Experimental Study on Combustion and Emissions Characteristics of a


Experimental Study on Combustion and Emissions Characteristics of a...

3 downloads 137 Views 3MB Size

2930

Energy & Fuels 2009, 23, 2930–2936

Experimental Study on Combustion and Emissions Characteristics of a Spark Ignition Engine Fueled with Gasoline-Hydrogen Blends Changwei Ji* and Shuofeng Wang* College of EnVironmental and Energy Engineering, Beijing UniVersity of Technology, Beijing 100124, China ReceiVed March 10, 2009. ReVised Manuscript ReceiVed April 27, 2009

Concerning the low thermal efficiency and high emissions of spark ignition (SI) engines, an experimental study was performed aiming at improving SI engine economic and emissions performance under a typical city driving condition of 1500 rpm and at the stoichiometric equivalence ratio with four intake manifold absolute pressures (MAP) of 39.3, 48.1, 56.4, and 67.8 kPa and four hydrogen volumetric fractions of 1, 1.5, 2, and 3% on a modified 4-cylinder gasoline-fueled SI engine on which hydrogen can be injected into the intake ports sequentially via a self-developed electronic control unit (DECU) and mixed with gasoline online. The DECU can be used to adjust the injection timings and durations of hydrogen and gasoline to accomplish the specified volumetric fractions of hydrogen in air and the mixture equivalence ratios. The engine test results demonstrated that, compared with the original gasoline engine, the brake thermal efficiency and the cycle-by-cycle variation of the peak in-cylinder pressure were averagely increased from 25.12 to 28.35% and reduced from 6.8 to 3.62%, respectively, at the hydrogen addition fraction of 3% and four specified MAPs. Both ignition delay and the rapid burning duration were shortened; the peak in-cylinder pressure and temperature were increased; CO2 and HC emissions were obviously lower with the increasing fraction of hydrogen in the intake. However, NOx and CO emissions were also raised. In a word, hydrogen addition is a potentially applicable measure to improve SI engine performance.

1. Introduction Concerns about the energy consumption and environmental pollutions have been more and more pressing in recent years. Hydrogen has been proved to be a promising alternative energy for IC engines.1,2 Hydrogen can be used to fuel IC engines and feed fuel cells, which are main automobile movers now and in the future.3 Although the fuel cells gain many advantages, such as high power output and zero emissions while operating, the high cost and short lifetime have become the concerns of their wide application. At the same time, there are still many headaches in their manufacturing and commercialization that block their popularization.4 Hydrogen is more suitable to be used on SI engines than CI engines due to its high ignition temperature.5-9 The diffusion coefficient of hydrogen is 10 times as large as that of gasoline, which benefits the homogeneity of the in-cylinder charge. The cyclic variation of hydrogen-fueled engines is usually lower than that of gasoline-fueled engines * Corresponding author. Phone: +86 1067392126; fax: +86 1067392126; e-mail addresses: [email protected](C.J.), [email protected]. (1) Ferguson, C. R.; Kirkpatrick, A. T. Internal Combustion Engines; John Wiley & Sons: New York, 2001; p 326. (2) Veziroglu, T. N.; Barbir, F. H. Int. J. Hydrogen Energy 1992, 17, 391–404. (3) Rankin, D. D. Lean Combustion; Elsevier: London, 2008; pp 214215. (4) Shudo, T. Int. J. Hydrogen Energy 2007, 32, 4285–4293. (5) Andrea, T. D.; Henshaw, P. F.; Ting, D. S. K. Int. J. Hydrogen Energy 2004, 29, 1541–1552. (6) DAS, L. M. Int. J. Hydrogen Energy 1996, 21, 703–715. (7) Heywood, J. B. Internal Combustion Engine Fundamentals; McGrawHill: New York, 1988; pp 463-464. (8) Kahraman, E.; Ozcanl, S. C.; Ozerdem, B. Int. J. Hydrogen Energy 2007, 32, 2066–2072. (9) Padiyar, S. Properties of hydrogen. In Proceedings of Summer School of Hydrogen Energy; IIT Madras: Chennai, India, 1985.

since the addition of hydrogen helps increase the fuel burning speed, which permits a short combustion duration.10-12 Also, the low ignition energy of hydrogen eases the engine starting process, although it sometimes causes abnormal combustion such as preignition and backfire. The pure-hydrogen-fueled engines sometimes produce a weak power output because of the low volume energy density of hydrogen,13 And the NOx emissions from a pure H2 fueled engine are much higher than that from a gasoline-fueled engine at the stoichiometric conditions, due to the high adiabatic flame temperature of H2.14 Different from pure-hydrogen-fueled engines, using hydrogen as an additive to SI engines takes advantages of both gasoline and hydrogen and improves the engine efficiency and emissions while keeping the engine power output.5 The use of hydrogen as an additive to enhance engine performance has been tested on gasoline, methane, ethanol, and biogas fueled engines.15-23 Ma et al.24-27 investigated the effect of hydrogen addition fraction, spark timing, and excess air factor on the cycle-by(10) Ma, F.; Wang, Y.; Liu, H.; Li, Y.; Wang, J.; Ding, S. Int. J. Hydrogen Energy 2008, 33, 823–831. (11) Huang, Z.; Zhang, Y.; Wang, Q.; Wang, J.; Jiang, D.; Miao, H. Energy Fuels 2006, 20, 2385–2390. (12) Huang, Z.; Liu, B.; Zeng, K.; Huang, Y.; Jiang, D.; Wang, X.; Miao, H. Energy Fuels 2007, 21, 2594–2599. (13) Ganesh, R. H.; Subramanian, V.; Balasubramanian, V. Renewable Energy 2008, 33, 1324–1333. (14) Haroun, A.; K, S. A.; Maher, A. R. S. A. Int. J. Hydrogen Energy 1999, 24, 363–375. (15) Maher, A. R. S. A. B.; Haroun, A.; K, S. A. J. Energy ConVers. Manage. 2000, 41, 77–91. (16) Akansu, S. O.; Kahraman, N.; Ceper, B. Int. J. Hydrogen Energy 2007, 32, 4279–4284. (17) Porpatham, E.; Ramesh, A.; Nagalingam, B. Int. J. Hydrogen Energy 2007, 32, 2057–2065. (18) Maher, A. R. S. A. B. Int. J. Hydrogen Energy 2000, 25, 1005– 1009.

10.1021/ef900209m CCC: $40.75  2009 American Chemical Society Published on Web 05/18/2009

Spark Ignition Engine Fueled with Gasoline-H2 Blends

Energy & Fuels, Vol. 23, 2009 2931 Table 1. Engine Specifications rated rated torque @ power@ No. of bore/ stroke/ displacement/ compression 4500 6000 cylinders mm mm L ratio rpm/Nm rpm/kW 4

Figure 1. Gasoline and hydrogen fuel rails and injectors installed on the intake manifolds.

cycle variation, combustion process, and emissions characteristics of a CNG engine with hydrogen enrichment. Varde28 tested a single-cylinder engine with the mixture of hydrogen and gasoline and found that the lean burn limit was extended and that the flame propagation speed was increased with the addition of hydrogen. Apostolescu and Chiriac29 studied the effect of hydrogen addition on the combustion process at mid-to-low loads with the results showing that the cyclic variation and 10-90% burn duration were greatly reduced while hydrogen mass fraction varied from 1.5 to 3%. In the previous studies of hydrogen-enriched gasoline engines, hydrogen was mostly premixed with air in a special tank or inducted into the plenum of intake manifolds.5,8 However, those methods increase the possibilities of backfire and have to face the poor test flexibility and accuracy. In this paper, a modified intake manifolds with a hydrogen fuel rail and four hydrogen injectors that permits the electronic port injection of hydrogen was introduced, as shown in Figure 1. Since hydrogen has much higher flame and diffusion speeds than gasoline, which are powerful to stimulate gasoline combustion, and because hydrogen is still at high cost, using small hydrogen addition fractions is more favorable for its application. In this study, an electronically controlled gasoline and hydrogen injection system has been developed to investigate the effect of hydrogen addition on gasoline engine combustion and emissions characteristics under various engine loads using four hydrogen volume fractions below 3% at a typical city driving engine speed of 1500 rpm and stoichiometric condition.

77.4

85.0

1.599

10

143.28

82.32

manufactured by Beijing Hyundai Motors. The engine has the specifications shown in Table 1. To avoid backfire and large space requirement caused by hydrogen and air premixing system in the previous researches, the intake manifolds was modified to realize hydrogen port injection, as shown in Figure 1. Four hydrogen injectors were mounted below the intake manifolds without any modifications to the original gasoline fuel injection system. The hydrogen injectors were mechanically coupled and sealed to the hydrogen rail containing 0.3 MPa hydrogen. The closing and opening of hydrogen injectors were controlled by a DECU, which was dedicatedly developed for this research. The DECU communicates with the engine original ECU (OECU) and a calibration computer to govern the hydrogen and gasoline injection timings and durations based on the sensor signals obtained from the OECU and commands from the calibration computer. The experimental system is shown in Figure 2. Engine test and control system includes a GW160 eddy current dynamometer, a FC2210 gasoline mass flow meter, and a FC2010 control panel. In-cylinder pressure data acquisition and combustion analysis system consists of a Kistler 2613B optical encoder, a Kistler 6117BCD17 cylinder pressure transducer with a spark plug, and a Dewetron combustion analyzer. The cylinder pressure transducer with a spark plug is screwed into the cylinder head of the fourth cylinder to collect the combustion cylinder pressure and enforce ignition. The optical encoder is connected to the front of the crankshaft producing 1800 pulses per rotation for obtaining crank angles and triggering the sampling of combustion pressure. The cylinder pressure transducer and optical encoder are connected to the Dewetron combustion analyzer via the screened cables. Cylinder pressure and crank angle signals for over 300 consecutive cycles are sampled and treated via DEWE-CA combustion analysis software embedded in the Dewetron combustion analyzer to obtain cylinder pressure against crank angle profiles, heat release rate against crank angle profiles, heat release fraction against crank angle profiles, etc.

2. Experimental Setup and Procedure 2.1. Experimental Setup. The test engine is an in-line, fourcylinder, 1.6 L port fuel injection, spark-ignited gasoline engine (19) Raju, A. V. R.; Ramesh, A.; Nagalingam, B. Performance of a spark ignited engine using natural gas and hydrogen supplemented natural gas. In Proceedings of the Fourth International Conference ICE99, Italy, September 12-16, 1999; pp 285-289. (20) Liu, B.; Huang, Z.; Zeng, K.; Chen, H.; Wang, X.; Miao, H.; Jiang, D. Energy Fuels 2008, 22, 273–277. (21) Huang, Z.; Liu, B.; Zeng, K.; Huang, Y.; Jiang, D.; Wang, X.; Miao, H. Energy Fuels 2006, 20, 2131–2136. (22) Yousufuddin, S.; Mehdi, S. N.; Masood, M. Energy Fuels 2008, 22, 3355–3362. (23) Ji, C.; Wang, S. Int. J. Hydrogen Energy 2009, 34, 3546–3556. (24) Ma, F.; Ding, S.; Wang, Y.; Wang, Y.; Wang, J.; Zhao, S. Int. J. Hydrogen Energy 2008, 33, 7245–7255. (25) Ma, F.; Wang, Y. Int. J. Hydrogen Energy 2008, 33, 1416–1424. (26) Ma, F.; Wang, Y.; Liu, H.; Li, Y.; Wang, J.; Zhao, S. Int. J. Hydrogen Energy 2007, 32, 5067–5075. (27) Ma, F.; Wang, J.; Wang, Y.; Wang, Y.; Li, Y.; Liu, H.; Ding, S. Energy Fuels 2008, 22, 1880–1887. (28) Varde, K. S. Combustion characteristics of small spark ignition engines using hydrogen supplemented fuel mixtures. SAE paper 810921. (29) Apostolescu, N.; Chiriac, R. A study of combustion of hydrogenenriched gasoline in a spark ignition engine. SAE paper 960603.

Figure 2. The schematics of the experimental system. (1) hydrogen cylinder assembly fence, (2) hydrogen pressure adjusting valve, (3) hydrogen pressure meter, (4) hydrogen mass flow meter, (5) backfire arrestor, (6) hydrogen injector, (7) throttle, (8) air mass flow meter, (9) idle valve 10. original ECU (OECU), (11) developed ECU (DECU), (12) calibration computer, (13) fuel tank, (14) fuel mass flow meter, (15) fuel pump, (16) fuel injector, (17) ignition module, (18) pressure transducer with a spark plug, (19) optical encoder (20) charge amplifier, (21) A/D converter, (22) combustion analyzer, (23) O2 sensor, (24) A/F analyzer, (25) emissions sampling pipe, (26) MEXA-7100 emissions analyzer. (a) Signals from OECU to DECU, (b1) calibration and control signals from the calibration computer to DECU, (b2) data signals from DECU to calibration PC.

2932

Energy & Fuels, Vol. 23, 2009

Ji and Wang

Table 2. Uncertainties and Sensitivities of Experimental Instruments name

uncertainty/sensitivity

dynamometer fuel flow meter optical encoder pressure transducer exhaust gas analyzer

air flow meter hydrogen flow meter A/F analyzer

type

uncertainty ( 1 rpm of speed uncertainty ( 0.4% F.S. of torque uncertainty ( 0.4% F.S. sensitivity 0.2 °CA uncertainty e0.01 °CA uncertainty ( 0.6% FSO NOx: sensitivity 1 ppm; uncertainty e1% CO2: sensitivity 0.01%; uncertainty e1% HC: sensitivity 1 ppm; uncertainty e1% CO: sensitivity 1 ppm; uncertainty e1% uncertainty ( 1% F.S uncertainty ( 1% F.S uncertainty ( 0.1 A/F at A/F ) 14.7

Table 3. Test Matrix of Spark Timings against MAPs MAP/kPa

measured spark timing range (°CA BTDC)

39.3 48.1 56.4 67.8

29.8 27.3 23.1 19.3

The exhaust emissions of NOx, HC, CO, and CO2 from the test engine were measured by a Horiba MEXA-7100D emissions analyzer. NOx was measured by chemiluminescent method; HC emissions were determined by hydrogen flame ionization detection (FID) method; and CO and CO2 were measured by nondispersive infrared (NDIR) method. The analyzer was calibrated prior to the experimental tests with zero and span gases. The air mass flow rate was monitored by an EPI-800 thermal mass flowmeter. Hydrogen is stored in an assembly fence containing 16 hydrogen cylinders at 16 MPa and injected into the engine intake ports at 0.3 MPa after two steps of pressure reduction, and hydrogen mass flow rate was measured by a D07-19BM thermal mass flowmeter. The uncertainty and/or sensitivity for each test instrument are shown in Table 2. 2.2. Experimental Procedure. All engine tests in this study were carried out at 1500 rpm and stoichiometric condition with four intake manifolds absolute pressures (MAP) of 39.3, 48.1, 56.4, and 67.8 kPa. Under a specified MAP, first the original engine tests were finished using pure gasoline, and then the gasoline-hydrogen mixture fueled engine tests were carried out with four hydrogen volume fractions of 1, 1.5, 2, and 3% in the intake. During the experiment, the spark timing was controlled by the OECU automatically. A data acquisition system was used for detecting the spark timing at each test point. The measured results showed that engine spark timings were kept the same at the specified MAPs for different hydrogen blending levels. Table 3 lists the measured spark timings at four MAPs. The various hydrogen volume fractions in the intake can be accomplished by increasing the hydrogen injection duration and decreasing the gasoline injection duration simultaneously to keep the gasoline-hydrogen mixture burnt at stoichiometric condition. Hydrogen volume fraction in total intake gas is represented by R ) V˙H2/(V˙H2 + V˙air) × 100%, where V˙H2 and V˙air are the hydrogen and air volume flow rates at normal states (L/min). The total equivalence ratio of hydrogen-enriched SI engine can be defined as the following equation:30

φ ) (m ˙ gAFst,g + m ˙ H2AFst,H2)/m ˙ air

manufacturer

GW160

Powerlink

FC2210 2613B

Powerlink Kistler

6117BFD17 MEXA-7100D

Kistler Horiba

EPI-800 D07-19BM MEXA -110

EPI Seven Star Horiba

The equivalence ratio was also monitored by a Horiba MEXA110 A/F analyzer ((0.1 A/F at A/F ) 14.7) with an O2 sensor inserted into the exhaust pipe. The H/C molar ratio of the instrument was adjusted with the measured gasoline and hydrogen mass flow rates, so that it can well measure the A/F ratio of the hydrogenenriched SI engine. The experiment results showed that deviation between the calculated and measured equivalence ratios of the hydrogen-enriched SI engine was within 3%. The combustion and emissions characteristics of the engine were investigated by burning pure gasoline and the mixtures of gasoline and hydrogen with four hydrogen addition levels. DEWE-CA software was applied to calculate heat release fraction and the percentage coefficient of variation (% COV) of peak cylinder pressure (Pmax) of 300 consecutive cycles for comparison purpose. The coolant temperature and lubricant oil temperature were kept at 90 °C ( 1 °C and 95 °C ( 1 °C, respectively, to minimize its effect on test results.

3. Results and Discussion 3.1. Engine Brake Thermal Efficiency. Better gasoline atomization, evaporation, and mixing with air will accelerate the gasoline combustion and produce a higher thermal efficiency. Hydrogen fast diffusion collision and low ignition energy are beneficial to gasoline atomization and earlier combustion. Engine brake thermal efficiency against MAP profiles are shown in Figure 3 at 1500 rpm, stoichiometric condition, and four hydrogen volume fractions. Figure 3 shows that engine brake thermal efficiency increases obviously with hydrogen volume fraction at all tested MAPs. Engine brake thermal efficiency on the average increases from 25.12 to 28.35% when hydrogen addition level is 3% by volume compared with that of the original engine at four specified MAPs.

(1)

In eq 1, m ˙ g, m ˙ H2, and m ˙ air represent the measured gasoline, hydrogen, and air mass flow rates respectively [kg/h]; AFst,g and AFst,H2 are the stoichiometric air-to-fuel ratios of gasoline and hydrogen (AFst,g ) 14.6, AFst,H2 ) 34.37). The gasoline flow rate was adjusted according to the calculation results of eq 1 to ensure the hydrogen-gasoline mixture burnt at the stoichiometric condition. (30) Li, J.; Guo, L.; Du, T. Int. J. Hydrogen Energy 1998, 23, 971– 975.

Figure 3. Variation of brake thermal efficiency with MAP at different H2 volume fractions.

Spark Ignition Engine Fueled with Gasoline-H2 Blends

Energy & Fuels, Vol. 23, 2009 2933

Figure 5. Variation of in-cylinder pressure with cylinder volume at MAP ) 67.8 kPa.

Figure 4. Variation of Pmax and θPmax with MAP at different H2 volume fractions. (a) Peak in-cylinder pressure. (b) Crank angle relevant to Pmax.

Because hydrogen propagates and combusts fast, gasolinehydrogen mixture fueled engine can work much closer to the ideal constant volume cycle. Meanwhile, as hydrogen propagates quickly and produces a short quenching distance, as a result, hydrogen addition to a SI engine is beneficial for enhancing the evaporation of the liquid gasoline and the mixing of the gasoline vapor with the air, causing a more complete combustion and a higher thermal efficiency. 3.2. Engine Combustion Analysis. 3.2.1. Peak In-cylinder Pressure and Its Corresponding Crank Angle. The values of engine peak in-cylinder pressure and its corresponding crank angle represent the engine working capability to some large extent. Engine peak in-cylinder pressure Pmax and its corresponding crank angle θPmax (°CA ATDC) against MAP profiles are shown in Figure 4 at 1500 rpm, stoichiometric condition and four hydrogen volume fractions. Figure 4a shows that engine peak cylinder pressure Pmax increases obviously with the increase of MAP and hydrogen volume fraction. The larger hydrogen volume fraction is the bigger increase in Pmax is. Compared with the original one, Pmax is averagely increased by 580 kPa for the 3% hydrogen-enriched engine at four MAPs. It can be seen in Figure 4b that θPmax advances with the increase of hydrogen volume fraction. The original engine θPmax

is from 15 to 16.5 °CA ATDC, and in the hydrogen-enriched gasoline engine θPmax is from 8.7 to 12.6 °CA ATDC at 3% hydrogen addition. When hydrogen volume fraction is 3%, θPmax on the average advances 5.6 °CA compared with that of the original engine at four specified MAPs. Moreover, hydrogen enrichment is very effective for reducing θPmax, especially at low loads. This is because, at low loads, gasoline fuel evaporates and mixes with air in a poor condition, resulting in the lower combustion velocity, but hydrogen-enriched SI engine combustion can be improved distinctively due to the fact that hydrogen propagates and burns fast. However, it also can be seen from Figure 4b that θPmax is not gradually advanced with the increase of MAP as expected. This is because the spark timing of each hydrogen enrichment level is set the same as the spark timing of the original engine under the specified MAPs. At high loads, the OECU does not apply the MBT spark timing but a retarded spark timing to avoid engine knock. As shown in Table 3, spark timing is retarded from 29.8 °CA BTDC at the MAP of 39.3 kPa to 19.3 °CA BTDC at the MAP of 67.8 kPa. So θPmax is not linearly advanced with the increase of MAP at the same hydrogen blending level. 3.2.2. Analysis of Constant-Volume Combustion by P-V Diagram. The indicated diagram is a kind of description of engine working efficiency. The larger pressure-volume (P-V) area is the higher engine power output is. Figure 5 displays the P-V diagram of compression and expansion strokes excluding the intake and exhaust strokes for five hydrogen enrichment levels at 1500 rpm and MAP ) 67.8 kPa. As it is seen in Figure 5, due to the faster burning of the gasoline-hydrogen mixture, the degree of constant volume combustion is gradually enhanced with the increase of hydrogen addition level, which helps the increase in engine thermal efficiency.31 Moreover, the postcombustion duration is also reduced with the increase of hydrogen blending level. 3.2.3. In-cylinder Temperature. In-cylinder temperature is heavily related to combustion efficiency and the formation of emissions. The higher and more advanced peak in-cylinder temperatures are the higher combustion efficiency and NOx emissions are. The in-cylinder temperature can be simply calculated by gas state equation.32 The variation of in-cylinder (31) Shudo, T.; Nabetani, S.; Nakajima, Y. Int. J. Engine Res. 2001, 2 (1), 81–92. (32) Ma, F.; Liu, H.; Wang, Y.; Li, Y.; Wang, J.; Zhao, S. Int. J. Hydrogen Energy 2008, 33, 816–822.

2934

Energy & Fuels, Vol. 23, 2009

Ji and Wang

Figure 6. Variation of in-cylinder temperature for 0 and 3% hydrogen fractions with crank angle at MAP ) 67.8 kPa.

temperature with crank angle for 0 and 3% hydrogen fractions at 1500 rpm and MAP ) 67.8 kPa are compared in Figure 6. It can be seen that the peak in-cylinder temperature was increased after hydrogen addition, because hydrogen has a higher adiabatic flame temperature than gasoline. Moreover, since the flame speed of the fuel-air mixture in the cylinder was enhanced with the addition of hydrogen, the heat release of the hydrogen-gasoline mixture can be accomplished within a much shorter time, causing the increased peak in-cylinder temperature. However, the in-cylinder temperature of the hydrogen-enriched engine drops much faster than the original one after reaching its peak value due to the short and fast burning and the reduced postcombustion duration after hydrogen addition. Generally, the reduced postcombustion temperature means a reduced exhaust loss and an increased thermal efficiency.1,7 3.2.4. Flame DeVelopment Period and Flame Propagation Period. The two important parameters for evaluating engine combustion quality are flame development period and flame propagation period. The shorter flame development and flame propagation periods indicate the earlier ignition and faster combustion of the fuel-air mixture in the cylinder. Flame development period (CA0-10) is defined as the combustion crank angle duration of 0-10% heat release of the total gasolinehydrogen fuel mixture. Flame propagation period (CA10-90) is defined as the combustion crank angle duration of 10-90% heat release of the total gasoline-hydrogen fuel mixture. CA0-10 and CA10-90 against MAP profiles are shown in Figure 7, a and b, respectively, at 1500 rpm, stoichiometric condition, and four hydrogen volume fractions. It can be seen from Figure 7 that CA0-10 and CA10-90 gradually decrease with the increase of MAP and drop sharply with the increase of hydrogen volume fraction, causing much shorter combustion durations using gasoline-hydrogen mixtures than using pure gasoline. When hydrogen volume fraction is 3%, fast burning period CA10-90 on the average reduces 5.2 °CA compared with that of the original engine at four specified MAPs. Moreover, although the spark timings were kept unchanged before and after the addition of hydrogen at the same MAP, the CA0-10 drops obviously with the increase of hydrogen volume fraction, producing much earlier ignition and quicker combustion using gasoline-hydrogen mixtures than using pure gasoline. This is attributed to that the gasoline-hydrogen mixture is much easier to ignite and combust than pure gasoline due to the early ignition of hydrogen and propagation of hydrogen

Figure 7. Variation of CA0-10 and CA10-90 with MAP at different H2 volume fractions.

Figure 8. HRF against crank angle for 0 and 3% hydrogen fractions at MAP ) 67.8 kPa.

flame, which results in the increase of charge temperature, consequently stimulating the combustion of gasoline-hydrogen mixture and surely reducing CA0-10 and CA10-90 of the gasoline-hydrogen mixture. The addition of hydrogen helps

Spark Ignition Engine Fueled with Gasoline-H2 Blends

Energy & Fuels, Vol. 23, 2009 2935

angle for 0 and 3% hydrogen volume fractions at MAP ) 67.8 kPa. From Figure 8 we can find that the heat release is well accelerated by hydrogen enrichment, and this phenomenon means a gasoline-hydrogen mixture fueled engine can obtain higher thermal efficiency than a pure gasoline engine because higher gasoline-hydrogen mixture flame speed reduces combustion duration and the heat transfer to the cylinder wall. 3.2.5. Cyclic Variation of Peak In-cylinder Pressure. It has been universally agreed upon that very serious combustion instability exists in SI engines, especially at low loads. Engine combustion cyclic variation can be estimated by statistically analyzing the fluctuations of engine key combustion parameters such as indicated mean effective pressure, peak in-cylinder pressure, etc. Percentage coefficient of variation of peak incylinder pressure, COVPmax is defined as

∑ q

Figure 9. Variation of COVPmax with MAP at different H2 volume fractions.

the formation of OH radical, which is thought to be effective on improving the chain reactions and reducing the fuel burning duration.33 Figure 8 shows the variation of accumulative heat release fraction of gasoline-hydrogen fuel mixture (HRF) against crank

COVPmax ) [

(Pmax i - Pmax)2 /(m - 1)/Pmax] × 100%

i)1

(2)

j max ) (∑im) 1Pmax i)/m, where Pmax i is the peak inIn eq 2, P j max is the average of cylinder pressure of the ith engine cycle; P the peak in-cylinder pressures of m cycles; and m is the number of engine consecutive cycles. Percentage coefficient of variation of peak in-cylinder pressure, COVPmax against MAP profiles

Figure 10. Variations of exhaust emissions with MAP at various H2 addition levels.

2936

Energy & Fuels, Vol. 23, 2009

are shown in Figure 9 at 1500 rpm, stoichiometric condition and four hydrogen volume fractions. Figure 9 shows that COVPmax decreases with the increase of hydrogen volume fraction. The higher engine loads are the smaller COVPmax is. When hydrogen volume fraction is 3%, COVPmax was averagely decreased from 6.8 to 3.62% at four specified MAPs. Furthermore, the lower engine load or MAP is the larger COVPmax drop appears. At 3% hydrogen volume fraction and the lowest test load of MAP ) 39.3 kPa, COVPmax decreases from 8.8 to 4% with a drop rate of 55.6%, which is the biggest drop rate compared with other 3 MAPs. This can be attributed to the fast hydrogen combustion causing engine combustion process closer to the ideal constant volume cycle with the increase of hydrogen volume fraction, consequently improving the combustion instability and cyclic variation in engine cylinders. 3.3. Emissions. For SI engines burning hydrocarbon fuels, HC, CO, NOx, and CO2 in combustion pollutants are regulated to meet different but more and more stringent emissions standards. Better combustion reduces some pollutants but increases other emissions, needing some trade-off measures. In this study, HC, CO, NOx, and CO2 emissions before a 3-way catalytic converter were measured to know the effect of hydrogen addition on a SI gasoline engine emissions. The profiles of HC, CO, NOx, and CO2 emissions against MAP are shown in Figure 10 at 1500 rpm, stoichiometric condition, and four hydrogen volume fractions. Figure 10 shows that HC and CO2 decrease and CO and NOx increase with the increase of hydrogen volume fraction. When hydrogen volume fraction is 3%, HC and CO2 on the average decrease by 21.57 and 14.76%, respectively, at four specified MAPs. However, when hydrogen volume fraction is 3%, CO and NOx on the average increase by 62.03 and 32.56%, respectively, at four specified MAPs. The possible reason for the reduction of HC is that the short quenching distance of hydrogen makes the gasoline-hydrogen-air mixture burn much closer to the crevices between piston and cylinder wall, consequently effectively reducing the HC emissions caused by the crevice effect. Meanwhile, at stoichiometric condition, hydrogen addition results in the reduction of gasoline fueling rate, and no carbon is contained in hydrogen, as a result, CO2 emission is reduced. The inhomogeneity of the in-cylinder charge and fast burn characteristic of hydrogen may be the possible reason for the increased CO emission. Due to the inhomogeneity and the fast hydrogen propagation, hydrogen may be burned in some oxygen-rich area, the fast burning properties and high stoichiometric air-to-fuel ratio of hydrogen make it consume more adjacent oxygen, causing CO existing in the relatively oxygen-lean area and damping the reaction kinetics of CO into CO2. Another possible explanation can be described as that the faster flame speed of hydrogen-gasoline mixture (fast burning of hydrogen-gasoline mixture produces a short postcombustion period, providing insufficient temperature and time for CO conversion to CO2) decreases the reaction time for CO to be fully oxidized into CO2. The formation of NOx emissions is thought to be related with the peak in-cylinder pressure. Generally, the higher peak in-cylinder temperature indicates the larger amount of NOx emissions.7 Since the peak in-cylinder temperature increases with hydrogen addition (as (33) Conte, E.; Boulouchos, K. Influence of hydrogen-rich-gas addition on combustion, pollutant formation and efficiency of an IC-SI engine. SAE paper 2004-01-09724; 2004.

Ji and Wang

Figure 6 shows), the NOx emissions from a hydrogen-enriched gasoline engine are also higher than the original one. 4. Conclusions In this paper, an experimental investigation aiming at the effect of hydrogen addition to a gasoline fueled SI engine on improving its economic and emissions performance has been carried out at 1500 rpm and stoichiometric conditions. The main results are summarized as follows: (1) Engine brake thermal efficiency increases with hydrogen volume fraction under the tested MAPs. When hydrogen volume fraction is 3%, engine brake thermal efficiency on the average increases from 25.12 to 28.35% compared with the original engine. (2) Pmax increases and θPmax advances with the increase of hydrogen volume fraction. Pmax and θPmax on the average increase by 580 kPa and advances 5.6 °CA, respectively. The fast burning period CA10-90 on the average reduces 5.2 °CA when hydrogen volume fraction is 3% compared with that of the original engine. (3) The peak in-cylinder temperature increases and advances, while the temperature for postcombustion duration decreases with the addition of hydrogen. (4) COVPmax decreases with the increase of hydrogen volume fraction and COVPmax on the average decreases from 6.8 to 3.62% at 3% hydrogen volume fraction. At 3% hydrogen addition level and MAP ) 39.3 kPa, COVPmax decreases from 8.8 to 4% with a drop rate of 55.6%, which is the biggest drop rate compared with the other three MAPs. (5) HC and CO2 decrease and CO and NOx increase with the increase of hydrogen volume fraction. When hydrogen volume fraction is 3%, HC and CO2 on the average decrease by 21.57 and 14.76%, respectively, within the test MAP range (from 39.3 to 67.8 kPa), and CO and NOx on the average increase by 62.03 and 32.56%, respectively. Acknowledgment. This work was supported by Beijing Municipal Natural Science Foundation Project (Grant No. 3082004) and Beijing Municipal Committee of Education-Scientific Research Base Construction Project (Grant No. 0050005366901). The authors also want to acknowledge all students in the group for their dedicated help in the experiment.

Nomenclature AF ) air to fuel ratio ATDC ) after top dead center CA ) crank angle MAP ) intake manifolds absolute pressure (kPa) HRF ) accumulative heat release fraction of gasoline-hydrogen fuel mixture (%) φ ) total equivalence ratio m ˙ g ) the measured gasoline mass flow rate (kg/h) m ˙ H2 ) the measured hydrogen mass flow rate (kg/h) m ˙ air ) the measured air mass flow rate (kg/h) AFst,g ) the stoichiometric air to fuel ratio of gasoline (14.6) AFst,H2 ) the stoichiometric air to fuel ratio of hydrogen (34.3) R ) hydrogen volume fraction (%) V˙H2 ) hydrogen volumetric flow rate at 298 K and atmospheric pressure (L/min) V˙air ) hydrogen volumetric flow rate at 298 K and atmospheric pressure (L/min) CA0-10 ) CA duration of 0 to 10% heat release (°CA) CA10-90 ) CA duration of 10 to 90% heat release (°CA) Pmax ) peak in-cyliner pressure (100 kPa) θPmax ) crank angle relevant to Pmax (°CA ATDC) COVPmax ) coefficient of variation in Pmax (%) EF900209M