Experimental Study on Particulate Emissions of a Methanol Fumigated Diesel Engine Equipped with Diesel Oxidation Catalyst

Abstract

In this article, the effects of fumigation methanol, diesel oxidation catalyst, and engine operation parameters (engine load and engine speed) on diesel smoke opacity, particulate mass concentration, particulate number concentration and the soluble organic fraction (SOF) in the particulate were investigated at certain selected operation conditions. Experiments were performed on a 4-cylinder direct injection diesel engine operating at three engine speeds and three loads for each engine speed. For each engine speed, there was a decrease of smoke opacity with increase in the level of fumigation methanol. The reduction was particularly obvious at the high engine load but was not significant at the low and medium engine loads. For all test conditions, fumigation methanol could effectively reduce the particulate mass and number concentrations. However, fumigation methanol increased the fraction of SOF in the particles. The DOC could further reduce the particulate mass and number concentrations as well as the fraction of SOF in the particles when the exhaust gas temperature was sufficiently high.

1. INTRODUCTION

Diesel engines have been increasingly used in comparison to spark ignition engines for trucks and passenger cars due to their higher thermal efficiency and durability. However, the high particulate matter (PM) emission from diesel engine is a major drawback. Epidemiologic and clinical investigations have suggested a strong link between particulate air pollution and certain diseases (Pope et al. 1995). Combustion optimization and advanced emission controls for engine emission reduction are mostly applicable to new engines, but not to in-use engines. With increasing public concern on the issues of energy security and air pollution, the development of alternative fuels such as biodiesel, ethanol and methanol to partly or totally replace diesel fuel, and to reduce PM emission, is becoming more important (Agarwal 2007).

Methanol has the advantage of being readily available. It is free of sulfur and aromatic compounds and its high hydrogen-to-carbon ratio may result in lower PM formation and emission. Moreover, it can be easily applied to in-use diesel engines and vehicles and at a relatively low cost through the fumigation method. Previous work has shown that diesel engines using fumigation methanol emit less NOx and PM but more CO and HC (Udayakumar et al. 2004; Cheng et al. 2008; Song et al. 2008). The increase in HC emission could lead to more particulate nucleation and condensation during the dilution process and thus affects particulate emission.

The reduction of particulate mass emission can be caused by a reduction in large particles, while there might be an increase in submicron particles. The finer particles are of major concern in diesel emission control because they are more difficult to handle by after-treatment devices. Many study mainly focused on total particle number but little is known about the ultra-fine (diameter < 100 nm) and nano-particles (diameter < 50 nm) emissions from methanol fumigated diesel engine. The application of oxygenated fuel in diesel engines might lead to particles containing a larger percentage of SOF. This effect has rarely been investigated for the methanol fumigated diesel engine.

Diesel oxidation catalysts (DOC) have been widely used in diesel engines to reduce CO and HC emissions. Also DOC has been found to reduce the particulate mass, mainly by reducing the SOF of the particulates (Bagley et al. 1998; Vaaraslanti et al. 2006). However, the DOC is assumed to reduce the amount of HC in the gas phase, which may result in smaller mass transfer onto the particles when the exhaust cools (Klein et al. 1998), resulting in smaller sized particles. Thus, there is a need to investigate the influence of the DOC on particle number concentration and size distribution.

In this work, the main objective was to investigate the effects of the level of fumigation methanol, the diesel oxidation catalyst and the engine operation parameters on the particulate mass concentration, and the particle number concentration and size distribution. The influence on the SOF was also investigated at selected engine loads and engine speeds.

2. EXPERIMENTAL SETUP

2.1. Test Engine and Fuels

The experimental setup is shown in Figure 1. A naturally aspirated, 4-cylinder direct injection diesel engine, with specifications shown in Table 1, was used for the experiments. This type of engine is commonly used in the trucks in Hong Kong and mainland China. The engine was coupled with an eddy-current dynamometer and the engine speed and torque were controlled by the Ono Sokki diesel engine test system. A methanol fuel rail and four fuel injectors were added to the inlet manifold of the engine to inject methanol into the intake manifold at a pressure of 0.35 MPa to form a lean homogenous methanol/air mixture. An electronic control unit was used to control the amount of methanol injected into the air intake by adjusting the width of the injection pulse. An Engelhard CCX8772A DOC was installed at the downstream end, 850 mm from the exhaust manifold, to further reduce the pollutants in the exhaust gas. The DOC is a catalyzed flow-through device which can oxidize carbon monoxide, gaseous hydrocarbons and liquid hydrocarbon particles (Hammond et al. 2007).

FIG. 1 Schematic of the experimental setup.

TABLE 1 Specifications of the test engine

Model	Isuzu 4HFI
Type	Water-cooled, In-line, four-cylinder, DI engine
Maximum power	88 kW/3200 rev/min
Maximum torque	285 Nm/1800 rev/min
Bore/stroke	112 mm/110 mm
Displacement	4334cc
Compression ratio	19.0 : 1
Injection pump type	Bosch in-line type
Injection nozzle	5-hole nozzle

The fuels used include Euro V diesel fuel with less than 10-ppm by weight of sulfur and industrial grade methanol. The lubricating oil used was a SAE 15W-40 full mineral oil. Major properties of the fuels are shown in Table 2. Fuel consumptions were measured using an electronic balance with a precision of 0.1 gram for methanol and a measuring cylinder for the diesel fuel. The fuel consumptions at each engine load are shown in Table 3.

TABLE 2 Properties of Euro V diesel fuel and methanol

Properties	Diesel	Methanol
Molecular formula	—	CH3OH
Molecular weight, g/mol	—	32
Density, kg/m^3	830	790
Lower heating value, MJ/kg	42.5	19.7
Heat of vaporization, kJ/kg	270	1178
Viscosity at 20°C, mPa·s	2.8	0.59
Oxygen content, wt.%	0	50
Sulfur content, ppm wt	<10	—
Stoichiometric fuel/air ratio	0.07	0.15
Flash point, °C	78	107
Ignition temperature, °C	316	470
Flame temperature, °C	2054	1890

TABLE 3 Diesel (D) and methanol (M) mass consumption rates and exhaust gas temperature (T)

	0.13 MPa			0.35 MPa			0.55 MPa
	D kg/h	M kg/h	T °C	D kg/h	M kg/h	T °C	D kg/h	M kg/h	T °C
1280 rev/min
M0	2.01	—	158	3.75	—	266	5.52	—	385
M10	1.88	0.42	156	3.52	0.78	262	5.07	1.01	378
M20	1.78	0.78	155	3.21	1.49	260	4.53	2.19	374
M30	1.69	1.11	157	2.90	2.23	259	4.05	3.28	368
1920 rev/min
M0	3.28	—	213	5.82	—	341	8.51	—	492
M10	3.14	0.51	211	5.35	1.12	337	7.78	1.71	477
M20	2.95	1.15	208	4.97	2.22	334	6.99	3.31	471
M30	2.83	1.79	209	4.64	3.23	331	6.25	4.82	462
2560 rev/min
M0	5.17	—	267	8.61	—	419	12.65	—	586
M10	4.86	0.97	259	8.04	1.57	416	11.59	2.39	576
M20	4.59	1.95	260	7.35	3.39	413	10.33	4.70	563
M30	4.41	2.91	264	6.81	4.78	409	9.39	6.84	559

2.2. Sampling and Analysis

A two stage Dekati mini-diluter (Wong et al. 2003) was used for diluting the exhaust gas for particle sampling. The dilutor provides primary dilution in the range of 8:1 to 6:1 depending on the engine operating conditions while the secondary dilution system provides a dilution of 8:1 to keep the sampling gas temperature below 52°C. The actual dilution ratio was evaluated based on measured CO₂ concentrations in the raw exhaust, in the background air and in the diluted exhaust.

Smoke opacity of the raw exhaust gas was measured with a Dieseltune smoke-meter (SPX DX.210). The primary dilution was delivered to a tapered element oscillating microbalance (TEOM 1105, Rupprecht & Patashnick Co., Inc.) to measure the particulate mass concentration, in which the main flow rate of sample was 1.5 l/min. The TEOM inlet was held at 47°C so that any sampled water droplets are driven off; however, this may also lead to the loss of semivolatile organic material attached to the particles (Soutar et al. 1999). The number concentration and size distribution of particles in the secondary dilution was measured by a TSI 3934 scanning mobility particle sizer (SMPS). The gas sample entering the SMPS was at atmospheric pressure and close to ambient temperature. The SMPS consists of a TSI 3071A differential mobility analyzer (DMA) and a TSI 3022 condensation particle counter (CPC). The SMPS was set to measure particles in the size range of 15 to 750 nm. Although the diluted exhaust particulate mass and number concentrations were measured, the results presented in this paper are the actual concentrations in the engine exhaust obtained by multiplying the measured concentrations with the corresponding dilution ratios.

The particulate samples for SOF analysis were collected from the secondary diluted exhaust using φ47 mm Teflon-coated glass fiber filter (Pallflex T60A20, Pallflex Inc., USA). Before taking the samples, the glass fiber filters were washed in ethanol and then placed in an oven at 150°C for one hour to burn off any organic compounds that might be present in the filters. Finally, the cleaned filters were stored in a desiccator for at least eight hours before weighing. The soluble organic fraction was determined by weighing the particulate samples before and after the Soxhlet extraction (Wheaton Science Products Inc., USA) for 12 h. The extraction solution contains toluene with a purity of more than 99.5% and anhydrous ethanol with a purity of more than 99.7%. The SOF extraction followed the procedures described in Bagley et al. (1998).

2.3. Test Procedure

In this study, experiments were performed at the engine speeds of 1280, 1920, and 2560 rev/min, corresponding to 40, 60, and 80% of the maximum engine speed. At each engine speed, three engine loads, namely, 0.13, 0.35, and 0.55 MPa, representing respectively low, medium, and high engine loads, were selected. Experiments were firstly carried out on the diesel fuel alone to build up a data base for comparison with those obtained with fumigation methanol. Experiments were then carried out with diesel fuel to take up 90% of the desired engine load while the rest of the desired load was supplemented by fumigation methanol. Experiments were repeated with the diesel fuel taking up 80% and 70% of the desired engine load, with fumigation methanol providing respectively 20% and 30% of the desired engine loads. Therefore in this article and in the figures, when we mention x% fumigation methanol or x% MeOH, we are referring to the case that fumigation methanol takes up x% of the engine load.

At each mode of operation, the engine was allowed to run for a few minutes until the exhaust gas temperature, the lubricating oil temperature and the cooling water temperature have attained steady-state values and data were measured subsequently. Throughout the tests, the cooling water temperature varied from 80 to 85°C while the lubricating oil temperature varied from 90 to 100°C, depending on the engine load and engine speed. Particulate mass concentration was continuously measured for five minutes at the exhaust tailpipe of the diesel engine and the average results are presented. Each experiment was conducted three times and the results were found to agree with each other within the 95% confidence level. For particle number concentration and size distribution, four measurements were made at each test condition and the average values are presented. The experimental standard errors in the measurements have been determined based on the method of Kline and McClintock (1953). The standard error at 95% confidence level is 1.8% for particulate mass concentration, 1.7% for particle number concentration and 2.5% for smoke opacity.

3. RESULTS AND DISCUSSION

3.1. Smoke Opacity and Particulate Mass Concentrations

Figure 2 shows the effect of fumigation methanol and DOC on smoke opacity. For each engine speed, the smoke opacity is very low at the low and medium loads but increases obviously at the high engine load of 0.55 MPa due to the increase of the amount of fuel burned in the diffusion mode. In all cases, the smoke opacity is less than 15 HSU. Thus it is difficult to identify the influence of engine speed on the smoke opacities, taking into consideration the resolution of the smoke-meter and the experimental uncertainties. However, there is a decrease of smoke opacity with increase in the level of fumigation methanol. The reduction is particularly obvious at the high engine load but is not significant at the low and medium engine loads. Similar results were also reported in Cheng et al. (2008).

FIG. 2 Effect of fumigation methanol and DOC on smoke opacity.

With the DOC, from low to high engine speed, there is no significant change in smoke opacity at low and medium engine loads, and there is less than 9% smoke opacity reduction at the high engine load of 0.55 MPa, indicating that the DOC has only minor or no effect on smoke opacity.

The reduction of smoke opacity with increase of fumigation methanol can be attributed to several factors. Firstly, there is a reduction of diesel fuel burned with an increase of fumigation methanol. Diesel fuel contains aromatics, which is a soot precursor, while methanol does not contain aromatics (Chang and Gerpen 1998; Lapuerta et al. 2008). Secondly, the reduction of diesel fuel consumption also means less diesel fuel burning in the diffusion mode and hence a reduction in soot formation. Thirdly, methanol does not contain C–C bonds, which can result in the decrease of soot formation. At higher engine load and higher level of fumigation, more diesel fuel is replaced by fumigation methanol, leading to more obvious reduction in soot formation and hence reduction in smoke opacity.

The variation of PM mass concentration, expressed in milligram per cubic meter, is shown in Figure 3. At each engine speed, there is an increase in PM mass concentration with engine load. At the engine speed of 1920 rev/min, for different level of fumigation methanol, the PM mass concentration varies from 8.7 to 11.8 mg/m³ for the engine load of 0.13 MPa, from 11.4 to 17.3 mg/m³ for the engine load of 0.35 MPa, and from 23.9 to 43.1 mg/m³ for the engine load of 0.55 MPa. Similar trends can also be found in Ning et al. (2004), Cheng et al. (2008), and Di et al. (2009). Also, for the same engine load, the PM mass concentration in general increases with increase in engine speed. At the engine load of 0.55 MPa, for diesel-only operation, the PM mass concentration is 41.0, 43.1, and 77.6 mg/m³, respectively, for the engine speeds of 1280, 1920, and 2560 rev/min. Di et al. (2008) also found higher PM mass concentration with higher engine speed in their investigation of ethanol–diesel blended fuel.

FIG. 3 Effect of fumigation methanol and DOC on particulate mass concentration.

However, there is a decrease in PM mass concentration when the engine is operated with fumigation methanol. Moreover, the reduction increases with increasing fumigation methanol. From low to high engine load, with 30% fumigation methanol, there is about 8–25, 26–45, and 39–45% reduction in PM mass emission, respectively, for the engine speeds of 1280, 1920, and 2560 rev/min. In Cheng et al. (2008), they found a maximum reduction of 49% in PM mass concentration when operating on ultra low sulfur diesel with fumigation methanol taking up 30% engine load at an engine speed of 1800 rev/min, the percentage reduction is similar to our results.

PM consists mainly of soot on which some organic or hydrocarbon compounds and sulfates have been adsorbed (Wang et al. 1997). Carbonaceous soot is formed in the center of the fuel spray where the air/fuel ratio is low. At high engine load, more fuel is sprayed and burned, some regions of the combustion chamber might be scavenged of oxygen and hence the soot oxidation process would be inhibited in these regions, resulting in a rapid increase in the PM mass concentration. The combustion deteriorates at higher engine speed, due to the reduction in volumetric efficiency associated with higher engine speed, leading to higher PM concentration at the higher engine speed, especially at high engine load. In the fumigation mode, less diesel fuel is consumed and hence less soot is formed, leading to a reduction in the PM concentration. The reduction of PM mass concentration seems to be dependent on the mass of methanol. Thus, the PM emission decreases with increasing level of fumigation methanol, reducing engine load and reducing engine speed.

In addition, the molecular weight and molecular structure of the fuel also influence engine-out hydrocarbon compositions and thus PM emissions (Wang et al.1997). The molecular weight of methanol is 32, which is much less than that of the diesel fuel. Diesel fuel consists of a blend of complex, heavy molecules that include aromatics and a wide range of unsaturated compounds. The hydrogen to carbon ratio is low, and there is a tendency to form soot precursors. The composition of the unburned and partially oxidized hydrocarbons in the diesel exhaust is much more complex and extends over a larger range of molecular size than those for methanol. The heavier hydrocarbons associated with diesel exhaust might condense to form particulate upon cooling. Also, methanol does not contain any sulfur. All these factors contribute to the lower PM emissions when the engine is operated with fumigation methanol.

However, the results from this investigation reveal that the fumigation methanol is more beneficial in reducing PM mass concentration at the high engine load. At the low engine load, PM emissions are not significantly reduced. Other oxygenated fuel also has been shown to be more effective in reducing PM at high engine loads than at low engine loads (Hallgren and Heywood 2001; Cheng et al. 2002; Song et al. 2002). Two factors likely contribute to this effect. The first is that less fuel is injected at low engine load and therefore less fuel is burned during the mixing-controlled phase of combustion. Since soot formation occurs primarily during this mixing-controlled phase, the effect of the oxygenate on PM reduction would be less pronounced at low engine load. Secondly, at low engine load, the unburned hydrocarbon emission is high when fumigation methanol is applied (Cheng et al. 2008; Zhang et al. 2009). Thus, the PM mass associated with adsorbed or condensed HC might have increased, leading to less significant reduction in PM (Cheng et al. 2002).

Previous research showed that the DOC can effectively reduce the emission of total particulate matter due to the oxidation of the SOF in the particulate (Pataky et al. 1994; Hosoya and Shimoda 1996) and hence a reduction of health hazardous substances in the particulate matter (Bagley et al. 1998). In this study, after passing through the DOC, the PM mass concentration can be further reduced except at the engine load of 0.13 MPa at 1280 rev/min and 1920 rev/min. For example, at the engine load of 0.13 MPa, there is only about 2–4% reduction of the PM mass concentration, however, there is about 20–45% reduction at other engine loads and at the engine speed of 2560 rev/min. Moreover, the percentage reduction is higher at the engine speed of 2560 rev/min than at the lower engine speeds due to the higher exhaust gas temperature.

3.2. Soluble Organic Fraction (SOF) in Particles

The SOF in the PM consists of unburned, pyrolyzed, or partially oxidized fuel and lubricating oil which is transferred from the gas phase to the particulate phase when the exhaust cools (Alander et al. 2004). In this study, the engine speed of 1920 rev/min was selected to investigate the effects of fumigation methanol and the DOC on SOF in the PM, for diesel fuel and for 10% and 30% fumigation methanol. The proportion of SOF in the PM was analyzed using the Soxhlet extraction method. The results are shown in Figure 4.

FIG. 4 Effect of fumigation methanol and DOC on SOF in the particle at 1920 rev/min.

The SOF in the PM decreases with increasing engine load; however, it increases with increasing fumigation methanol. For the baseline engine operating on Euro V diesel fuel, the SOF in the PM decreases from 79 to 50% with the engine load increasing from 0.13 to 0.55 MPa, while the corresponding SOF decreases from 85 to 58% with 10% fumigation methanol and from 90 to 65% with 30% fumigation methanol. Lin and Chao (2002) also found increase of SOF when the engine was operated with diesel–methanol blends at different operating conditions. At high engine loads, the higher exhaust temperature favors the oxidation of hydrocarbon resulting in less SOF in the PM. With fumigation methanol, the soot in the particles is reduced and the unburned HC, including unburned methanol and formaldehyde, adsorbed on the PM may increase, contributing to the increase of SOF proportion in the PM.

Due to oxidation of the SOF, the percentage of SOF in the PM with DOC is about 5–32% less than that without DOC. However, the percentage of SOF in the PM is still 5–10% higher at 10% fumigation and 9–20% higher at 30% fumigation compared with that of the baseline engine operating on the diesel fuel.

3.3. Particle Number Concentration

The influence of PM to the environment and human health depends on its mass concentration as well as on its number concentration. In this investigation, the SMPS was set to measure particles within the size range of 15–750 nm. Besides the total particle number concentration within the measured size range, concentrations of nano-particles with diameter Dp < 50 nm and ultra-fine particles with diameter Dp < 100 nm were also investigated. Since only particles larger than 15 nm were measured, nano-particles less than this size have been excluded. The nano-particles and the ultra-fine particles may have more influence on human health because they have higher specific surface areas for adsorbing harmful chemicals (Peters et al. 1997; Somers et al. 2004; Pope, III and Dockery 2006).

As shown in Figure 5, for each engine speed, regardless of the mode of fueling the engine, there is an increase of particle number concentration with increase of engine load. And, for the same engine load, the total particle number concentration increases with increase in engine speed. For example, for 30% fumigation methanol, the total number concentration increased by 25% when the engine load increases from 0.13 to 0.55 MPa for the engine speed of 1280 rev/min, while the corresponding increase is 44 and 46% for the engine speeds of 1920 and 2560 rev/

min, respectively. As suggested by Lapuerta et al. (2008), at higher engine load and speed, with the higher amount of burning fuel mass, combustion takes place with lower excess oxygen, which coupled with higher pressure and temperature levels in the combustion chamber, contributing to soot nucleation and promoting the growth of the existing soot nuclei. It is also noticed from Table 4 that for each fuel, the geometrical mean diameter (GMD) of the particles increases with engine load as well. In fact, at high engine loads, more fuel is consumed in the diffusion mode and hence more particles are formed. With an increase in the number of particles, coagulation rate increases and hence larger particles are formed, leading to the increase of GMD.

FIG. 5 Effect of fumigation methanol and DOC on particle number concentration.

TABLE 4 Geometric mean diameter (GMD) of particles

	0.13 MPa		0.35 MPa		0.55 MPa
	Before DOC/nm	After DOC /nm	Before DOC/nm	After DOC/nm	Before DOC/nm	After DOC/nm
1280 rev/min
M0	53.3	52.2	58.6	58.2	67.8	68.4
M10	54.2	54.1	57.9	57.1	68.3	68.6
M20	54.2	54	58.8	57	67.8	68.8
M30	54.1	53.2	57.5	56.2	67.6	68.2
1920 rev/min
M0	48.3	47.2	53.1	51.4	61.6	62.7
M10	48.2	47.1	51	50.9	57.8	61
M20	48.3	47.0	50.6	49.2	56.3	59.8
M30	47.6	46.9	49.5	48.3	53.8	58.4
2560 rev/min
M0	49.4	49.8	52.9	52.6	62.7	64.0
M10	48.4	48.8	53.7	52.4	61.3	62.5
M20	47.4	47.7	53.3	52.7	60.1	61.2
M30	47.3	48.5	51.5	51.3	58.1	60.4

However, there is a decrease in the number of total particles with increase in level of fumigation methanol. At each engine speed, the percentage reduction decreases with increase of the engine load. For example, at the engine load of 0.13 MPa, with different fumigation methanol, there is about 26–44%, 18–49%, and 27–44% reduction, for the engine speeds of 1280, 1920, and 2560 rev/min. However, the corresponding values are 9–22%, 14–42%, and 12–35% at the engine load of 0.55 MPa. Cheng et al. (2008) found that, with fumigation methanol, the total particle number concentration decreased at low and medium engine load but slightly increased at high engine load at the engine speed of 1800 rev/min. Their results are similar to ours despite ultra low sulfur diesel fuel instead of Euro V diesel fuel was used in their investigation. The reduction of particles with increase of fumigation level is a direct consequence of the reduction of diesel fuel burned in the diffusion mode. The particle number concentration and size distribution is affected by the combustion process as well as the condensation and adsorption of hydrocarbon and the coagulation and agglomeration of particles in the exhaust gas. Actually, the increase in HC emission with increase of fumigation level could lead to increase in the number of particles formed. Ning et al. (2004) have shown that increase in HC emission could lead to more particulate nucleation, condensation and coagulation to occur during the dilution process, especially at high engine load when the exhaust gas at higher temperature is cooled. The combined effect of these factors is a higher percentage reduction in particle number concentration at low engine load and a lower percentage reduction at high engine load.

The fraction of nano-particles and ultra-fine particles in the total particle number concentration was also investigated. The results are shown in Table 5. In this study, over 35% of the particles are nano-particles with Dp < 50 nm and no less than 70% of them are ultra-fine particles with Dp < 100 nm. For the same engine load, the fraction of nano-particles is smaller at low engine speed than at medium and high engine speeds leading to the larger GMD at the low engine speed, as shown in Table 4. While for each engine speed, regardless of the mode of fueling the engine, the fraction of both nano-particles and ultra-fine particles decreases with increase in engine load, indicating that the particles tend to be larger in size with increasing engine load. Similar results can also be found in Di et al. (2009) for blended fuels. Probably, with an increase of engine load, the increase of particle concentration enhances particle collision and agglomerations, resulting in the formation of larger particles and a reduction in the fraction of nano-particles and fraction of ultra-fine particles. When fumigation methanol is applied, the total particle number concentration, as well as the nano-particle and ultra-fine particle concentrations, all decrease, resulting in no significant change in the GMD and slight change in the fraction of nano-particles and the fraction of ultra-fine particles.

TABLE 5 Fraction of nano-particles and ultra-fine particles (F) and percentage reduction of the particles after passing through the DOC (PR)

	0.13 MPa				0.35 MPa				0.55 MPa
	Dp < 50 nm		Dp < 100 nm		Dp < 50 nm		Dp < 100 nm		Dp < 50 nm		Dp < 100 nm
	F %	PR %	F %	PR %	F %	PR %	F %	PR %	F %	PR %	F %	PR %
1280 rev/min
M0	47.1	10.7	84.7	14.7	42.5	13.7	80.9	17.1	32.6	23.4	73.6	25.8
M10	46.0	8.8	84.0	12.4	43.1	14.4	81.1	18.2	30.0	16.8	71.9	27.8
M20	46.0	8.6	83.8	12.5	43.8	15.1	80.7	18.1	31.9	24.9	72.2	26.8
M30	46.1	8.7	83.9	11.8	44.3	13.0	82.0	15.3	32.1	29.5	72.6	33.0
1920 rev/min
M0	56.1	10.7	90.7	12.4	50.2	23.1	85.1	23.5	38.0	21.0	78.6	24.1
M10	56.4	8.7	90.5	10.3	53.0	26.4	88.8	27.2	41.5	19.7	80.9	18.2
M20	56.2	8.7	90.3	10.5	52.9	27.3	88.6	29.6	41.5	17.5	81.0	16.7
M30	56.0	9.3	89.9	11.3	54.0	25.1	88.9	27.4	43.4	20.1	82.0	20.3
2560 rev/min
M0	54.1	23.2	90.6	24.9	49.2	30.0	87.8	24.4	37.2	20.1	79.8	20.4
M10	56.1	24.0	90.0	24.3	48.7	27.9	85.9	22.4	39.0	34.2	82.8	25.2
M20	56.7	18.9	90.2	20.9	45.3	28.3	83.3	23.7	40.4	27.7	81.3	26.4
M30	56.9	15.7	89.9	16.9	51.2	23.8	88.3	24.7	42.7	27.1	82.2	24.1

After passing through the DOC, particles in the different groups can be further reduced by 9–31% at different engine loads and speeds with different percentage of fumigation methanol. For each engine speed, the percentage reduction is larger at the medium engine load of 0.35 MPa than at the other two engine loads. For example, at the engine speed of 1920 rev/min, there is 23–27% reduction in total particle number concentration with different fumigation level at 0.35 MPa. While the corresponding values are 10–12% and 12–14%, respectively, for the engine loads of 0.13 and 0.55 MPa. There is a slight increase in the GMD at high engine load after passing through the DOC. The DOC is effective in reducing the unburned hydrocarbon in the exhaust gas and the soluble organic fraction of the particles. The former might lead to a reduction in the particle number concentration due to a reduction of the particles formed by the nucleation of the unburned hydrocarbon, especially at high engine loads. On the other hand, the DOC provides a large number of channels through which the particles pass through. There is a higher tendency for the particles to interact with each other and with the channel walls, resulting in the formation of larger particles and a reduction in the number of particles. The overall result is a higher reduction of the small-size particles and lower reduction of the large-size particles; leading to reduction in the number concentration but increase in the GMD, particularly at the high engine load. Moreover, the function of DOC is related to the exhaust gas temperature and duration of time the particles reside in the DOC. Lower exhaust temperature associated with low engine load and shorter residence time in the DOC associated with the higher exhaust flow at high engine load result in lower reduction of particle number concentration at low and high engine loads than at medium engine load.

4. CONCLUSIONS

Experiments were conducted on a 4-cylinder direct-injection diesel engine operating on Euro V diesel fuel and fumigation methanol under different engine loads and engine speeds to investigate the PM emissions from the engine. A DOC was equipped for investigating its influence on the PM emissions. The following conclusions can be drawn from this study.

For each engine speed, the smoke opacity is very low at the low and medium loads but increases obviously at high engine load due to the increase of the amount of fuel burned in the diffusion mode. However, there is a decrease of smoke opacity with increase in the level of fumigation methanol. The reduction is particularly obvious at the high engine load but not significant at the low and medium engine loads.

Regardless of the mode of fueling the engine, the PM mass and number concentrations increase with increase in engine load and engine speed. For each engine speed, the fraction of both the nano-particles and the ultra-fine particles decrease with increase in engine load, indicating that the particles tend to be larger in size with increasing engine load. Fumigation methanol can effectively reduce the particulate mass and number concentrations, including the nano-particle and ultra-fine particle concentrations, resulting in no significant change in the GMD under all the tests conducted. However, fumigation methanol increases the fraction of SOF in the particles.

The particulate mass and number concentrations both decreases with increasing fumigation methanol. However, the percentage reduction of particulate mass concentration is higher at high engine load while the percentage reduction of particle number concentration is higher at lower engine load.

The DOC can further reduce the mass and number concentrations as well as the fraction of SOF in the particles when the exhaust gas temperature is sufficiently high. Moreover, the percentage reduction increases with engine load and speed due to the high exhaust gas temperature. At high engine load, there is an increase of GMD of the particles after passing through the DOC but the changes are less obvious at light and medium engine loads.

Acknowledgments

The authors are thankful for the financial support from The Hong Kong Polytechnic University, the Research Grants Council of the Hong Kong SAR (Project No. PolyU 5139/07E), and the National Science Foundation Committee of China (Contract No. 50876075).