability problems in the vehicles. The information generated by this study may be used to direct future renewable diesel blending formulations (e.g. seasonal bio-diesel for winter per conventional practice) and by standard-setting bodies to set specifications that will ensure "fit for service" fuel products. 2. EXPERIMENTAL 2.1 Properties of Materials Used The properties of the base fuel and the CME used in this study are summarized in Table 1 and 2, respectively. The approach was to spike a B5 CME bio-diesel fuel containing an insignificant amount of SMG with three different levels of "pure" saturated monoglyceride(s) such as glyceryl monostearate acquired from a supplier. The testing program was performed at -16 C or 1 C above the cloud point of the B5 CME (Table 1). Glyceryl monostearate (monostearin) was acquired from two different suppliers: Spectrum Chemicals and PCM/Ashland. Although both suppliers specified "pure" glyceryl monostearate, a significant amount of glyceryl palmitate (monopalmitin) was present in the products. Both products were deemed acceptable since monopalmitin and monostearin are present in the FAME. The glyceryl monostearate from Spectrum Chemicals was selected for this study. The C of A is presented in the Appendix 1. The GC-MS analysis of this product revealed the presence of 48.3 wt% monopalmitin and 51.7wt% monostearin (Appendix 2). 2.2 Definition of the SMG Spiked Fuels A B5 CME (Table 1) comprising of 95. vol% ULSD-29 with 5. vol% CME was selected as the base fuel for the study. Based on the filter blocking tendency results obtained from the low temperature storage stability study (9), we have focused on fuel blends containing from 1 to 35 mg/l SMG to give a FBT range of ~1to 6. The key challenge was to dissolve the saturated monoglycerides in the fuel since their melting points are between 71 and 81 C. Several solvents

7 3 such a diesel fuel and CME were examined. The CME was found to be the most appropriate solvent. As an example, 8 mg of SMG was added to 2 ml CME. The mixture was placed in an oven at 7-8 C for 3 to 4 hours until all the SMG has dissolved. The clear SMG/CME solution was then added to the B fuel at ambient temperature to produce 4 ml of a 2 mg/l SMG spiked B5 CME bio-diesel fuel. This procedure is similar to that used by ADM (11). Due to the limitations for the AWCD testing ( number of tests, 3 trucks and three different spiked fuels, replicate testing and time constraints), the test fuels could not be soaked more than 4-days prior to the test. Laboratory B5 CME blends containing from 1 mg/l to 35 mg/l of saturated monoglycerides (SMG) were stored at -16 C for 4 days followed by a filter blocking tendency determination (ASTM D268). The filter blocking tendency results on the spiked SMG samples are compared in Figure 1 (green circles) with the data from the previous study (9) and are in reasonable agreement considering that both the cold soak times and soak temperatures were different. Figure 1. Effect of SMG on FBT 12. FBT y =.1x x + 1 R 2 =.776 CME TME SME AWCD-P2 B SMG, mg/l in blended fuel 2.3 Preparation of the AWCD Test Fuels The AWCD test fuels were prepared in a 1L tote. The B ULSD-29 (47454) diesel fuel (76L) was poured into a clean tote and placed inside a garage at ambient (2 C) temperature. Forty litres of CME (BIO-5345) was poured into a clean metal pail. The required amount of Spectrum SMG was added to the CME. The CME/SMG mixture was heated to 7-8 C for 3 to 4 hours until complete dissolution of the SMG. The 4L of CME containing the SMG was added to the 76L base fuel. The bio-diesel fuel was then mixed using a recirculation pump. The fuel was then soaked at -16 C (test temperature) for a total of 84 to 9 hours. The protocol for the preparation of the test fuels is further described in the Appendix 3.The properties of the B ULSD-29 base fuel and the B5 CME fuel are presented in Table 1. The cloud point of the B5 CME was -26 C. Two 8 litre batches of each test fuel were prepared and used for the AWCD test program per the table below:

8 4 Test Fuel Base Fuel SMG, mg/l TF#1 Batch 1 B5 CME 25 TF#1 Batch 2 B5 CME 25 TF#2 Batch 1 B5 CME 2 TF#2 Batch 2 B5 CME 2 TF#3 Batch 1 B5 CME 15 TF#3 Batch 2 B5 CME Test Vehicles Used in the AWCD Three vehicles were selected as representative of the North American Heavy Duty truck market; one Freightliner mounted with a Detroit Diesel Series 6 engine and two Internationals equipped with a Cummins ISM engines. The 27-market share for these engines was about 18% and 25% respectively. The balance of the market is other OEMs. A description of the trucks with their fuel filter characteristics is presented below. Vehicle Engines Year Truck Fuel Filter Pore Size, µm Freightliner DD Series 6 24 D 7 International Cummins ISM 25 A 7 International Cummins ISM 25 B 7 All the trucks were chosen as standard production models with no special add-on equipment. Trucks A and B were essentially the same. 2.5 Fuel System and Fuel Filter Configurations The fuel system and fuel filter configurations are critical to the operability limit determination. The engine manufacturers were consulted and their recommendations followed. The two International test vehicles have the primary fuel filter housing located outside of the engine compartment and behind the cab. The Freightliner test vehicle has the fuel filter housing within the engine compartment. All fuel filters run at negative gauge pressure. For the Cummins ISM engines, the return fuel from the engine passes through the filter housing to warm up the in-coming fuel before returning to the fuel tank. For the Detroit Diesel engine, a 22 Watt electric heater is included in the fuel filter housing to warm up the in-coming fuel. The Freightliner Detroit Diesel truck has a faster fuel recirculation rate than the International Cummins ISM trucks. The fuel system schematics for the test trucks indicating the locations of pressure and temperature sensors are illustrated in Appendices 4 and 5.

9 5 2.6 All Weather Chassis Dynamometer (AWCD) Imperial Oil s AWCD is a state-of-the-art facility capable of cooling and testing three HDD trucks a day. The system consists of a dynamometer test cell and a cold soak room, which are cooled independently. Typically one truck is placed on the dynamometers and cooled in place overnight. Two other trucks are located in the cold soak room and cooled overnight to the same or different temperature as required. All vehicles are fitted with transducers and thermocouples to monitor parameters during cooling and test runs. The AWCD capabilities are described below. Independent temperature control in separate test cell and pre-soak area Temperature range: -4 C to +4 C Dual dynamometers for front, rear and 4WD vehicles, tandem axe trucks 112 kw absorption power per dynamometer Inertia simulation to 5, kg Maximum vehicle speed: 15 kph Synchronized wind velocity to 12 kph for cars and 65 kph for trucks Humidity range: 2 to 95% RH (above 5 C) Full computer control, data acquisition/data reduction and plotting 25 channels (2 thermocouples, 5 analogs) at up to 1 Hz The AWCD is illustrated with few photos presented in the Appendix AWCD Test Protocol The test protocol used in the dynamometer testing was the same as employed in previous studies (12-15) with the exception of the pre-soak procedure. Phase separation of a "waxy" material was observed in previous AWCD studies (8, 14, and 15). The CRC study (15) found that movement of the vehicle or deliberate mixing of the fuel prior to testing resulted in operability failure. The report concluded that "...movement of the truck could cause mixing of the vehicle fuel tank and suspension of the materials that had precipitated to the bottom causing a failure. This may indicate that vehicle dynamomoter tests do not capture all failure modes that can be experienced in actual vehicle operations". The Cosmo Oil study (8) showed that the formation of crystals of C14, C16 and C18 monoglycerides was complete in days. This is similar to the results reported in Reference 11. Based on this information, a procedure was developed in which the fuel was presoaked at -16 C from 84 to 9 hours before the test began. This temperature was chosen because it is well above the -26 C cloud point of the fuel and near to the temperature at which field issues as a result of SMG phase separation have been reported (1). The pre-soaked fuel was mixed via a recirculation pump prior to use. The trucks were fitted with pressure transducers before and after the fuel filter to measure any pressure drop across the filter due to filter plugging. Thermocouples were placed in the fuel tank, in the return line, and both before and after the fuel filter to measure temperature at those points. Each truck was flushed with fresh fuel and filled with approximately 2 liters of the test

10 6 fuel. Only one of the fuel tanks was used. Two trucks were then placed in the pre-soak room and one truck was placed directly on the dynamometers, in the test cell. During soaking, both air and fuel tank temperatures were recorded. After the soak period, the truck on the dynamometer was started using fully charged, warm batteries and ether spray to assist in the cold start. Heaters were used to keep the engine oil and brakes warm to avoid failures unrelated to the fuel system. Upon completing the test, the dynamometer test cell temperature was kept at -16 C to match the pre-soak room temperature. After the test, the truck was removed from the test cell and replaced by one of the trucks from the pre-soak room without starting the truck. This test protocol allowed the testing of three trucks per day. After the truck started, it was held at idle for 1 minutes. At the end of the idle period, the truck was accelerated through the normal gear changes to a speed of 8 kph and held at that speed. The total drive time was 6 minutes including acceleration. The acceleration to 8 kph typically requires 5 minutes or less. The fuel delivery system was monitored continuously for pressure and temperature at the measurement points illustrated in Appendices 3 and 4. The fuel delivery system associated with the International Cummins ISM Heavy Duty engine (Appendix 4) was found to be the most severe, likely due to slower fuel re-circulation and/or lack of an electric heater in the fuel filter housing. A passing AWCD result was one that completed the entire cycle without loss in speed and power. Failing results could include: failure to start, rough or stall at idle and failure to reach 8 kph. In addition to the measurements described above, the air temperature, engine speed, wind speed, tractive force, and oil and coolant temperatures were continuously monitored. 2.8 Testing of the SMG Spiked Bio-Diesel Fuels in the AWCD The fuel in the tote was well mixed with a re-circulation pump before transferring into the truck fuel tank. The fuel tank test temperature was set at -16 C. Depending upon the AWCD result, the next test was either repeated or continued using the same fuel filter. The decision on the next test was based on the pressure drop observed across the fuel filter obtained from the first test. The operability failure corresponds to a vehicle power loss due to the plugging of the fuel delivery system. A total of 18 AWCD tests were performed. A summary of the AWCD test results is presented in Table 3 with additional details provided in Appendix 7. The test profile for the 18 AWCD tests is presented in the Appendix 8. In general the program ran smoothly once the issue of fuel mixing was resolved after the first test. The data in Table 3 show that the tests were repeatable, both for the same truck (i.e. tests 7 & 12 for the Freightliner and tests 13 & 18 for the International truck A) and between the International trucks (i.e. tests 8 & 1 and 13 & 14). Nine out of the eighteen tests were selected to illustrate the effect of the SMG on the low temperature operability of the trucks. The results of the selected tests are discussed in the following sections. 3. RESULTS AND DISCUSSION 3.1 Test with 25 mg/l SMG in the Fuel

11 7 The results of three consecutive tests (#3, #4 & #9) on the International Truck B with the same fuel filter, is shown in Figures 2 to 4. Test #3 used Test Fuel # 1, Batch # 1 while tests #4 & #9 used Test Fuel #1, Batch #2. During test #3 the pressure drop across the fuel filter increased to 88 kpa after one hour and remained steady when the test was extended by 15 minutes. An additional 4 litres of fuel was added to the fuel tank while the truck was in the test cell and Test #4 initiated. The pressure drop across the fuel filter quickly rose to 8 kpa and lined out to 89 kpa by end of test. The truck was moved to the garage and kept at ambient until the next day at which time it was refuelled without changing the fuel filter and Test #9 initiated. After starting the engine, the pressure drop across the fuel filter increased quickly to 85 kpa after the 1 minute idle period and the truck could not accelerate resulting in a test failure. Two photos of the plugged fuel filter after the test are presented in Figures 5 and 6. The SMG can be observed on the fuel filter. 1 8 Figure 2 Truck Performance with 25 mg/l SMG in Fuel Truck "B" -16 C Test #3, TF# 1 Batch : :14 :28 :43 :57 1:12 1:26 1:4 FUELTANK C FRFILTER Kpa Vehicle Speed KPH

12 8 Figure 3 Truck Performance with 25 mg/l SMG in Fuel Truck "B" -16 C Test #4, TF# 1 Batch 1 ("old" fuel filter) : :14 :28 :43 :57 1:12 1:26 FUELTANK C FRFILTER Kpa DYM_Speed KPH Figure 4 Truck Performance with 25 mg/l in Fuel Truck "B" -16 C Test #9, TF# 1 Batch 2 ("old" fuel filter) : :1 :2 :4 :5 :7 :8 :1 :11 :12 :14 FUELTANK C FRFILTER kpa Vehicle Speed KPH

13 9 Figure 5. Plugged Filter with SMG Figure 6. Plugged Filter with SMG The Gas Chromatography-Mass Spectrometry (GC-MS) analysis of the material on the fuel filter confirmed it is a mixture of monopalmitin and monostearin (SMG). The gas chromatogram is presented in the Figure 7. Figure 7 GC-MS Analysis of the Waxy Deposit on the Plugged Fuel Filter Test #9 Abundance 1.6e+7 1.4e+7 monopalmitin TIC: 9243.D monostearin 1.2e+7 1e+7 Methyl oleate Time--> Methyl Oleate = residual fuel from B5 CME

14 1 These tests show that the SMG builds up on the fuel filter and does not re-dissolve in the fuel even after the truck was warmed up to ambient conditions. This is not unexpected due to the high melting point of monoplamitin and monosearin at 71 C and 81 C, respectively. The build up of SMG on the fuel filter results in fuel re-circulation issues as evidenced by the decrease in the endof-test (EOT) fuel tank temperature (see Table 3) ultimately ending in a test failure. 3.2 Test with 2 mg/l SMG in the Fuel The results of two consecutive tests (#8 & #11) on the International Truck A with the same fuel filter, is shown in Figures 8 & 9. Test #8 used Test Fuel # 2, Batch # 1 while test #11 used Test Fuel #2, Batch #2. During test #8 the pressure drop across the fuel filter only increased to 9 kpa after one hour but the fuel tank temperature at EOT was -6 C suggesting fuel re-circulation was restricted. The truck was moved to the garage and kept at ambient until the next day at which time it was re-fuelled without changing the fuel filter and Test #11 initiated. After starting the engine, the pressure drop across the fuel filter gradually increased to 84 kpa after 5 minutes of steady state drive and then stabilized. These two tests also show that SMG builds up on the fuel filter and does not re-dissolve when the truck is warmed up to ambient. Vehicle operability is affected as evidenced by the high pressure drop across the fuel filter and the low EOT fuel tank temperature. Figure 8 Truck Performance with 2 mg/l SMG in Fuel Truck "A", -16 C Test 8, TF# 2 Batch : :14 :28 :43 :57 1:12 1:26 FUELTANK C FR FILTER KPa Vehicle Speed KPH

15 11 Figure 9 Truck Performance with 2 mg/l SMG in Fuel Truck "A", -16 C Test #11, TF# 2 Batch : :14 :28 :43 :57 1:12 1:26 FUELTANK C FR FILTER KPa Vehicle Speed KPH 3.3 Tests with 15 mg/l SMG in the Fuel The results of two consecutive tests (#15 & #16) on the Freightliner Truck D with the same fuel filter, is shown in Figures 1 & 11. Test #15 used Test Fuel #3, Batch # 1 while test #16 used Test Fuel #3, Batch #2. During test #15 the pressure drop across the fuel filter increased to 26 kpa after one hour but the fuel tank temperature at EOT was 14 C suggesting fuel re-circulation wasn't restricted. The truck was moved to the garage and kept at ambient until the next day at which time it was re-fuelled without changing the fuel filter and Test #16 initiated. After starting the engine, the pressure drop across the fuel filter gradually increased to 7 kpa after 3 minutes of steady state drive and then stabilized. These two tests also show that SMG builds up on the fuel filter and does not re-dissolve when the truck is warmed up to ambient. The high pressure drop across the fuel filter did not appear to have caused a fuel re-circulation problem as indicated by the EOT fuel tank temperature.

16 12 Figure 1 Truck Performance with 15 mg/l SMG in Fuel Truck "D", -16 C Test #15, TF# 3 Batch : :14 :28 :43 :57 1:12 1:26 1:4 FUELTANK C FR FILTER KPa Vehicle Speed KPH Figure 11 Truck performance with 15 mg/l SMG in Fuel Truck "D" -16 C Test #16, TF# 3 Batch 2 ("old" fuel filter) : :14 :28 :43 :57 1:12 1:26 1:4 1:55 FUELTANK C FR FILTER kpa Vehicle Speed KPH

17 13 The above tests were repeated (#14 & #17) with the International Truck B with the same fuel filter and the results are shown in Figures 12 & 13. Test #14 used Test Fuel #3, Batch # 1 while test #17 used Test Fuel #3, Batch #2. During test #14 the pressure drop across the fuel filter increased to 1 kpa after one hour but the fuel tank temperature at EOT was 1 C suggesting fuel re-circulation wasn't restricted. The truck was moved to the garage and kept at ambient until the next day at which time it was re-fuelled without changing the fuel filter and Test #17 initiated. After starting the engine, the pressure drop across the fuel filter gradually increased to 65 kpa after 1 hour of steady state drive and the EOT fuel tank temperature increased to 2 C. These two tests also show that SMG builds up on the fuel filter and does not re-dissolve when the truck is warmed up to ambient. The high pressure drop across the fuel filter did not appear to have caused a fuel recirculation problem as indicated by the EOT fuel tank temperature. The performance of the Freightliner Truck D and the International Truck B were very similar in terms of EOT fuel filter pressure drop and fuel tank temperature. Figure 12 Truck Performance with 15 mg/l SMG in Fuel Truck "B" -16 C Test #14, TF# 3 Batch : :14 :28 :43 :57 1:12 1:26 FUELTANK C FR FILTER KPa Vehicle Speed KPH

18 14 1 Figure 13 Truck Performance with 15 mg/l in Fuel Truck "B" -16 C Test #17, TF# 3 Batch : :14 :28 :43 :57 1:12 1:26 1:4 FUELTANK C FR FILTER KPa Vehicle Speed KPH 4. CONCLUSIONS and RECOMMENDATION 1. The performance of the fuel delivery systems associated with the heavy duty diesel trucks tested was negatively impacted by the phase separation of SMG from the renewable diesel fuel at IE -16 well above its -26 C cloud point. A SMG content of 25 mg/l resulted in an operability failure while 2 mg/l resulted in a high pressure drop across the fuel filter and restricted fuel re-circulation. The 15 mg/l SMG content fuel caused high fuel filter pressure drop without causing fuel recirculation problems. 2. The SMG on the fuel filter did not re-dissolve in the fuel as it warmed up, either via fuel re-circulation or by letting the truck warm up to ambient conditions. This is due to the high melting point (71-81 C) of individual SMG components. 3. The performance of the fuel systems associated with the heavy duty diesel trucks tested correlated reasonably well with the FBT of the spiked fuel post a low temperature pre-soak. The data suggests that the renewable diesel blend should have a FBT in the 1 to 1.4 range similar to conventional diesel fuels. The results of this study further confirm the previous reports in the literature regarding the deleterious impact of saturated monoglycerides in FAME on the low temperature operability of filters in fuel handling systems. This is a very important consideration when formulating renewable diesel fuels for a Canadian climate. The information generated by this study may be used to direct future renewable diesel blending formulations (e.g. seasonal bio-diesel for winter per conventional

19 15 practice) and by standard-setting bodies to set specifications that will ensure "fit for service" fuel products. Possible future work should focus on determining if the SMG spiking procedure/level used in this study had a significant effect on precipitate formation. 5. REFERENCES 1. Charley Selvidge, Scott Blumenshine, Kurt Campbell, Cathy Dowell and Julie Stolis, "Effect of Biodiesel Impurities on Filterability and Phase Separation from Biodiesel and Biodiesel Blends", IASH 27, the 1 th International Conference on Stability, Handling and Use of Liquid Fuels, Tucson, AZ, October 5-11, Lisa Pfaltzgraf, Inmok Lee, James Foster and George Poppe, "The Effect of Minor Components on the Cloud Point and Filterability", Biodiesel Magazine, November Inmok Lee, Lisa M. Pfalzgraf, Geroge B. Poppe, Erica Powers and Troy Haines, "The Role of Sterol Glucosides on Filter Plugging", Biodiesel Magazine, April Innospec Performance Specialties Technical Memo, "Biodiesel-Potential Causes of Filter Blocking, Issue 1, February ASTM D6751-8, Standard Specification for Biodiesel Fuel Blend Stock (B1) for Middle Distillate Fuels. 5. M. Brewer, "Identification of precipitate found in depot storage tanks containing Swedish Klass1 B5 fuels", International Congress on Biodiesel, November 27, Vienna, Austria. 6. R. Faucon, A.Gendron, and O. Cottalorda, "Diesel Fuel B7 Specifications Need to be Reinforced for Cold Weather Conditions", World Refining Fuels Conference, Bruxelles, May FAME Cold Flow Properties - The Challenges of Measurement, Brian Davis and Vincent Denecker, Infineum, April Nobuyasu Ohshio, Kazuhisa Saito, Shuichi Kobayashi and Shigeyuki Tanaka, "Storage Stability of FAME Blended Diesel Fuels", SAE Paper , presented at the Powertrains, Fuels and Lubricants Meeting, Chicago, IL, October 6-9, Low Temperature Storage Test Phase 2 -Identification of Problem Species report prepared By Imperial Oil for Natural Resources Canada under the NRDDI program. 1. ASTM D268-8, Standard Test Method for Determining Filter Blocking Tendency.

20 Inmok Lee, Lisa M. Pfalzgraf, Geroge B. Poppe, Erica Powers and Troy Haines, "The Role of Sterol Glucosides on Filter Plugging", Biodiesel Magazine, April Chandler John E., "Comparison of All Weather Chassis Dynamometer Low Temperature Operability Limits for Heavy and Light Duty trucks with Standard Laboratory Test Methods", SAE Paper Chandler John E. and Zechman Jennifer A., "Low Temperature Operability Limits of Late Model Heavy Duty Diesel Trucks and the Effect of Operability Additives and Changes to the Fuel Delivery System have on Low Temperature Performance", SAE Paper Marc-André Poirier, Patrick K. Lai and Larry Lawlor, " The Effect of Biodiesel Fuels on the Low Temperature Operability of North American Heavy Duty Diesel Trucks", SAE Paper , presented at the Powertrains, Fuels and Lubricants Meeting, Chicago, IL, October 6-9, CRC Project: DP-2a-7, June 28.

21 17 6. TABLES

22 18 Table 1. Properties of the Petroleum Diesel Fuel Sample ID Product ULSD-29 B5 CME CGSB Type Type B Type B Type B ULSD No. 2 Limits Properties Min Max Appearance C&B C&B Density, kg/m Flash Point, C C, mm 2 /s Aromatics, wt% Sulphur, mg/kg Cetane Index Cloud Point, C C, ps/m Distillation D86 IBP % % % % % % % % % % % FBP SMG, mg/l -calculated 1 1. For a fuel designed for an operability temperature colder than -2 C.

23 19 Table 2. Properties of CME Sample ID BIO ASTM D6751 EN Properties CME Min Max Min Max 15 C, kg/m C, mm 2 /s Total sulphur,mg/kg Water & Sediment, vol% Water, mg/kg Cloud Point, C Flash Point, C Acid Number, mg KOH/g Carbon Residue, wt% Carbon Residue, 1% distillation residu, wt% Copper Corrosion 1a 3 Cetane Number Oxidation Stability, hours Sulfated Ash, wt%..2.2 Free Glycerin, wt% Total Glycerin, wt% Monoglycerides, wt% Diglycerides, wt% Triglycerides, wt% < SMG,mg/kg 236 Cold Soak Filtration, sec For a fuel designed for an operability temperature colder than -12 C.

24 2 Table 3. Summary of the AWCD Test Results Test # Truck Fuel Filter P Fuel Tank Comments Volume, L kpa, (EOT) Temp. C (EOT) A. Test Fuel #1 Spiked with 25 mg/l SMG 1 D Fuel not properly mixed - Pass 5 D Repeat of Test #1 - Pass 2 A Fuel re-circulation problems - Pass 6 A Repeat of Test #2 - Pass 3 B High P at EOT - Pass 4 B Continuation of Test #3 - Pass 9 B Con't Test # 3 & 4 - Fail B. Test Fuel #2 Spiked with 2 mg/l SMG 7 D Pass 12 D Repeat of Test #7 - Pass 8 A Fuel re-circulation problems - Pass 11 A Continuation of Test #8 - Pass 1 B Repeat of Test #8 - Pass C. Test Fuel #3 Spiked with 15 mg/l SMG 13 A Pass 18 A Repeat of Test #13 - Pass 14 B Repeat of Tests 13 & 18 - Pass 17 B Continuation of Test #14 - Pass 15 D Test run 1h3 min - Pass 16 D Continuation Test #15, 1h4 min - Pass Truck D = Freightliner Truck A = International Truck B = International

25 21 7. APPENDICES

26 22 Appendix 1 CERTIFICATE OF ANALYSIS Item Number GL149 Lot Number WS393 Item GLYCERYL MONOSTEARATE, POWDER, FOOD GRADE CAS Number Molecular Formula C 21 H 42 O 4 Molecular Weight TEST SPECIFICATION MIN MAX RESULT MONOGLYCERIDE CONTENT 9. % 97.3 % FREE GLYCEROL (DIGLYCEROL AND GLYCEROL) %.4 % ACID VALUE IDENTIFICATION TO PASS TEST PASSES TEST EXPIRATION DATE 8-AUG-29 MANUFACTURE DATE 8-AUG-26

27 23 Abundance Appendix 2. GC/MS of SMG Purchased from Spectrum Ion range 33. to 65.: D 6 5 Glyceryl palmitate Glyceryl stearate 4 3 Degradation products 2 1 Time-->

28 24 Appendix 3 Protocol for the Preparation of the AWCD Fuels Preparation of the 25 mg/l SMG spiked fuels (Test Fuel 1, Batch # 1 and Batch 2) 1. Transfer 76L of the ULSD-29 (47454) using a calibrated metering pump into a 1L tote at 2-24 C. 1. Place 4L of Canola Methyl Ester (CME) (BIO-5345) in a clean metal container(s). 2. Weigh 2. ±.1 gm of SMG (Chemical-51513) and add it into the 4L CME. 3. Place the container(s) with the CME in an oven and heat to 7-8 C and until complete dissolution of the SMG. Stir occasionally with a paddle to help the dissolution of the SMG. This step took approximately 3 to 4 hours. 4. Add the warm 4L CME spiked SMG to the 76L fuel. 5. Mix the 8L B5 CME spiked SMG using a recirculation pump for about 1 hour. 6. The tote is then soaked at -16 C (test temperature) in the pre-soak room, for 84 to 9 hours. Preparation of the 2 mg/l SMG spiked fuels (Test Fuel 2, Batch # 1 and Batch 2) 1. Transfer 76L of the ULSD-29 (47454) using a calibrated metering pump into a 1L tote at 2-24 C. 2. Place 4L of Canola Methyl Ester (CME) (BIO-5345) in a clean metal container(s). 3. Weigh 16. ±.1 gm of SMG (Chemical-51513) and add it into the 4L CME. 4. Repeat steps 4 to 7 in the above section. Preparation of the 15 mg/l SMG spiked fuels (Test Fuel 3, Batch # 1 and Batch 2) Transfer 76L of the ULSD-29 (47454) using a calibrated metering pump into a 1L tote at 2-24 C. 2. Place 4L of Canola Methyl Ester (CME) (BIO-5345) in a clean metal container(s). 3. Weigh 12. ±.1 gm of SMG (Chemical-51513) and add it into the 4L CME. 4. Repeat steps 4 to 7 in the above section.

29 25 Appendix 4. Fuel Delivery System for the Detroit Diesel Series 6 (Truck D)

30 26 Appendix 5. Fuel Delivery System for the Cummins ISM (Trucks A and B)

31 27 Appendix 6

32 28 Appendix 7 Summary of AWCD Tests Tes t No File Name Truck Fuel IO1139A Truck D TF #1 B1 IO1139B Truck A TF #1 B1 IO1139 C Truck B TF #1 B1 IO1159A Truck B TF #1 B1 IO1159B Truck D TF #1 B2 IO1159 C Truck A TF #1 B2 IO1189A Truck D TF #2 B1 IO1189B Truck A TF #2 B1 IO1189 C Truck B TF #1 B2 Comments 68 hr soak in tote, 16 hr soak in truck. Fuel in tote re-circulated by pump for 2 minutes before used. Sample taken when filling truck looked clear at ambient temp. and after 16 hr cold soak. Very small delta pressure (13 kpa max) build-up during 1 hr steady state drive. Fuel likely not mixed enough resulted in less SMG concentration 68 hr soak in tote, 16 hr soak in truck. Sample taken when filling truck looked hazy at ambient temp. and after 16 hr cold soak. Delta pressure build-up to 6 kpa at end of 1 hr steady state drive. After filling first truck from tote, it was moved back into pre-soak room to be kept cold until usage. 68 hr soak in tote, 16 hr soak in truck. Sample taken when filling truck looked hazy at ambient temp. and after 16 hr cold soak. Delta pressure build-up to 86 kpa after 1 hr of steady state drive and maintained steady when test extended for another 15 minutes. Truck remained in test cell at -16 C after the previous test. Added 4L of TF#1 B2 already has 48 hr cold soak. Filter was not changed. Further soaked truck with fuel for ~4 hr. After test started, delta pressure went quickly to >8 kpa but only increased slightly up to 89 kpa at the end of 1 hr steady state drive. 68 hr soak in tote, 16 hr soak in truck. Fuel in tote re-circulated by pump for 6 minutes, and physically agitated before used. Sample taken when filling truck looked slightly hazy and moderate precipitation at ambient temp. and after 16 hr cold soak. Delta pressure build-up to 28 kpa at 5 minutes of 1 hr steady state drive, and then stabilized 68 hr soak in tote, 16 hr soak in truck. Truck filled from tote without further recirculation. Sample taken when filling truck looked slightly hazy and moderate precipitation at ambient temp. and after cold soak. Delta pressure eventually built up to 33 kpa after extending the steady state drive by 15 minutes. It then stabilized. 44 hr soak in tote, 4 hr soak in truck due to weekend schedule. Fuel in tote recirculated by pump for 6 minutes, and physically agitated before used. This mixing procedure became standard from this point on. Sample taken when filling truck looked slightly hazy and moderate precipitation at ambient temp. and after 16 hr cold soak. Delta pressure remained low (12 kpa max) through out 1 hr steady state drive. 44 hr soak in tote, 4 hr soak in truck. Truck filled from tote without further recirculation. Sample taken when filling truck looked very slightly hazy and moderate precipitation at ambient temp. and after cold soak. Delta pressure remained low (9 kpa max) through out 1 hr steady state drive. After the previous test (#4), truck moved to garage and kept at ambient until following day. Fuel drained from tank. Added 2L of TF#1 B2 previously has 84 hr of cold soak. Filter was not changed. After engine started, delta P went quickly to 85 kpa by the end of 1 minute idle. Could not accelerate (remained at first gear) and test terminated.

33 29 Appendix 7 Summary of AWCD Tests (Cont'd) Tes t No. File Name Truck Fuel Comments IO129A Truck B TF #2 B1 IO129B Truck A TF #2 B2 IO129 C Truck D TF #2 B2 68 hr soak in tote, 16 hr soak in truck. Sample taken when filling truck looked slightly hazy and moderate precipitation at ambient temp. and after 16 hr cold soak. Filter out pressure decreased continuously to -28 kpa at end of 1 hr steady state drive. After the previous test (#8), truck moved to garage and kept at ambient until following day. Fuel tank was drained and added 2L of TF#2 B2 already cold soaked for 68 hr. Filter was not changed. Truck further soaked for 16 hr. After engine started, filter out pressure went gradually to -84 kpa by 5 minutes of steady state drive, and then stabilized. 68 hr soak in tote, 16 hr soak in truck. Sample taken when filling truck looked very slightly hazy and moderate precipitation at ambient temp. and after 16 hr cold soak. Filter out pressure decreased continuously to -21 kpa by 4 minutes of steady state drive, and then stabilized and increased slightly. 13 IO1229A Truck A TF #3 B1 68 hr soak in tote, 16 hr soak in truck. Sample taken when filling truck looked hazy at ambient temp. and after cold soak. Delta pressure build-up to 15 kpa at end of 1 hr steady state drive. 14 IO1229B Truck B TF #3 B1 68 hr soak in tote, 16 hr soak in truck. Sample taken when filling truck looked hazy at ambient temp. and after cold soak. Delta pressure build-up to 19 kpa at end of 1 hr steady state drive IO1229 C Truck D TF #3 B1 IO1259A Truck D TF #3 B2 IO1259B Truck B TF #3 B2 IO1259 C Truck A TF #3 B2 68 hr soak in tote, 16 hr soak in truck. Added 317 L of fuel. Delta pressure build-up to 28 kpa at end of 1 hr steady state drive and started to decrease. After the previous test (#15), truck moved to garage and kept at ambient until following day. Fuel tank was drained and added 35L of TF#3 B2 already cold soaked for 68 hr. Filter was not changed. Truck further soaked for 16 hr. Delta pressure build-up to 7 kpa at about 3 minutes of steady state drive and reached 75 kpa at one hour at its maximum. After the previous test (#14), truck moved to garage and kept at ambient until following day. Fuel tank was drained and added 2L of TF#3 B2 already cold soaked for 68 hr. Filter was not changed. Truck further soaked for 16 hr. Delta pressure build-up to 72 kpa after one hour of steady state drive then started to recover. 68 hr soak in tote, 16 hr soak in truck. Delta pressure build-up to 25 kpa at end of 1 hr steady state drive. Note: EOT (end of test) samples were taken from the fuel tank of all tests. They were generally clear at ambient temperature.

34 3 Appendix 8 SSE Biodiesel Program Truck "D" -16 C Test # 1, TF# 1 Batch Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYF_Speed KPH ENGSPD RPM SSE Biodiesel Program Truck "A" -16 C Test # 2, TF# 1 Batch Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYM_Speed KPH ENGSPD RPM

35 31 Appendix 8 SSE Biodiesel Program Truck "B" -16 C Test #3, TF# 1 Batch Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 1:4 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYM_Speed KPH ENGSPD RPM SSE Biodiesel Program Truck "B" -16 C Test #4, TF# 1 Batch 1 (continued) Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYM_Speed KPH ENGSPD RPM

36 32 Appendix 8 SSE Biodiesel Program Truck "D" -16 C Test #5, TF# 1 Batch Temp. (C) or Press. (kpa) or Speed (kph) Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYF_Speed KPH ENGSPD RPM SSE Biodiesel Program Truck "A" -16 C Test #6, TF# 1 Batch Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 1:4 1:55 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYM_Speed KPH ENGSPD RPM

37 33 Appendix 8 SSE Biodiesel Program Truck "D" -16 C Test # 7, TF# 2 Batch Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYF_Speed KPH ENGSPD RPM SSE Biodiesel Program Truck "A" -16 C Test 8, TF# 2 Batch Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYM_Speed KPH ENGSPD RPM

38 34 Appendix 8 SSE Biodiesel Program Truck "B" -16 C Test #9, TF# 1 Batch 2 (old filter) Engine Speed (rpm) : :1 :2 :4 :5 :7 :8 :1 :11 :12 :14 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYM_Speed KPH ENGSPD RPM SSE Biodiesel Program Truck "B" -16 C Test #1, TF# 2 Batch Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYM_Speed KPH ENGSPD RPM

39 35 Appendix 8 SSE Biodiesel Program Truck "A" -16 C Test #11, TF# 2 Batch 2 (old filter) Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYM_Speed KPH ENGSPD RPM SSE Biodiesel Program Truck "D" -16 C Test #12, TF# 2 Batch Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYF_Speed KPH ENGSPD RPM

40 36 Appendix 8 SSE Biodiesel Program Truck "A" -16 C Test #13, TF# 3 Batch Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYM_Speed KPH ENGSPD RPM SSE Biodiesel Program Truck "B" -16 C Test #14, TF# 3 Batch Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYM_Speed KPH ENGSPD RPM

41 37 Appendix 8 SSE Biodiesel Program Truck "D" -16 C Test #15, TF# 3 Batch 1 (317 L) Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 1:4 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYF_Speed KPH ENGSPD RPM SSE Biodiesel Program Truck "D" -16 C Test #16, TF# 3 Batch 2 (old filter + 35 L) Engine Speed (rpm) -1 : :14 :28 :43 :57 1:12 1:26 1:4 1:55 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYF_Speed KPH ENGSPD RPM

42 38 Appendix 8 SSE Biodiesel Program Truck "B" -16 C Test #17, TF# 3 Batch 2 (old filter + 2 L) Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 1:4 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYM_Speed KPH ENGSPD RPM SSE Biodiesel Program Truck "A" -16 C Test #18, TF# 3 Batch Temp. ( C) or Press (kpa) or Speed (kph) Engine Speed (rpm) : :14 :28 :43 :57 1:12 1:26 TOFFILTER C FRFFILTER C FUELTANK C TOFILTER Kpa FRFILTER Kpa DYM_Speed KPH ENGSPD RPM