An Experimental Study on the Spray, Combustion, and Emission

---

Article pubs.acs.org/EF

An Experimental Study on the Spray, Combustion, and Emission Characteristics of Two Types of Biodiesel Fuel Hyungik Kim, Yungjin Kim, and Kihyung Lee* †

Department of Mechanical Engineering, Hanyang University, 1271 Sa3-dong, Sangrok-gu, Ansan-si, Gyeonggi-do, 426-791, Korea ABSTRACT: In this study, we produced pure biodiesel and analyzed its properties. Also, pure biodiesel was blended with conventional diesel fuel in volumetric mixing rates of 5%, 20%, and 35%. Therefore, a total seven kinds of fuels were tested for this study. Injection quantity was measured according to the blending rate. Spray penetration, distribution, and evaporation characteristics were also clarified through a spray visualization system. Furthermore, the combustion flame was investigated using a single cylinder optical engine and a high-speed camera to understand how evaporation characteristics affect the combustion flame. Moreover, exhaust emissions were measured with an exhaust emission analyzer and a smoke meter. According to the results of the injection test, there were few differences between the characteristics of diesel and biodiesel-blended fuel regarding the injection quantity and penetration length when the fuel temperature was maintained at 60 °C. Hence, there were not any problems such as additional correction of the injection duration and pressure found to be used for the engine tests as long as the fuel temperature was properly controlled. From the spray RMS images, two types of biodiesel-blended fuels showed nonuniform spray distributions compared to diesel, indicating that biodiesel-blended fuel had worse evaporation characteristics. The flame visualization experiment showed that the combustion flame characteristics of diesel and biodiesel-blended fuels were similar; however, white smoke was observed at the start of ignition with the biodiesel-blended fuel due to its poor atomization characteristics. From the emission results, the biodiesel-blended fuels showed higher levels of NOx emissions compared to diesel, and it could be expected that the oxygen content and rapid combustion of the biodiesel-blended fuel amplified NOx generation. Also, as a result of combustion image analysis through an image processing, the PM distribution rates of biodiesel-blended fuels were significantly lower than that of diesel, and it showed a similar trend to the exhaust emission data. Therefore, it was judged to be a useful and valid method for analyzing the relationship between combustion and emission. disappeared above 70 °C, and the value decreased in the order of canola > soybean > fish oil. Kinast3 analyzed the production processes of biofuels and studied the characteristics of various biofuels derived from a diverse collection of seeds. Also, he suggested test results about biodiesel/diesel blended fuel. He conducted a componential analysis after biofuel production through the ASTM method and mentioned that componential analysis is essential for potential solutions to problems encountered during the production of biofuels. Furthermore, quantification of the component characteristics classified by concentration was studied through the componential analysis of biodiesel/diesel blended fuel. Canola and soybean blended fuels with mixing rates of 5, 20, and 35% had no significant differences in viscosity. Lee et al.4 analyzed spray and combustion characteristics of biodiesel for common-rail diesel engines. They analyzed physical characteristics of biodiesel according to mixing rate using a PDPA system and compared its penetration with that of diesel. Also, they measured characteristics of combustion and exhaust emissions using a single cylinder engine. Higher concentrations of biodiesel increased kinematic viscosity, surface tension, and cetane number. This is the cause that goes into the other attributes compared with commercial diesel

1. INTRODUCTION Recently, the internal combustion engine has been strongly challenged by hybrid and electric vehicles due to depletion of fossil fuels and environmental factors. In other words, internal combustion engines are facing a new assignment to reduce carbon dioxide emissions and increase fuel economy. For the diesel engine, many emission control techniques have been developed to produce better fuel economy than that of gasoline. Furthermore, numerous alternative energies have been applied to the engine system. Biodiesel, one alternative energy source, has been studied for use in the diesel system.1 Biodiesel can be produced using various sources such as vegetables, timber, and algae. Also, its characteristics are very similar to diesel fuel. Moreover, it is possible that conventional diesel combustion systems can be used without modification. Therefore, the biofuel characteristics of spray behavior and engine performance have been studied. Tate et al.2 compared the characteristics of viscosity according to temperature from 20 to 300 °C. They quantitatively studied the change of viscosity with temperature using canola, soybean, and ethyl esters of fish oil. They measured kinematic viscosity from 20 to 300 °C at intervals of 20° based on ASTM D 88 using modified saybolt viscometry. The correlation between MSS and kinematic viscosity is listed by formula in conjunction with the associated error. The viscosity values of the three biofuel types decreased with increasing temperature and became similar to that of diesel. Also, the viscosity difference between biofuel and diesel almost © 2013 American Chemical Society

Received: October 4, 2012 Revised: July 26, 2013 Published: July 30, 2013 5182

dx.doi.org/10.1021/ef400936a | Energy Fuels 2013, 27, 5182−5191

Energy & Fuels

Article

fuel. Also, the high viscosity value of biodiesel leads to increased friction at the nozzle so that the injection pressure of biodiesel needs to be higher than that of commercial diesel fuel. Finally, HC decreased but NOx increased because of the increased oxygen content. Qi et al.5 carried out an experimental study using soybean biodiesel with various mixing rates. They suggested the results of BSFC, BTE, emission characteristics, cylinder pressure, and RoHR under the several load conditions using a single cylinder DI diesel engine. They identified, experimentally, that the BSFC of the soybean biodiesel increased, and the BTE decreased because the biodiesel has a lower LHV compared to diesel fuel. In addition, the biodiesel had better combustion and high combustion chamber temperatures caused by the oxygen contents. Therefore, they claimed that NOx increased, and HC had no evident variation for all tested fuels. Also, they concluded that the characteristics of PM and CO were improved. From this paper, they concluded that the oxygen content was a key factor, and biodiesel was very likely to replace diesel fuel. In this paper, we analyzed canola and soybean blended fuel. We produced biodiesel and then analyzed its properties based on ASTM methods. In addition, we carried out spray visualization using a high pressure chamber and high-speed camera in order to analyze spray characteristics. Furthermore, we conducted a combustion visualization test using a single cylinder optical engine to investigate relationships between combustion flames and spray characteristics. Therefore, we analyzed the combustion and emission characteristics of the soybean and canola biodiesel.

Table 2. Experimental Conditions of Injection Quantity Test description

conditions

fuel type injection pressure (bar) injection duration (msec) fuel temperature (°C) ambient pressure (bar) ambient temperature (°C)

diesel, BD5%, 20%, 35% (canola, soybean) 800, 1000, 1200 0.3−1.2 60 1 (atmosphric) 25

Figure 1. Schematic of the injection quantity measurement device.

2. EXPERIMENTAL DEVICES AND CONDITIONS manufactured by TEMS Co. and Zenobalti Co., respectively. Furthermore, the common-rail injector was a seven-hole solenoid type injector, and both high-pressure and low-pressure pumps, commercialized in CRDI engines, were employed. 2.3. Spray Visualization. To analyze both spray uniformity and characteristics of atomization, we designed a spray visualization device to simulate ambient temperature and pressure of the engine cylinder. We can control temperature and pressure using an electric heater and N2 gas, respectively. The high pressure chamber has three visual windows composed of quartz glass. Furthermore, we used a xenon lamp as a light source. A high-speed camera that can collect 12 000 frames per second was set up at the bottom of the chamber. Specifications of the high-speed camera are listed in Table 3. The spray image size was set to 400 × 400. Also, camera exposure time and frame speed were 50 μs and 12 000 fps, respectively. A Bosch seven-hole solenoid type injector with a 148° spray angle was used. Figures 2 and 3 show the schematic of the spray visualization device and the spray angle and penetration, respectively. As shown in Figure 2, we used a pulse generator, injector driver, PCV controller, high-pressure pump, and low-pressure pump to control this system. We fixed the injection

2.1. Biodiesel Production. To produce pure biodiesel, we designed a biodiesel-production machine which uses a transesterification process to exchange the organic group R″ of an ester with the organic group R′ of an alcohol.6,7 These reactions are often catalyzed by the addition of an acid or base catalyst. Table 1 lists specifications of the biodiesel-production machine.

Table 1. Specifications of the Biodiesel-Producing Machine production standard type one-time production total production per day operation characteristics of composition rated voltage

530(D) × 500(W) × 810(H) batch 36∼40 L 144∼160 L semiautomatic installation of cooking oil filter tray 220 V, 60 Hz

In this study, we produced two types of pure biodiesel, one based on soybean and another on canola oil, to analyze their properties. These biodiesel fuels were blended with conventional diesel in mixing rates of 5%, 20%, and 35% by volume for the experiments of injection quantity, spray visualization, and combustion flame visualization. 2.2. Injection Quantity. We organized the injection quantity measurement device as follows in order to determine injection quantity according to injection pressure and duration. This device was composed of a pulse generator, injector driver, PCV controller, highpressure pump, and low-pressure pump. We repeated the same process 200 times according to injection pressure and duration. Then, we measured total injection quantities and calculated the average. Fuel type and experimental conditions are listed in Table 2. Also, the schematic of the injection quantity measurement device is shown in Figure 1, where the model 555 pulse/delay generator was manufactured by BNC. The injector driver and PCV controller were

Table 3. Specifications of the High-Speed Camera description

specification

model sensor sensitivity pictures per second (PPS) exposure time

Phantom 7.0 (Vision Research) 800 × 600 pixel 24 bit color array 1200 ISO/ASA color full sensor: to 4800 pps 400 × 400 to 12 000 pps variable to 2 ms independent of sample rate (pps) continuously variable pre/post Nikon mount standard TTL pulse

trigger lens mounts sync image 5183

dx.doi.org/10.1021/ef400936a | Energy Fuels 2013, 27, 5182−5191

Energy & Fuels

Article

mean of pixels of each image from eq 1, and the RMS average image (Srms) was expressed by using eq 2 as the value of pixels of each frame.11 n

Smean(x , y) =

∑k = 1 Sk(x , y) (1)

n n

Srms(x , y) =

∑k = 1 {Sk(x , y) × Sk(x , y)} n

(2)

where Smean(x,y) is the mean averaged image, Srms(x,y) is the RMS averaged spray image, Sk(x,y) is the spray (foreground) image after correcting and denoising, and n is the number of images. 2.4. Flame Visualization and Exhaust Emission Measurement. In this study, we used a single cylinder optical engine to directly observe the combustion flame. This equipment has an extended piston with a quartz window to see the inside of the cylinder head, as shown in Figure 4, and is driven by a DC dynamometer. The quartz was

Figure 2. Schematic of the spray visualization device.

Figure 4. Schematic of the single cylinder optical engine. attached to the top of the expended piston to allow collection of flame images using the high-speed camera. To control combustion, we manipulated the injector pressure and injection signal through a PCV controller and injector driver, respectively. Also, we controlled the coolant temperature using a water temperature controller (WTC) to maintain a constant temperature at 80 °C. A schematic of this device is shown in Figure 5, and specifications of a single cylinder optical engine are listed in Table 5. As shown in Figure 5, the timing pulse generator receives signals from both the TDC sensor at a cam-shaft and from the encoder at the fly wheel in order to control the injection signal. The timing pulse generator (TPG-28MP) was manufactured by Blue-planet Co. and the injector driver (TDA-3200H) by TEMS Co. We also used a pressure sensor (6056A) made by Kisler Co. to obtain combustion pressure. Furthermore, we used an exhaust gas analyzer (MEXADEGR 7100) made by Horiba and a smoke meter made by AVL to analyze exhaust emissions. Table 6 lists specifications of exhaust emission analyzers. In this study, we applied fixed conditions to compare combustion flames depending on the type of fuel. We fixed the rotational speed of the engine at 1000 rpm and injection pressure at 1000 bar. We also maintained the injection quantity as 20 mm3. Furthermore, injection timing was fixed at 5° BTDC, and fuel temperature and coolant temperature were maintained at 60 and 80 °C, respectively. Experimental conditions are listed in Table 7.

Figure 3. Definition of spray angle and radial penetration. quantity at 20 mm3. The injection pressure was 1000 bar, and ambient pressure was 40 bar. Also, we maintained fuel temperature at 60 °C. We used three types of fuel (diesel and biodiesel mixes of 5, 20, and 35% soybean and canola). Details are shown in Table 4.

Table 4. Experimental Conditions of Spray Visualization description

conditions

fuel type injection pressure (bar) ambient pressure (bar) injection quantity (mm3) fuel temperature (°C) ambient temperature (°C)

diesel, BD5%, 20%, 35% (canola, soybean) 1000 40 20 (fixed) 60 80

In general, the quantitative analyzing method of spray image is to average many images.8−10 However, this usual procedure is needed a certain amount of times. Therefore, in this study, we carried out image processing using Matlab code (Matrox) with high-speed camera images. The RMS (root mean square) method calculated a comparison of spray stems during the injection period. The distribution of spray can be measured by using the RMS method. The arithmetic mean image (Smean) was extracted using the arithmetic

3. EXPERIMENTAL RESULTS 3.1. Fuel Characteristics of Produced Biodiesel. In this study, we produced pure biodiesel made of soybean and canola using a biodiesel-producing machine. From these biodiesel 5184

dx.doi.org/10.1021/ef400936a | Energy Fuels 2013, 27, 5182−5191

Energy & Fuels

Article

Table 8. Properties of Diesel and Biodiesel description

Table 5. Specifications of a Single Cylinder Optical Engine conditions

engine type number of cylinders bore × stroke (mm) displacement volume (cc) compression ratio connecting rod length (mm)

four-stroke direct injection 1 91 × 96 624 17.7:1 148.5

Table 6. Specifications of Exhaust Emission Analyzers emissions

measurement principle

model

THC CO, CO2, O2 NOx PM (soot)

flame ionization detector non-dispersive infrared rays chemiluminescent filter smoke number

Horiba-7100DEGR Horiba-7100DEGR Horiba-7100DEGR AVL-415S

Table 7. Experimental Conditions of Combustion Visualization description

conditions

fuel type engine speed (rpm) injection pressure (bar) injection quantity (mm3) injection timing (deg) fuel temperature (°C) coolant temperature (°C)

diesel, BD5%, 20%, 35% (canola, soybean) 1000 1000 20 (fixed) BTDC 5 60 80

canola

soybean

40−55 42.8 870 2.67

53.9 38.4507 885.1 4.425

59 39.11 886.3 4.207

−15 −35 60−80 87 13 0 0.05 max

CFPP (°C)

−33

−9 −11 84 77.68 12.25 10.79 0.0−0.0024 max −1

0 −2 171 77.95 11.98 11.40 0.0−0.0024 max −9

and tear of parts. Also, biodiesel has a higher cloud and pour point than those of diesel. Because of this, driving during winter may be disadvantageous. Also, the oxygen content is very important in these fuels. Biodiesel has an almost 10% oxygen content which is helpful for complete combustion. Therefore, it is well-known that exhaust emissions will be reduced.5,12−16 According to the report of Tate et al.,2 kinematic viscosity at higher fuel temperature is lower than that at lower fuel temperature. Therefore, in this study, the temperature of the fuel was 60 °C. 3.2. Injection Quantity. 3.2.1. Characteristics of Injection Quantity According to the Mixing Rate of Canola Biodiesel. Tate et al.2 reported that, as the temperature of biodiesel fuel increases, biodiesel has similar kinematic viscosity characteristics to those of diesel. Furthermore, Yoon et al.17 found that all viscosities of blends, diesel, and biodiesel decreased with increased fuel temperature. Also, as the fuel temperature increased, the difference between the diesel and biodiesel decreased. Therefore, we maintained the fuel temperature at 60 °C for this experiment. As shown in Figure 6, when both injection pressure and injection duration were increased, injection quantity increased linearly. There were small differences compared with diesel. Also, changes in mixing rate did not affect the characteristics of injection quantity due to injection pressure and duration. 3.2.2. Characteristics of Injection Quantity According to the Mixing Rate of Soybean Biodiesel. Figure 7 shows a comparison of injection quantity for different soybean biodiesel blending rates at different injection pressures. As well as significant results for canola, when both injection pressure and duration were increased, injection quantity increased linearly. Also, soybean biodiesel had few differences from diesel fuel. Furthermore, changes of mixing rate do not affect the characteristics of injection quantity due to injection pressure and duration. 3.2.3. Comparison of Canola and Soybean Biodiesels. When comparing the two kinds of fuel, the injection quantity of soybean biodiesel showed a slight tendency to decrease compared to canola biodiesel, but the difference was very small. Therefore, these results indicate that there was no significant difference of total injection quantity between soybean and canola biodiesels. 3.3. Spray Visualization. In this study, we carried out spray visualization experiments according to the fuel type and injection pressure (IP) under the same injection quantity conditions. As seen from Table 8, the fuel properties have a

Figure 5. Schematic of the combustion flame visualization device.

description

diesel

cetane number calorific value (MJ/kg) density@15°C (kg m3) kinematic viscosity@40°C (mm2/s) cloud point (°C) pour point (°C) flash point (°C) carbon (wt %) hydrogen (wt %) oxygen (wt %) sulfur (wt %)

fuels, we blended conventional diesel with mixing rates of 5%, 20%, and 35% by volume. Biodiesel has different properties from those of conventional diesel fuel. Properties of these fuels are listed in Table 8. As shown in Table 8, biodiesel has a lower cetane number and caloric value than diesel. Although the difference in density between the two is small, the difference in kinematic viscosity is very large. If the kinematic viscosity is high, characteristics of atomization will deteriorate. In addition, a high kinematic viscosity leads to poor combustion and wear 5185

dx.doi.org/10.1021/ef400936a | Energy Fuels 2013, 27, 5182−5191

Energy & Fuels

Article

Figure 6. Comparison of injection quantity for different canola biodiesel blending rates at different injection pressures.

visualization data of diesel as reference data to compare with the biodiesel results. 3.3.2. Spray Characteristics of Canola-Blended Fuel. Canola 5%, as shown in Figure 8, had very similar results to diesel. Of course, only 5% of canola biodiesel was blended. However, even canola 20% and 35% showed small differences compared with the results of diesel. From the RMS image and Figure 9, first, spray penetration during the early period was reduced compared to that of diesel. The shorter spray penetration caused by higher kinematic viscosity was observed under 1000 bar conditions. In the case of 5% canola blended fuel, a difference due to the initial injection delay was observed, but the spray penetration became similar to the diesel as time passed. Also, the slope increased sharply. The reason is that although the injection delay due to the kinematic viscosity occurred at the initial injection period, the penetration of the 5% blended fuel became longer than that of diesel due to the relatively lower evaporation characteristics at the end stage of injection. However, in the cases of 20% and 35% canola blended fuels, the spray penetration was slightly irregular due to the increased biodiesel content. Second, nonuniform distribution of the spray was observed in RMS image processing. The 5% canola blended fuel had a similar spray distribution to that of diesel, and only the amount of evaporation decreased at the same time. However, the 20% and 35% canola blended fuels had an irregular spray distribution, and their evaporation characteristics were poorer. Third, when the mixing rate increased, an injection delay due to the influence of kinematic viscosity was observed. Therefore, canola-blended fuel has

large influence on the fuel atomization and combustion characteristics.17 The kinematic viscosity, in particular, expressed as a ratio of the dynamic viscosity and the density was a representative factor.4 Therefore, the characteristics of spray visualization for biodiesel-blended fuels were investigated to determine the influence of kinematic viscosity. In this part, the injection pressure was fixed at 1000 bar for all cases to compare the spray characteristics of each fuel. 3.3.1. Spray Characteristics of Diesel Fuel. As shown in Figure 8, the injection delay of the diesel fuel was shorter than that of biodiesel at the initial injection period. Also, the spray distribution of each hole was very regular. The reasons for those phenomena are that the injector was well developed for pure diesel fuel and the low kinematic viscosity was one of the factors. Furthermore, as seen in Figure 9, the initial spray penetration of diesel was longer than biodiesel fuel. However, the slope of the diesel decreased as the time passed, and it can be confirmed from the RMS image result. The fuel atomization of the diesel is fast paced due to the low kinematic viscosity and high evaporation characteristics. Therefore, as seen from the RMS result, the outskirt distribution of the spray became lower, and then although the initial spray penetration was longer, penetration was shorter as time passed. At lower injection pressures, spray penetration seems to be reduced compared to that at higher injection pressure. Also, at higher ambient pressures, spray penetration seems to be reduced compared to that at lower ambient pressure. Furthermore, the distribution of the spray tended to be very uniform because this injector was developed for diesel fuel. In this paper, we used spray 5186

dx.doi.org/10.1021/ef400936a | Energy Fuels 2013, 27, 5182−5191

Energy & Fuels

Article

Figure 7. Comparison of injection quantity for different soybean biodiesel blending rates at different injection pressures.

Figure 9. Spray tip penetration according to injection pressure.

kinematic viscosity than pure canola biodiesel, soybean-blended fuel and canola-blended fuel had a few differences. First, soybean-blended fuel had slightly higher spray penetration compared to canola-blended fuel. Second, despite the difference in kinematic viscosities, soybean-blended fuel’s injection delay was similar to that of the canola-blended fuel. Tate et al.2 reported that, when the fuel temperature increased, the kinematic viscosity of canola and soybean oil differed by more than 0.4 mm2/s. However, the present study showed that a difference in kinematic viscosity of about 0.4 mm2/s did not affect spray penetration or injection delay because of the high injection pressure. Third, from the RMS image, soybeanblended fuel had a nonuniform distribution.

Figure 8. Characteristics of spray visualization and RMS images at 1000 bar.

lower spray penetration and higher injection delay than diesel and a nonuniform distribution. 3.3.3. Spray Characteristics of Soybean-Blended Fuel. As shown in Figure 8, results of the soybean-blended fuel were not much different from the results of canola-blended fuel, although it had lower penetration and a higher injection delay than diesel. However, although pure soybean biodiesel had a lower 5187

dx.doi.org/10.1021/ef400936a | Energy Fuels 2013, 27, 5182−5191

Energy & Fuels

Article

Figure 10. Combustion visualization for different fuel types.

3.4. Combustion Visualization and Exhaust Emission. As shown in Table 6, we applied the same conditions to

compare the combustion characteristics according to the fuel types. The results of the combustion visualization test are 5188

dx.doi.org/10.1021/ef400936a | Energy Fuels 2013, 27, 5182−5191

Energy & Fuels

Article

Figure 11. Combustion pressure and RoHR for different fuel types.

Figure 12. Exhaust emissions for different fuel types.

arranged according to the crank angle in Figure 10. As seen from the overall pictures, rapid combustion, known as a characteristic of conventional diesel combustion, occurred. Also, canola and soybean blended fuels had higher peak values of both combustion pressure and RoHR (Rate of Heat Release) than diesel fuel, see Figure 11.4 The reason for the higher combustion pressure is that the biodiesel has a greater oxygen content, which increases NOx emissions.13,18−20 In other words, the evaporation characteristics of the biodiesels are poorer than those of diesel. Thus, the rapid temperature rise compared to diesel combustion reflected the high combustion temperature and pressure. In the cases of both 5% canola and soybean blended fuels, the ignition timings were advanced compared to diesel, as seen from Figure 11. It can be judged that the oxygen content in the biodiesel fuel facilitated the combustion and shortened both the ignition delay and combustion period. As a result, it also affected the exhaust emissions such as NOx, PM, CO, and THC. NOx emission of biodiesel blended fuels including oxygen content was a little higher than diesel. PM was lower than that of diesel because of combustion promoted by the oxygen content. In the case of CO emission, it was lower than diesel because of the similar evaporation characteristics to diesel and promoted combustion. THC was also low because the wall impingement decreased because of the shorter spray penetration, and HC was burnt under the high temperature conditions during the combustion period. Also, under the 5% conditions, WS was not found in the results of combustion visualization. However, in the cases of 20% and 35% canola and soybean blended fuels, the combustion period decreased, but the ignition delay increased

due to the poor evaporation characteristics affected by the high kinematic viscosity. Therefore, as shown in Figure 12, it can be observed that NOx increases and PM decreases when the biodiesel content increases. Also, yellow flame related to the PM (particulate matter) occurrence was almost not observed in the flame visualization images. And as shown in Figure 12, the concentration of PM was strikingly lower compared to diesel. It can be said that the oxygen contents in the biodiesel promoted the combustion process. To support this conclusion, the RGB signals of combustion images were analyzed using the Matlab code. The yellow flame signals20 were extracted from the combustion images, and then its area was calculated. Finally, as seen in Figure 13, the distribution rate of the PM region during the combustion period was derived. It can be observed that the result seems to have a similar trend to the exhaust emission

Figure 13. PM distribution rate by RGB analysis. 5189

dx.doi.org/10.1021/ef400936a | Energy Fuels 2013, 27, 5182−5191

Energy & Fuels

Article

result. Therefore, it is carefully said that this approach can be a reasonable method. The value of CO under 5% biodiesel blended conditions was lower, but above this concentration, the results were higher than diesel. The cause of CO emission is due to incomplete combustion or chemical reaction under high temperature conditions.20 In this study, from the results of both spray penetration and THC emission, it can be judged that the higher CO level except 5% biodiesel blended fuel is due to both the worse evaporation characteristics and chemical reaction under the high temperature conditions. It could be reasonable considering the results of NOx emission generated under high temperature conditions. Furthermore, in the cases of both 20% and 35% canola and soybean blended fuels, white smoke (WS) was observed at the start of combustion. This can be the effect of microscopic residual water in the biodiesel-blended fuel or poor evaporation characteristics due to high kinematic viscosity. However, from Figure 12, the value of THC was still lower than that of diesel, so that generated WS at the SOI is burned during the rest of the combustion period.13,18 Other engine tests will be prepared to study this phenomenon in detail. As a result, although 5% biodiesel blended fuel has a slightly higher NOx level than diesel, it showed however the best results in terms of combustion and emission among all the fuels. In the cases of 20% and 35%, it can be judged that a way to improve evaporation characteristics is needed in terms of combustion and emission. In addition, canola biodiesel-blended fuels generally showed a better result than soybean blended fuels. From the results, it can be said that as the mixing rate of soybean biodiesel increases, the results were seriously influenced by fuel properties such as oxygen content and kinematic viscosity compared to canola biodiesel fuel.

blended fuels or poor evaporation characteristics due to the high kinematic viscosity. Yellow flame related to PM generation was calculated from the combustion images using RGB image processing, and the result showed a similar trend to the PM emission characteristics. Therefore, it was demonstrated that this kind of approach could be a useful and reasonable method. Biodiesel blended fuels generally had higher NOx values than diesel because the included oxygen contents and rapid combustion after the longer ignition delay increased the combustion temperature; PM levels were much lower than diesel because of combustion promoted by the oxygen contents in the fuel. THC levels of blended fuels were also lower than diesel because of the decreased wall impingement by the shorter spray penetration length and the promoted combustion. In the case of CO emissions, when the mixing rate of biodiesel increased over 10%, the levels became higher than diesel. It can be expected that the worse evaporation characteristics of the blended fuels and the chemical reaction at the higher combustion temperature increased CO generation.

■

AUTHOR INFORMATION

Corresponding Author

*Tel.: +82-31-400-5251. Fax: +82-31-400-4064. E-mail: hylee@ hanyang.ac.kr. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Industrial Strategic Technology Development Program (No. 10039673, Development of Design Technologies of Core Control Algorithms, and an ECU for clean diesel engines) funded by the Ministry of Knowledge Economy (MKE, Korea).

4. CONCLUSION The effects of different mixing rates of biodiesel-blended fuels on the spray, combustion, and exhaust emission characteristics were analyzed using a single cylinder optical engine, spray visualization system, and exhaust gas analyzer. To understand the spray characteristics of various biodiesel blended fuels, the injection quantity, spray images, RMS images, and spray penetration were analyzed using a spray visualization system and Matlab code. Also, the combustion and the exhaust emission characteristics were analyzed using combustion images, RoHR, and image processing. The following conclusions are derived from the results of this study. The injection quantity of the blended fuels increased linearly according to the injection pressure and duration increasing. Furthermore, the changes of the mixing rate did not significantly affect the characteristics of injection quantity. Therefore, when the fuel temperature was maintained at 60 °C, the effect of kinematic viscosity of biodiesels decreased. On the basis of spray visualization results, both types of biodiesel blended fuels had poor spray distribution and nonlinear spray penetration compared to diesel. This indicates that biodiesel has a higher kinematic viscosity than diesel, resulting in a nonuniform spray distribution and poor evaporation characteristics. From the combustion visualization, in cases of both 20% and 35% canola and soybean biodiesel-blended fuels, white smoke was observed at the start of combustion. It can be expected from the effect of microscopic residual water in the biodiesel-

■

REFERENCES

(1) Senatore, A.; Cardone, M.; Buono, D. Combustion study of a common rail diesel engine optimized to be fueled with biodiesel. Energy Fuels 2008, 22, 1405−1410. (2) Tate, R. E.; Watts, K. C.; Allen, C. A. W.; Wilkie, K. I. The viscosities of three biodiesel fuels at temperatures up to 300°C. Fuel 2006, 85, 1010−1015. (3) Kinast, J. A. Production of Biodiesels from Multiple Feedstocks and Properties of Biodiesels and Biodiesel/Diesel Blends, NREL Subcontractor Final Report 1 in a series of 6, 2003. (4) Lee, C. S.; Park, S. W.; Kwon, S. I. An experimental study on the atomization and combustion characteristics of biodiesel-blended fuels. Energy Fuels 2005, 19, 2201−2208. (5) Qi, D. H.; Chen, H.; Geng, L. M.; Bian, Y. Z. H. Experimental studies on the combustion characteristics and performance of a direct injection engine fueled with biodiesel/diesel blends. Energy Convers. Manage. 2010, 51, 2985−2992. (6) Knothe, G.; Gerpen, J. V.; Krahl, J. The Biodiesel Handbook; AOCS Press: Urbana, IL, 2005. (7) Conceicao, M. M.; Candeia, R. A.; Silva, F. C.; Bezerra, A. F.; Fernandes, V. J., Jr.; Souza, A. G. Thermoanalytical characterization of castor oil biodiesel. Renewable Sustainable Energy Rev. 2007, 11, 964− 975. (8) Shao, J.; Yan, Y.; Greeves, G.; Smith, S. Quantitative characterization of diesel sprays using digital imaging techniques. Meas. Sci. Technol. 2003, 14, 1110−1116. (9) Cronhjort, A.; Wahlin, F. Segmentation algorithm for diesel spray image analysis. Appl. Opt. 2004, 43, 5971−5980.

5190

dx.doi.org/10.1021/ef400936a | Energy Fuels 2013, 27, 5182−5191

Energy & Fuels

Article

(10) Jeong, H. Y.; Lee, K. H. Investigation of the relationship between liquid characteristics and spray mean diameter. 19th Annual Conference on Liquid Atomization and Spray System, ILASS-Americas, Toronto, Canada, 2006. (11) Oh, J. M.; Kim, K. B.; Lee, K. H. Effects of various mixer types on the spatial distribution of a De-NOx reductant. Fuel 2012, 94, 93− 101. (12) Lapuerta, M.; Armas, O.; Rodriguez-Fernandez, J. Effect of biodiesel fuels on diesel engine emissions. Prog. Energy Combust. Sci. 2008, 34, 198−223. (13) Sharp, C. A.; Howell, S. A.; Jobe, J. The effect of biodiesel fuels on transient emissions from modern diesel engines. Part I: regulated emissions and performance. SAE paper 2000-01-1967, 2000. (14) Scholl, K. W.; Sorenson, S. C. Combustion of soybean oil methyl ester in a direct injection diesel engine. SAE paper 930934, 1993. (15) Schumacher, L. G.; Borgelt, S. C.; Fosseen, D.; Goetz, W.; Hires, W. G. Heavy-duty engine exhaust emission tests using methyl ester soybean oil/diesel fuel blends. Bioresour. Technol. 1998, 212, 25− 42. (16) Senthil, K. M.; Ramesh, A.; Nagalingam, B. Use of hydrogen to enhance the performance of a vegetable oil fuelled compression ignition engine. Int. J. Hydrogen Energy 2003, 28, 1143−1154. (17) Yoon, S. H.; Park, S. W.; Lee, C. S. Experimental investigation of the fuel properties of biodiesel and its blends at various temperatures. Energy Fuels 2008, 22, 652−656. (18) Agarwal, A. K. Biofuels applications as fuels for internal combustion engines. Prog. Energy Combust. Sci. 2007, 33, 233−271. (19) Senatore, A.; Cardone, M.; Rocco, V.; Prati, M. V. A comparative analysis of combustion process in D.I. Diesel engine fueled with blends biodiesel and low sulfur diesel fuel. SAE paper 972998, 1997. (20) Heywood, J. B. Internal Combustion Engine Fundamentals; McGraw-Hill: New York, 1998.

5191

dx.doi.org/10.1021/ef400936a | Energy Fuels 2013, 27, 5182−5191