Numerical Study on Combustion and Emission Characteristics of


Numerical Study on Combustion and Emission Characteristics of...

3 downloads 113 Views 5MB Size

Energy Fuels 2010, 24, 916–927 Published on Web 12/07/2009

: DOI:10.1021/ef901092h

Numerical Study on Combustion and Emission Characteristics of Homogeneous Charge Compression Ignition Engines Fueled with Biodiesel Sukkee Um and Sung Wook Park* Department of Mechanical Engineering, Hanyang University, 17 Haengdang-dong, Seongdong-gu, Seoul 133-791, Korea Received September 25, 2009. Revised Manuscript Received November 16, 2009

This paper describes a numerical study on the effect of the mixing ratio of biodiesel on combustion and emission characteristics of homogeneous charge compression ignition engines. The KIVA code coupled with Chemkin chemistry solver was used to simulated combustion and emission formation processes. A modified reduced methyl butanoate mechanism was used after combining with a reduced n-heptane mechanism to model ignition and combustion of biodiesel. The mixing ratio of biodiesel was varied from 0% (conventional diesel) to 100% (pure biodiesel). Operating ranges covered up to a 2.0 equivalence ratio. The fueling rate was fixed at 18 mg/cyc, which corresponds to a medium engine load. Results of the present study showed that the ignition delay of biodiesel is shorter than that of conventional diesel fuel because of the higher cetane number of biodiesel. Little differences of emission maps for CO and NOx emissions are found between biodiesel and conventional diesel. However, higher concentration regions for hydrocarbon (HC) and soot maps on the peak cycle temperature/equivalence ratio are reduced significantly using biodiesel instead of conventional diesel fuel.

sprays.4,7,8 Therefore, combustion and emission characteristics of biodiesel fuel are significantly different from those of conventional diesel fuel. To analyze engine performance and emissions of compression ignition engines fueled with biodiesel fuel, studies have been conducted actively in recent decades. With regard to engine performances, most of the studies showed that losses of torque and power ranged between 3 and 10%. Kanplan et al.9 revealed that the loss of torque is closer to 5% at low speed but is around 10% at high speed at full load for a compression ignition engine fueled with biodiesel derived from sunflower oil compared to conventional diesel fuel. Lin et al.10 showed similar results in engine performance. In their results, the loss of power caused by biodiesel was 3.5% with pure biodiesel and 1% with the 20% blend at full load. On the other hand, some studies showed similar or a slight increase in engine power using biodiesel instead of conventional diesel fuel. Lee et al.4 showed that engine power was kept similar or slightly increased by providing biodiesel blendes up to 40%. According to their analysis, biodiesel has a higher cetane number than that of conventional diesel fuel. Hence, ignition timing is advanced (i.e., approaches the top dead center for their experimental conditions), which makes similar engine power, despite the lower heating value of biodiesel. Altiparmak et al.11 also reported a 6.1% increase in the maximum torque with mixed fuel of 70% tall oil and 30% conventional diesel, with respect to that measured with biodiesel. Along with works on engine performance, the effect of biodiesel on emissions has also been studied. Most of the previous studies reported that NOx increases from the mixing

1. Introduction Biodiesel is an attractive alternative fuel because it is oxygenated and is a renewable bio-based fuel that is derived from soybean and rapeseed. Hence, biodiesel has the potential to reduce particulate emissions and hydrocarbons (HCs) in compression ignition engines.1,2 Biodiesel is a multi-component fuel that consists of monoalkyl ester of long-chain fatty acids derived from vegetable oils and animal fats. The soy- and rapeseed-derived biodiesels are mainly composed of five saturated and unsaturated methyl esters, such as methyl palmitate, methyl stearate, methyl oleate, methyl linoleate, and methyl linolenate.3 These methyl esters contain oxygen; hence, soy- and rapeseed-derived biodiesels contain around 11 vol % oxygen. The oxygen content in biodiesel changes combustion phasing and partially contributes to 9% less heating of biodiesel compared to conventional diesel fuel.4 It is also known that physical properties of biodiesel (e.g., surface tension and kinematic viscosity) are much different from those of conventional diesel fuel.5,6 It was also reported that differences in physical properties of biodiesel make differences in atomization characteristics of fuel *To whom correspondence should be addressed: Department of Mechanical Engineering, Hanyang University, 17 Haengdang-dong, Seongdong-gu, Seoul 133-791, Korea. Telephone: þ82-2-2220-0430. Fax: þ82-2-2220-4588. E-mail: [email protected]. (1) Agarwal, A. K. Prog. Energy Combust. Sci. 2007, 33, 233–271. (2) Lapuerta, M.; Armas, O.; Rodriguez-Fernandez, J. Prog. Energy Combust. Sci. 2008, 34, 198–233. (3) Herbinet, O.; Pitz, W. J.; Westbrook, C. K. Combust. Flame 2008, 154, 507–528. (4) Lee, C. S.; Park, S. W.; Kwon, S. I. Energy Fuels 2005, 19, 2201– 2208. (5) Demirbas, A. Fuel 2008, 87, 1743–1748. (6) Tang, H.; Salley, S. O.; Simon Ng, K. Y. Fuel 2008, 87, 3006–3017. (7) Im, S. Y.; Song, Y. S.; Ryu, J. I. Int. J. Automot. Technol. 2008, 9, 249–256. (8) Suh, H. K.; Park, S. H.; Lee, C. S. Int. J. Automot. Technol. 2008, 9, 217–224. r 2009 American Chemical Society

(9) Kanplan, C.; Arslan, R.; Surmen, A. Energy Sources, Part A 2006, 28, 751–755. (10) Lin, Y.-F.; Wu, Y.-P. G.; Chang, C.-T. Fuel 2007, 86, 1772–1780. (11) Altiparmak, D.; Deskin, A.; Koca, A.; Guru, M. Bioresour. Technol. 2007, 98, 241–246.

916

pubs.acs.org/EF

Energy Fuels 2010, 24, 916–927

: DOI:10.1021/ef901092h

Um and Park

4

of biodiesel. Lee et al. found around a 20% increase in NOx emissions with a 40% biodiesel blend with respect to conventional diesel fuel. They reported that ignition timing was advanced with the increase in the mixing ratio of biodiesel, which lead to a higher in-cylinder peak temperature. Marshall et al.12 performed an experimental study under transient conditions using a Cummins LE10E engine fueled with diesel fuel and 20% biodiesel blends. They also reported an increase in NOx emissions of 3.7% with the 20% blend. Choi and Reitz13 analyzed the reason for the increase in NOx for biodiesel blends numerically. They obtained a reduced ignition delay and higher extension of the high-temperature areas using biodiesel fuels. Although most previous work showed a slight increase in NOx emissions when using biodiesel fuel, some studies showing different effects have been found. An experiment was performed by Staat and Gateau14 using a 6-cylinder engine for an ECE R49 test cycle and an urban transient cycle named AQA F21 established by the French Agency of Air Quality. In their experimental results, NOx is increased by 9.5% in the ECE R49 mode while decreased by 6.5% in the transient urban cycle. In terms of soot, HC, and carbon monoxide (CO) emissions, most work observed decreases in these emissions.4,15,16 It is known that soot, HC, and CO emissions are formed mainly at high equivalent ratio regions (i.e., rich regions).17 Considering that biodiesel contains 11% oxygen, it is believed that biodiesel decreases soot, HC, and CO emissions because it is oxygenated. Despite extensive work on combustion and emission characteristics of biodiesel, some uncertainties still remain. Especially, it is not fully understood whether chemical characteristics (e.g., higher cetane number and oxygen in biodiesel) or physical properties (e.g., higher density and viscosity) are dominant on different combustion and emission characteristics of biodiesel. To analyze the effect of chemical characteristics of biodiesel on combustion and emission characteristics, a numerical study on homogeneous charge compression ignition (HCCI) engines fueled with biodiesel and its blends is useful because there is no consideration of the physical properties of fuel. To simulate HCCI engines, a proper surrogate and corresponding oxidation mechanism are required. As a surrogate for biodiesel, the surrogate should have an oxygen content similar to biodiesel (i.e., around 11%). Herbinet et al.3 developed a detailed chemical kinetic oxidation mechanism for methyl decanoate (C11H22O2) and found that the ignition delay associated with biodiesel predicted reasonably using the methyl decanoate oxidation mechanism. However, because methyl decanoate has 12 carbons, it is not proper with respect to the computational time cost for three-dimensional engine simulations. Brakora et al.18 suggested a reduced mechanism for biodiesel-fueled engine simulations using methyl butanoate (C5H10O2). To maintain the oxygen amount at 11% by mass, they assumed that 1 mol of biodiesel consisted of 1 mol of

Figure 1. Two-dimensional computational mesh at 40° before top dead center (BTDC) crank angle. Table 1. Engine Specifications and Calculation Conditions test fuels

D100 (conventional diesel) BD10 (10% rice oil þ 90% conventional diesel) BD20 (20% rice oil þ 80% conventional diesel) BD50 (50% rice oil þ 50% conventional diesel) BD100 (pure biodiesel) bore  stroke (mm) 75.0  84.5 compression ratio 17.8 fuel amount (mg/cyc) 18 373.3 displacement (cm3) engine speed (rpm) 1500 equivalence ratio 0.2-2.0 by 0.2 intake 300-460 by 20 temperature (K) boost 0.1-0.3 by 0.02 pressure (MPa) IVC -128° ATDC EVO 172° ATDC swirl ratio at IVC 1.63 wall temperatures (K) 553 (piston) 523 (head) 433 (cylinder)

methyl butanoate and 2 mol of n-heptane. In addition, the proposed surrogate of biodiesel (1 mol of methyl butanoate þ 2 mol of n-heptane) has a molecular weight of 302 g/mol, which is similar to that of soy-derived biodiesel (292 g/mol). However, their calculation conditions were limited to pure biodiesel and did not consider biodiesel blends. The objective of the present study is to investigate combustion and emission characteristics of HCCI engines fueled with biodiesel blends using the KIVA code coupled with Chemkin chemistry solver.19 To simulate ignition and combustion processes, an oxidation mechanism of methyl butanoate combined with a n-heptane mechanism was used, as suggested by Brakora et al.18 However, some reaction constants have been revised as validated in the previous work.20 2. Model Formulation A modified version of the KIVA-3 V code21 was used to simulate the combustion and emission characteristics of the HCCI engines fueled with biodiesel blends using two-dimensional computational mesh. The mechanism for biodiesel oxidation suggested by Brakora et al.18 was used after adjusting reaction constants for some low-temperature reactions of methyl butanoate to produce a reasonable ignition delay, as suggested by Um and Park.20

(12) Marshall, W.; Schumacher, L. G.; Howell, S. SAE Tech. Pap. 952363, 1995. (13) Choi, C. Y.; Reitz, R. D. J. Eng. Gas Turbines Power 1999, 131, 31–37. (14) Staat, F.; Gateau, P. SAE Tech. Pap. 950053, 1995. (15) Alfuso, S.; Auriemma, M.; Police, G.; Prati, M. V. SAE Tech. Pap. 932801, 1993. (16) Choi, C. Y.; Bower, G. R.; Reitz, R. D. SAE Tech. Pap. 970218, 1997. (17) Park, S. W.; Reitz, R. D. Combust. Sci. Technol. 2007, 179, 2279– 2307. (18) Brakora, J. L.; Ra, Y.; Reitz, R. D., McFarlane, J.; Daw, C. S. SAE Tech. Pap. 2008-01-1378, 2008.

(19) Kee, R. J.; Rupley, F. M.; Miller, J. A. Sandia Report SAND 898009, 1989. (20) Um, S; Park, S. W. Fuel 2010, DOI: 10.1016/j.fuel.2009.10.026. (21) Amsden, A. A. Los Alamos National Laboratory Report LAUR-99-915, 1999.

917

Energy Fuels 2010, 24, 916–927

: DOI:10.1021/ef901092h

Um and Park

Figure 2. Effects of the mixing ratio of biodiesel on combustion characteristics of HCCI engines (Tini = 360 K, Pboost = 0.12 MPa, and φ = 1.0).

2.2. Emission and Other Numerical Models. To predict CO and HC emissions, no additional emission model is needed because they are species represented in the mechanism. The HC emission is defined as the total mass of species that contain both hydrogen and carbon regardless of the combination with oxygen in the present study. To simulate soot emissions, a two-step phenomenological model that considers both soot formation and oxidation was used. Soot formation was calculated using a modified Hiroyasu model, which considers acetylene (C2H2) as a soot precursor.24 On the other hand, the Nagle-Strickland-Constable model was employed to simulate soot oxidation.25 The rate of soot mass change M_ s within a computational cell is calculated from the soot formation rate M_ sf and soot oxidation rate M_ so.24 dM_ s dM_ sf dM_ so ¼ ð1Þ dt dt dt

2.1. Fuel Oxidation Mechanism of Biodiesel Blends. To model the fuel ignition and combustion process, the KIVA code was coupled with Chemkin chemistry solver.19 Hence, a proper fuel oxidation mechanism is required for simulation of HCCI engines fueled with biodiesel blends. In the present study, a biodiesel oxidation mechanism developed by Brakora et al.18 was used. They developed a reduced mechanism for biodiesel oxidation, which consists of 156 reactions among 54 species. The mechanism consists of oxidation reactions for methyl butanoate and n-heptane. Key reactions for methyl butanoate were chosen from the reaction mechanism developed by Lawrence Livermore National Laboratory (LLNL)22 and then combined with those for n-heptane suggested by Patel et al.23 However, it is known that the methyl butanoate mechanism has discrepancy in predicting ignition delay of biodiesel.3 Thus, in the present study, a biodiesel mechanism was used to simulate the combustion process of the HCCI engine after modifying some reaction constants, as suggested by Um and Park.20 The biodiesel reaction mechanism of the present study is shown in the Appendix. Biodiesel contains around 11% oxygen. Thus, another issue in modeling biodiesel combustion is to keep the oxygen content of the biodiesel surrogate at 11%. To maintain the oxygen amount at 11% by mass, it is assumed that 1 mol of biodiesel consisted of 1 mol of methyl butanoate and 2 mol of n-heptane, as suggested by Brakora et al.18 In addition, the proposed surrogate of biodiesel (1 mol of methyl butanoate þ 2 mol of n-heptane) has a molecular weight of 302 g/mol, which is similar to that of soy-derived biodiesel (292 g/mol). Normal heptane (C7H16) was used as a surrogate for conventional diesel fuel.

where Asf is 200, Esf is 12 500 cal/mol, MC2H2 is the mass of C2H2, MWC is the carbon molecular weight, Fs is the soot density (2.0 g/cm3), Ds is the soot diameter (2.5  10-6 cm), and Rtotal is the net reaction rate.

(22) Fisher, E. M.; Pitz, W. J.; Curran, H. J.; Westbrook, C. K. Proc. Combust. Inst. 2000, 28, 1579–1586. (23) Patel, A.; Kong, S. C.; Reitz, R. D. SAE Tech. Pap. 2004-010558, 2004.

(24) Kong, S.-C.; Sun, Y.; Reitz, R. D. J. Eng. Gas Turbines Power 2007, 129, 245–251. (25) Han, Z.; Uludogan, A.; Hampson, G. J.; Reitz, R. D. SAE Tech. Pap. 960633, 1996.

An Arrhenius expression is used for the soot formation, given as

918

  dM_ sf Esf ¼ Asf MC2 H2 Pn exp dt RT

ð2-1Þ

dM_ so 6MWC _ M s Rtotal ¼ dt Fs Ds

ð2-2Þ

Energy Fuels 2010, 24, 916–927

: DOI:10.1021/ef901092h

Um and Park

Figure 3. Effect of the mixing ratio of biodiesel on emission characteristics of HCCI engines (Tini = 360 K, Pboost = 0.12 MPa, and φ = 1.0).

Figure 4. Effect of the equivalence ratio on combustion characteristics (Tini = 340 K and Pboost = 0.14 MPa). CA50 indicates the crank angle at which accumulated heat release reaches 50% of the total heat release.

species (i.e., N, NO, NO2, and N2O) to describe the formation of nitric oxides, as shown in the Appendix. In the present simulations, NOx was defined as the sum of NO and NO2. The reduced

A NOx mechanism that is reduced from the Gas Research Institute (GRI) NO mechanism24 was used for the present study. The reduced NOx mechanism consists of 12 reactions and 4 919

Energy Fuels 2010, 24, 916–927

: DOI:10.1021/ef901092h

Um and Park

Figure 5. Effect of the equivalence ratio on emission characteristics (Tini = 340 K and Pboost = 0.14 MPa).

volume at IVC are fixed for each calculation, the oxygen mole fraction can be calculated to make a given equivalence ratio. Then, it is assumed that the remaining portion of the mixture is N2 and CO2, which are the main species of the exhaust gas recirculation (EGR). In the present paper, indicated mean effective pressure (IMEP) was calculated using work performed from -180° after top dead center (ATDC) to þ180° ATDC. In-cylinder pressures before IVC and after EVO are assumed to be identical to those at IVC and EVO, respectively.

GRI NO mechanism was combined with the biodiesel reaction mechanism and solved by Chemkin chemistry solver.19 2.3. Calculation Conditions. A two-dimensional computational mesh of the present study was generated on the basis of a single-cylinder diesel engine with a 75 mm bore size and 84.5 mm stroke, as shown in Figure 1. Calculations were performed from intake valve closure (IVC) to exhaust valve opening (EVO) to save calculation time. To simulate HCCI engines, it is assumed that the fuel is completely evaporated and mixed with air homogeneously at IVC as performed in the previous studies.17,26 In the present study, HCCI engine emissions fueled with biodiesel and its blends were analyzed using maps on equivalence ratio/peak cycle temperature. Thus, numerous calculations were performed to complete equivalence ratio/peak cycle temperature maps for engine performance (e.g., total heat release) and emissions, covering a wide range of operating conditions. The equivalence ratio was varied from 0.2 to 2.0, and the intake temperature ranged from 300 to 460 K. In addition, boost pressures were set to cover 0.1-0.3 MPa. Thus, 990 calculations using a two-dimensional mesh for HCCI combustion were conducted for each biodiesel blend. The mixing ratio of biodiesel was changed from 0% (conventional diesel) to 100% (pure biodiesel). However, the fueling rate was fixed at 18 mg per cycle because biodiesel is applied to conventional diesel engines without modification in the engine electronic control unit (ECU) or structure generally. The fueling rate of the present study corresponds to the medium load of the test engine. Detailed engine operating conditions are listed in Table 1. To calculate HCCI combustion, the initial mole fraction of the mixture at IVC is required for each calculation. Because fuel mass, intake temperature, boost pressure, and in-cylinder

3. Results and Discussion Combustion and emission characteristics of HCCI engines fueled with D100 (conventional diesel), BD10, BD20, BD50, and BD100 (pure biodiesel) are analyzed for 360 K intake temperature, 0.12 MPa boost pressure, and at 1.0 equivalence ratio cases. Then, correlations between peak cycle temperature and emissions, such as CO, HC, and soot emissions, are found. Finally, emissions are plotted on peak cycle temperature/equivalence ratio maps for various biodiesel blends. On the basis of emission maps, effects of the mixing ratio on CO, HC, and soot emissions are analyzed. 3.1. Effect of the Mixing Ratio of Biodiesel on Engine Performance and Emissions of a HCCI Engine. Figure 2 shows the effects of the mixing ratio of biodiesel on combustion characteristics of HCCI engines for 360 K intake temperature, 0.12 MPa boost pressure, and 1.0 equivalence ratio. In Figure 2, mixing ratios of biodiesel are varied from 0% (D100) to 100% (BD100). In pressure histories of Figure 2a, ignition timing is advanced as the mixing ratio of biodiesel is increased (also can be seen in Figure 2b). It is known that biodiesel has a higher cetane number than that of

(26) Park, S. W. Energy Fuels 2009, 23, 3909–3918.

920

Energy Fuels 2010, 24, 916–927

: DOI:10.1021/ef901092h

Um and Park

Figure 6. Correlations between the peak cycle temperature and emissions of HCCI engines fueled with BD100.

Figure 7. Correlations between the peak cycle temperature and emissions of HCCI engines fueled with D100.

conventional diesel.27,28 Lee et al.4 mentioned that the oxygen content and higher cetane number of biodiesel reduces advances in ignition timing of biodiesel. However, considering that all cases in Figure 2 have the same equivalence ratio, it is believed that a shorter ignition delay of biodiesel mainly comes from a higher cetane number. An issue of using biodiesel in a conventional compression ignition engine is that the heating value of biodiesel is lower than that of conventional diesel fuel by 10-15%.18,27 Thus, the effect of the mixing ratio of biodiesel on cumulative heat releases is of interest. Cumulative heat releases are shown in Figure 2c. It can be seen that an increasing mixing ratio of biodiesel decreases the total heat release. When the total heat

release of BD100 is compared to that of D100, the total heat release of BD100 is lower by 12.9%, which agrees with previous studies on the heating value of biodiesel.18,27 Figure 2d shows the mean in-cylinder temperature as a function of the crank angle for various mixing ratios. The incylinder temperature is a significant result for NOx emissions because NOx emissions are greatly dependent upon the incylinder temperature. In Figure 2d, the peak in-cylinder temperature is decreased as the mixing ratio of biodisel increases because of the lower heating value of biodiesel. The effects of the mixing ratio of biodiesel on emission characteristics for CO, HC, and NOx are shown in Figure 3 at the same operating conditions as those of Figure 2. In CO and HC emissions of panels a and b of Figure 3, respectively, CO and HC emissions are both increased as the mixing ratio increases. On the other hand, NOx emissions are decreased. Although the equivalence ratio is kept at the stoichiometric

(27) Zhang, X.; Gao, G.; Li, L.; Wu, Z.; Hu, Z.; Deng, J. SAE Tech. Pap. 2008-01-1832, 2008. (28) Lin, Y. C.; Lee, W. J.; Wu, T. S.; Wang, C. T. Fuel 2006, 85, 2516– 2513.

921

Energy Fuels 2010, 24, 916–927

: DOI:10.1021/ef901092h

Um and Park

Figure 9. Emission map of NOx in units of g kW-1 h-1 for HCCI engines fueled with BD100 and D100.

Figure 8. Emission map of CO in units of g kW-1 h-1 for HCCI engines fueled with BD100 and D100.

in the equivalence ratio. For rich combustions, a significant amount of energy is lost because of CO and HC. Thus, the total heat release decreases, as shown in Figure 4c, and IMEP is reduced. In addition, as the mixing ratio is increased, IMEP is deteriorated for most equivalence ratios in Figure 4a. Seeing total heat releases in Figure 4c, it is believed that worse IMEP is caused by less heating of biodiesel. Furthermore, the required initial oxygen mole fraction to keep a constant equivalence ratio is decreased with an increased mixing ratio of biodiesel because 11% of biodiesel is oxygen content. 3.2. Correlations between the Peak Cycle Temperature and Emissions. Figure 5 shows the effect of the equivalence ratio on emissions, such as CO, HC, NOx, and soot, for 340 K initial temperature and 0.14 MPa boost pressure. In CO emission results of Figure 5a, CO is increased for biodiesel blends with higher mixing ratios generally. Although higher CO emission is observed for D100 at a 2.0 equivalence ratio, it is believed that corresponding IMEP is significantly low compared to other operating conditions (see Figure 4a). However, the mixing ratio of biodiesel has little influence on HC emissions for most operating conditions in Figure 5b. In NOx emissions at a 0.6 equivalence ratio of Figure 5c, it can be seen that BD10 shows a higher NOx emission than D100. However, decreased NOx emissions are observed as the mixing ratio of biodiesel increases from BD10 to BD100. The peak in-cylinder temperature of a biodiesel-fueled engine is mainly dependent upon two factors: ignition timing and total heat releases. Advanced ignition timing of biodiesel

value in Figure 3, Park and Reitz17 found that CO and HC emissions are mainly formed at an equivalence ratio over 1.2 and high NOx concentrations are found near a 0.7 equivalence ratio. Thus, it is hard to see that the results of Figure 3 represent the effect of the mixing ratio of biodiesel on emissions. Results of Figures 2 and 3 show the effect of the mixing ratio of combustion and emission characteristics for specific operating conditions, especially for a fixed equivalence ratio. Hence, the effect of the equivalence ratio on engine performances are analyzed for 340 K initial pressure and 0.14 MPa boost pressure, as shown in Figure 4. In Figure 4a, IMEP is increased from 0.6 to the stoichiometric value and is decreased over the stoichiometric value as the equivalence ratio increases for all tested fuels. At low equivalence ratios, such as 0.6 and 0.8, it is believed that IMEP is deteriorated as the ignition timing becomes too advanced. In results of ignition timing (i.e., CA50) of Figure 4b, it can be seen that combustion starts BTDC for equivalence ratios of 0.6 and 0.8. In Figure 4b, CA50 indicates a crank angle when the accumulated heat release reaches 50% of the total heat release, which shows the start of combustion.29 For rich combustions (i.e., equivalence ratios over the stoichiometric value), IMEP is deteriorated with an increase (29) Genzale, C. L.; Kong, S.-C.; Reitz, R. D. J. Eng. Gas Turbines Power 2008, 130, No. 052806.

922

Energy Fuels 2010, 24, 916–927

: DOI:10.1021/ef901092h

Um and Park

Figure 10. Emission map of HC in units of g kW-1 h-1 for HCCI engines fueled with BD100 and D100.

Figure 11. Emission map of soot in units of g kW-1 h-1 for HCCI engines fueled with BD100 and D100.

increases the in-cylinder temperature, which results in increased NOx emissions. On the other hand, a decreased total heat release reduced NOx emissions, as shown in Figure 2. Therefore, it is believed that the effect of the mixing ratio of biodiesel on NOx emissions is governed by which factor is more dominant among the two factors (i.e., advanced ignition timing and less total heat release of biodiesel). Soot emission results of Figure 5d show that soot decreases by mixing biodiesel to conventional diesel fuel. Strong correlations between the peak cycle temperature and emissions, such as CO, HC, NOx, and soot, are found for n-heptane by Park and Reitz.17 Figure 6 shows that the correlations can also be found for biodiesel (BD100). In Figure 6, correlations between the peak cycle temperature and emissions, such as CO, HC, NOx, and soot, are shown for equivalence ratios of 0.6, 1.0, and 1.4. Below stoichiometric cases of Figure 6a, CO and HC are formed below the 1400 K peak cycle temperature, which corresponds to the CO and HC oxidation limit found by Park and Reitz17 and Ekoto et al.30 On the other hand, NOx is increased with higher peak cycle temperatures (i.e., over 1800 K), which is contrary to the CO and HC emission trend. Soot emissions are observed from 1000 to 1600 K peak cycle temperatures, although the amount of soot is minimal.

For stoichiometric cases of Figure 6b, emission trends are similar to those of the 0.6 equivalence ratio in Figure 6a and soot emissions are still minimal. At the 1.4 equivalence ratio of Figure 6c, significant amounts of CO and HC emissions are observed for most peak cycle temperatures. In addition, a higher concentration of CO is found at a higher peak cycle temperature near 2000 K. On the other hand, the HC distribution has a higher concentration at a lower peak cycle temperature. As found by Park and Reitz,17 CO and HC are formed at similar operating conditions and two factors contribute to incomplete oxidations of CO and HC: too low of a temperature and too rich of a mixture (i.e., too high pf an equivalence ratio). Thus, it is believed that higher concentrations of HC emissions are observed at a low peak cycle temperature because fuel itself is treated as HC for HCCI engine calculations. Figure 7 shows correlations between the peak cycle temperature and emissions for conventional diesel (D100) combustion. For 0.6 and 1.0 equivalence ratios, emission trends are similar to those of BD100 of Figure 5. However, in the result of soot emissions for the 1.4 equivalence ratio, more soot emissions are observed for D100 compared to BD100. In addition, the soot formation range of the peak cycle temperature for D100 is around 1150-2100 K, which is wider than that of BD100 (1200-1900 K). 3.3. Emission Maps. To compare emission characteristics of biodiesel, emissions are plotted on peak cycle

(30) Ekoto, I. W.; Colban, W. F.; Miles, P. C.; Park, S.; Foster, D. E.; Reitz, R. D. SAE Tech. Pap. 2009-01-1446, 2009.

923

Energy Fuels 2010, 24, 916–927

: DOI:10.1021/ef901092h

Um and Park

temperature/equivalence ratio maps and compared to those of conventional diesel. Figure 8 shows the CO emission map on the peak cycle temperature/equivalence ratio map for a HCCI engine fueled with BD100 and D100. In both maps for BD100 and D100, CO emission is increased dramatically when the equivalence ratio becomes higher than the stoichiometric value. For the regions over the stoichiometric value, there is not enough oxygen to react with CO. Thus, CO persists until EVO. It is also found that there are no significant differences between CO emissions of B100 and D100. Therefore, it can be said that mixing biodiesel has little influence on CO emissions chemically. However, in real engine operating conditions, it is know that the local equivalence ratio of biodiesel is lower than that of commercial diesel, which decreases CO emissions.20 Emission maps of NOx for BD100 and D100 are shown in Figure 9. In the figure, high concentrations of NOx for both BD100 and D100 are found for low equivalence ratios (i.e., below the stoichiometric value) and high temperature regions. In maps for HC emissions in Figure 10, below the stoichiometric value, BD100 has a lower peak cycle temperature of the HC oxidation limit (i.e., 1650 K) compared to that of D100 (i.e., 1750 K), which shows that BD100 suppresses HC emissions. Because comparisons are made for the same equivalence ratio, it can be said that biodiesel is capable of reducing HC emissions, even not considering the low equivalence ratio compared to conventional diesel. Figure 11 shows soot emission maps for BD100 and D100. In this figure, it is observed that soot formation regions are reduced significantly for BD100 compared to those of D100. Equivalence ratios for high soot concentrations are similar for both BD100 and D100. However, peak cycle temperatures for soot formations differ significantly from each other.

4. Conclusions In this study, parametric studies for combustion and emission characteristics of HCCI engines fueled with biodiesel were performed using the KIVA code coupled with Chemkin chemistry solver. Mixing ratios of biodiesel were varied from 0% (conventional diesel) to 100% (pure biodiesel). Effects of the mixing ratio on combustion and emissions were investigated for specific operation conditions, and then calculation conditions were expanded to complete emission maps on peak cycle temperature and equivalence ratio plots. The conclusions of the present study can be summarized as follows: (1) As the mixing ratio of biodiesel increases, ignition timing is advanced for the same equivalence ratio. Thus, it is believed that a higher cetane number of biodiesel has significant effects on a shorter ignition delay. (2) Mixing biodiesel deteriorates IMEP mainly because of less heating of biodiesel. In addition, the required initial mole fraction at IVC is decreased by mixing biodiesel because biodiesel is oxygenated. (3) At the 1.4 equivalence ratio, more soot emissions are observed for BD100 compared to D100. Higher soot emissions are observed from 1200 to 1900 K for BD100. However, for conventional diesel, the higher soot concentration region is expanded to 1150-2100 K. (4) Emission maps for CO and NOx show little difference between biodiesel and conventional diesel. Thus, it is believed that CO is reduced mainly because of the low local equivalence ratio and NOx is increased because of the shorter ignition delay for compression ignition engines fueled with biodiesel instead of conventional diesel fuel. Acknowledgment. This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science, and Technology (2009-0072881).

Table A1. Bidiesel Reaction Mechanism k = ATb exp(-E/RT) reactions considered

A (mol cm s K)

1. MB2J þ H = MB 2. MB þ C2H3 = C2H4 þ MB2J 3. MB þ CH3 = CH4 þ MB2J 4. MB þ CH3O2 = CH3O2H þ MB2J 5. MB þ H = H2 þ MB2J 6. MB þ HO2 = H2O2 þ MB2J 7. MB þ O = OH þ MB2J 8. MB þ O2 = HO2 þ MB2J 9. MB þ OH = H2O þ MB2J 10. MB þ MB2OO = MB2OOH þ MB2J 11. MP2D þ CH3 = MB2J 12. MB2J þ O2 = MB2OO 13. MB2OO þ MB2J = MB2O þ MB2O 14. MB2OO þ CH3 = CH3O þ MB2O 15. CH3O2 þ MB2J = CH3O þ MB2O 16. HO2 þ MB2J = OH þ MB2O 17. MB2OO = MB2OOH4J 18. MB2OO þ HO2 = MB2OOH þ O2 19. MB2OO þ H2O2 f MB2OOH þ HO2 20. MB2OOH þ HO2 f MB2OO þ H2O2 21. MB2OO þ CH3O2 f MB2O þ CH3O þ O2 22. MB2OO þ MB2OO f O2 þ MB2O þ MB2O 23. MB2OOH = MB2O þ OH 24. ME2*O þ C2H5 = MB2O 25. C2H5CHO þ CH3OCO = MB2O 26. MB2OOH4J f ME2*O þ C2H4 þ OH 27. MB2OOH4J þ O2 = MB2OOH4OO

1.00  10 4.00  1011 2.00  1011 4.00  1012 2.52  1014 5.62  1014 2.20  1018 4.00  1016 1.15  1011 2.16  1012 1.00  1011 1.41  1015 7.00  1012 7.00  1012 7.00  1012 7.00  1017 2.62  1012 1.75  1010 2.40  1012 2.40  1012 1.40  1016 1.40  1016 5.95  1015 1.50  1011 1.50  1011 5.92  1019 4.52  1012 14

924

b

E (cal/mol)

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -1.6 -1.6 0.0 0.0 0.0 -1.9 0.0

0.0 14300.0 7900.0 14000.0 7300.0 14400.0 3280.0 41300.0 63.0 14400.0 7600.0 0.0 -1000.0 -1000.0 -1000.0 -1000.0 22066.9 -3275.0 10000.0 10000.0 1860.0 1860.0 42540.0 11900.0 11900.0 30290.0 0.0

Energy Fuels 2010, 24, 916–927

: DOI:10.1021/ef901092h

Um and Park Table A1. Continued k = ATb exp(-E/RT)

reactions considered

A (mol cm s K)

28. MB2OOH4OO = MB4OOH2*O þ OH 29. MB4OOH2*O f CH2O þ MP3J2*O þ OH 30. CO þ CH3O = CH3OCO 31. CO2 þ CH3 = CH3OCO 32. HCO þ CH3OCO = ME2*O 33. ME2*O þ H = ME2J*O þ H2 34. ME2*O þ OH = ME2J*O þ H2O 35. ME2*O þ O = ME2J*O þ OH 36. ME2*O þ CH3 = ME2J*O þ CH4 37. ME2*O þ HO2 = ME2J*O þ H2O2 38. CH3OCO þ CO = ME2J*O 39. MP2D þ CH3 f C2H3CO þ CH2O þ CH4 40. MP2D þ H f C2H3CO þ CH2O þ H2 41. MP2D þ O f C2H3CO þ CH2O þ OH 42. MP2D þ OH f C2H3CO þ CH2O þ H2O 43. MP2D þ HO2 f C2H3CO þ CH2O þ H2O2 44. MP2D þ O = CH3OCO þ CH2CHO 45. C2H3 þ CH3OCO = MP2D 46. CH2CO þ CH3OCO = MP3J2*O 47. C2H2 þ OH = CH2CO þ H 48. CH2CO þ H = CH3 þ CO 49. CH2CO þ O = CH2 þ CO2 50. CH2CO(þM) = CH2 þ CO(þM) low pressure limit: 0.36000  1016, 0.00000, and 0.59270  105 51. C3H6 þ O = CH2CO þ CH3 þ H 52. CH2CHO = CH2CO þ H 53. C2H4 þ O = CH2CHO þ H 54. C2H3 þ O2 = CH2CHO þ O 55. CH2CHO þ O2 = CH2O þ CO þ OH 56. C2H5CHO = C2H5 þ HCO 57. CH3O2 þ M = CH3 þ O2 þ M 58. CH3O2H = CH3O þ OH 59. CH3O2 þ CH2O = CH3O2H þ HCO 60. C2H4 þ CH3O2 = C2H3 þ CH3O2H 61. CH4 þ CH3O2 = CH3 þ CH3O2H 62. CH3O2 þ CH3 = CH3O þ CH3O 63. CH3O2 þ HO2 = CH3O2H þ O2 64. CH3O2 þ CH3O2 = O2 þ CH3O þ CH3O 65. C2H3CO = C2H3 þ CO 66. NC7H16 þ H = C7H15-2 þ H2 67. NC7H16 þ OH = C7H15-2 þ H2O 68. NC7H16 þ HO2 = C7H15-2 þ H2O2 69. NC7H16 þ O2 = C7H15-2 þ HO2 70. C7H15-2 þ O2 = C7H15O2 71. C7H15O2 þ O2 = C7KET12 þ OH 72. C7KET12 = C5H11CO þ CH2O þ OH 73. C5H11CO = C2H4 þ C3H7 þ CO 74. C7H15-2 = C2H5 þ C2H4 þ C3H6 75. C3H7 = C2H4 þ CH3 76. C3H7 = C3H6 þ H 77. C3H6 þ CH3 = C3H5 þ CH4 78. C3H5 þ O2 = C3H4 þ HO2 79. C3H4 þ OH = C2H3 þ CH2O 80. C3H4 þ OH = C2H4 þ HCO 81. CH3 þ HO2 = CH3O þ OH 82. CH3 þ OH = CH2 þ H2O 83. CH2 þ OH = CH2O þ H 84. CH2 þ O2 = HCO þ OH 85. CH2 þ O2 = CO2 þ H2 86. CH2 þ O2 = CO þ H2O 87. CH2 þ O2 = CH2O þ O 88. CH2 þ O2 = CO2 þH þ H 89. CH2 þ O2 = CO þ OH þ H 90. CH3O þ CO = CH3 þ CO2 91. CO þ OH = CO2 þ H 92. O þ OH = O2 þ H 93. H þ HO2 = OH þ OH 94. OH þ OH = O þ H2O

925

b

E (cal/mol)

9.98  1010 1.50  1016 1.50  1011 1.50  1011 1.00  1013 4.00  1013 2.69  1010 5.00  1012 1.70  1012 2.80  1012 1.50  1011 4.52  10-1 9.40  104 9.65  104 5.25  109 8.40  1012 5.01  107 1.00  1013 1.00  1011 2.19  10-4 1.10  1013 1.75  1012 3.00  1014

0.0 0.0 0.0 0.0 0.0 0.0 0.8 0.0 0.0 0.0 0.0 3.6 2.8 2.7 1.0 0.0 1.8 0.0 0.0 4.5 0.0 0.0 0.0

20351.0 42000.0 3000.0 36730.0 0.0 4200.0 -340.0 1790.0 8440.0 13600.0 3000.0 7154.0 6280.0 3716.0 1590.0 20440.0 76.0 0.0 9200.0 -1000.0 3400.0 1350.0 70980.0

2.50  107 3.09  1015 3.39  106 3.50  1014 2.00  1013 9.85  1018 4.34  1027 6.31  1014 1.99  1012 1.13  1013 1.81  1011 7.00  1012 1.75  1010 1.40  1016 2.04  1014 4.38  107 9.70  109 1.65  1013 2.00  1015 4.68  1011 2.25  1014 6.67  1014 9.84  1015 7.04  1014 9.60  1013 1.25  1014 9.00  1012 6.00  1011 1.00  1012 1.00  1012 5.00  1013 7.50  106 2.50  1013 4.30  1010 6.90  1011 2.00  1010 5.00  1013 1.60  1012 8.60  1010 1.57  1014 3.51  107 4.00  1014 1.70  1014 6.00  108

1.8 -0.3 1.9 -0.6 0.0 -0.7 -3.4 0.0 0.0 0.0 0.0 0.0 0.0 -1.6 -0.4 2.0 1.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.3 -0.5 0.0 1.3

76.0 50820.0 179.0 5260.0 4200.0 81710.0 30470.0 42300.0 11670.0 30430.0 18480.0 -1000.0 -3275.0 1860.0 31450.0 4760.0 1690.0 16950.0 47380.0 0.0 18232.7 41100.0 40200.0 34600.0 30950.0 36900.0 8480.0 10000.0 0.0 0.0 0.0 5000.0 0.0 -500.0 500.0 -1000.0 9000.0 1000.0 -500.0 11800.0 -758.0 0.0 875.0 0.0

Energy Fuels 2010, 24, 916–927

: DOI:10.1021/ef901092h

Um and Park Table A1. Continued k = ATb exp(-E/RT)

reactions considered

A (mol cm s K)

95. H þ O2 þ M = HO2 þ M H2O enhanced by 2.100  101 CO2 enhanced by 5.000 H2 enhanced by 3.300 CO enhanced by 2.000 96. H2O2 þ M = OH þ OH þ M H2O enhanced by 2.100  101 CO2 enhanced by 5.000 H2 enhanced by 3.300 CO enhanced by 2.000 97. H2 þ OH = H2O þ H 98. HO2 þ HO2 = H2O2 þ O2 99. CH2O þ OH = HCO þ H2O 100. CH2O þ HO2 = HCO þ H2O2 101. HCO þ O2 = HO2 þ CO 102. HCO þ M = H þ CO þ M 103. CH3 þ CH3O = CH4 þ CH2O 104. C2H4 þ OH = CH2O þ CH3 105. C2H4 þ OH = C2H3 þ H2O 106. C2H3 þ O2 = CH2O þ HCO 107. C2H3 þ HCO = C2H4 þ CO 108. C2H5 þ O2 = C2H4 þ HO2 109. CH4 þ O2 = CH3 þ HO2 110. OH þ HO2 = H2O þ O2 111. CH3 þ O2 = CH2O þ OH 112. CH4 þ H = CH3 þ H2 113. CH4 þ OH = CH3 þ H2O 114. CH4 þ O = CH3 þ OH 115. CH4 þ HO2 = CH3 þ H2O2 116. CH4 þ CH2 = CH3 þ CH3 117. C3H6 = C2H3 þ CH3 118. CH2 þ CH2 = C2H2 þ H2 119. CH2 þ CH2 = C2H2 þ H þ H 120. C2H4 þ M = C2H2 þ H2 þ M 121. C2H2 þ O2 = HCO þ HCO 122. C2H2 þ O = CH2 þ CO 123. C2H2 þ H þ M = C2H3 þ M 124. C2H3 þ H = C2H2 þ H2 125. C2H3 þ OH = C2H2 þ H2O 126. C2H3 þ CH2 = C2H2 þ CH3 127. C2H3 þ C2H3 = C2H2 þ C2H4 128. C2H3 þ O = C2H2 þ OH 129. C2H2 þ OH = CH3 þ CO 130. C2H3 = C2H2 þ H 131. H2O2 þ OH = H2O þ HO2 declared duplicate reaction 132. OH þ OH(þM) = H2O2(þM) low pressure limit: 0.30410  1031, -0.46300  101, and 0.20490  104 133. H2O2 þ H = H2O þ OH 134. H2O2 þ O = OH þ HO2 135. H2O2 þ H = H2 þ HO2 136. H2O2 þ OH = H2O þ HO2 declared duplicate reaction 137. CO þ HO2 = CO2 þ OH 138. HO2 þ O = OH þ O2 139. HCO þ HO2 = CH2O þ O2 140. CH3O þ O2 = CH2O þ HO2 141. HO2 þ H = H2 þ O2 142. C2H3 þ O2 = C2H2 þ HO2 declared duplicate reaction 143. C2H3 þ O2 = C2H2 þ HO2 declared duplicate reaction 144. C2H4 þ O2 = C2H3 þ HO2 145. N þ NO T N2 þ O 146. N þ O2 T NO þ O 147. N þ OH T NO þ H 148. N2O þ O T N2 þ O2 149. N2O þ O T 2NO

926

b

E (cal/mol)

3.60  1017

-0.7

1.00  1016

0.0

45500.0

1.17  109 3.00  1012 5.56  1010 3.00  1012 3.30  1013 1.59  1018 4.30  1014 6.00  1013 8.02  1013 4.00  1012 6.03  1013 2.00  1010 7.90  1013 7.50  1012 3.80  1011 6.60  108 1.60  106 1.02  109 9.00  1011 4.00  1012 3.15  1015 1.20  1013 1.20  1014 1.50  1015 4.00  1012 1.02  107 5.54  1012 4.00  1013 3.00  1013 3.00  1013 1.45  1013 1.00  1013 4.83  10-4 4.60  1040 1.00  1012

1.3 0.0 1.1 0.0 -0.4 0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.6 2.1 1.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -8.8 0.0

3626.0 0.0 -76.5 8000.0 0.0 56712.3 0.0 960.0 5955.0 -250.0 0.0 -2200.0 56000.0 0.0 9000.0 10840.0 2460.0 604.0 8700.0 570.0 5500.0 00.0 00.0 5800.0 8000.0 900.0 410.0 0.0 0.0 0.0 0.0 0.0 -2000.0 46200.0 0.0

1.24  1014

-0.4

0.0

2.41  1013 9.55  106 4.82  1013 5.80  1014

0.0 2.0 0.0 0.0

3.01  1013 3.25  1013 2.97  1010 5.50  1010 1.66  1013 5.19  10-15

0.0 0.0 0.3 0.0 0.0 -1.3

2.12  10-6

6.0

9484.0

4.00  1013 3.50  1013 2.65  1012 7.33  1013 1.40  1012 2.90  1013

0.0 0.0 0.0 0.0 0.0 0.0

58200.0 330.0 6400.0 1120.0 10810.0 23150.0

0.0

3970.0 3970.0 7950.0 9560.0 23000.0 0.0 -3861.0 2424.0 820.0 3310.0

Energy Fuels 2010, 24, 916–927

: DOI:10.1021/ef901092h

Um and Park Table A1. Continued k = ATb exp(-E/RT)

reactions considered

A (mol cm s K)

150. N2O þ H T N2 þ OH 151. N2O þ OH T N2 þ HO2 152. N2O(þM) T N2 þ O(þM) low pressure limit: 0.62000  1015, 0.00000, and 0.56100  105 H2 enhanced by 2.000 H2O enhanced by 6.000 CH4 enhanced by 2.000 CO enhanced by 1.500 CO2 enhanced by 2.000 153. HO2 þ NO T NO2 þ OH 154. NO þ O þ M T NO2 þ M H2 enhanced by 2.000 H2O enhanced by 6.000 CH4 enhanced by 2.000 CO enhanced by 1.500 CO2 enhanced by 2.000 155. NO2 þ O T NO þ O2 156. NO2 þ H T NO þ OH

927

b

E (cal/mol)

4.40  1014 2.00  1012 1.30  1011

0.0 0.0 0.0

18880.0 21060.0 59620.0

2.11  1012 1.86  1015

0.0 0.0

-480.0 0.0

3.90  1012 1.32  1014

0.0 0.0

-240.0 360.0