Exhaust Gas Characteristics According to the Injection Conditions in Diesel and DME Engines

: In this paper, the effect of high-pressure injection pressure on particulate matter (PM) and nitrogen oxide (NO x ) emissions is discussed. Many studies have been conducted by active researchers on high-pressure engines; however, the problem of reducing PM and NO x emissions is still not solved. Therefore, in the existing diesel (compression ignition) engines, the common rail high-pressure injection system has limitations in reducing PM and NO x emissions. Accordingly, to solve the exhaust gas emission problem of a compression ignition engine, a compression ignition engine using an alternative fuel is discussed. This study was conducted to optimize the dimethyl ether (DME) engine system, which can satisfy the emission gas exhaust requirements that cannot be satisﬁed by the current common rail diesel compression ignition engine in terms of efﬁciency and exhaust gas using DME common rail compression ignition engine. Based on the results of this study on diesel and DME engines under common rail conditions, the changes in engine performance and emission characteristics of exhaust gases with respect to the injection pressure and injection rate were examined. The emission characteristics of NO x , hydrocarbons, and carbon monoxide (CO) emissions were affected by the injection pressure of pilot injection. Under these conditions, the exhaust gas characteristics were optimized when the pilot injection period and needle lift were varied.


Introduction
In order to prevent air pollution caused by exhaust gas from automobiles, exhaust emission regulations such as Tier-III and EURO-VI have been announced in various countries such as USA, Europe and Japan. Many researchers are working to reduce harmful emissions by gradually increasing the allowable regulation of hazardous emissions from automobiles [1]. Diesel engines have been widely used as power sources for automobiles and ships because of their high thermal efficiency and fuel efficiency. It has the advantage of low carbon monoxide (CO) and unburned hydrocarbons (HC) because it burns under lean conditions where the fuel consumption rate is lower than that of the gasoline engine and the combustion state is excess oxygen. However, since it has the disadvantage of high PM (particulate matter) and NOx (nitrogen oxides) emissions, it cannot satisfy the increasingly stringent emission regulations [2,3].
Using diesel particulate filters (DPF) together with diesel oxidation catalysts (DOC) is a technology that can reduce the emissions of diesel cars to the level of the latest gasoline vehicles. There have been many studies on nitrogen oxides in various countries depending on the social environment and economic conditions. However, in Europe, selective catalytic reduction (SCR) and US prefer the NOx storage catalyst (LNT, Lean NOx trap) [4]. The exhaust emission reduction technique The experimental engine shown in Table 1 is an engine from Daedong, which is a water-cooled single-cylinder four-cycle direct-injection diesel engine. Its specifications are total displacement of 673 cc, maximum output of 9.11 kW with 2200 rpm, and maximum torque of 4.22 kg m with 1800 rpm.
The injection period pulse (using AM503B current probe amplifiers from Tektronix, Beaverton, OR, USA) was performed for the common rail injector injection characteristics and compared with the actual injection amount. The line pressure according to the two-way pulse was measured (using 4067A2000 amplifiers from KISTLER, Winterthur, Switzerland) to confirm the pilot of the injection lift operating period and the pressure fluctuation of the main injection. The injection needle lift was measured (using signal condition amplifiers from WOLF, Boston, MA, USA) to compare it with the line pressure under the loading condition of the engine. In addition, the injection rate, which is the core of the common rail high-pressure injection characteristics, was measured (Japan KYOWA Electric Corporation (Tokyo, Japan), signal condition: CDV700A, bridge box: DB120, strain gauge: KFG-03). Figure 2 shows the fuel injection control system and test engine configuration. The engine control system in this study applied the ECD-U2 common rail system manufactured by NIPPON DENSO (Tokyo, Japan). To input the engine revolutions and crank positions into the engine control unit (ECU), Ne pulse (crank position pulse) for injection characteristics experiments and Ge pulse (cylinder detect pulse) embedded in the supply pump were used. For the measurement of exhaust gases, non-dispersive infrared spectrometry (using MEXA 324JK from HORIBA) was performed for CO and THC, the chemiluminescence method (using the 10 AR model of the thermo environmental instrument) for NOx, and the reflection photometer method (AFT-2000 from World Environment) for smoke. The data obtained from these instruments were used as a representative value using the average value for the three measurement results.  The experimental engine shown in Table 1 is an engine from Daedong, which is a water-cooled single-cylinder four-cycle direct-injection diesel engine. Its specifications are total displacement of 673 cc, maximum output of 9.11 kW with 2200 rpm, and maximum torque of 4.22 kg m with 1800 rpm. The injection period pulse (using AM503B current probe amplifiers from Tektronix, Beaverton, OR, USA) was performed for the common rail injector injection characteristics and compared with the actual injection amount. The line pressure according to the two-way pulse was measured (using 4067A2000 amplifiers from KISTLER, Winterthur, Switzerland) to confirm the pilot of the injection lift operating period and the pressure fluctuation of the main injection. The injection needle lift was measured (using signal condition amplifiers from WOLF, Boston, MA, USA) to compare it with the line pressure under the loading condition of the engine. In addition, the injection rate, which is the core of the common rail high-pressure injection characteristics, was measured (Japan KYOWA Electric Corporation (Tokyo, Japan), signal condition: CDV700A, bridge box: DB120, strain gauge: KFG-03). Figure 2 shows the fuel injection control system and test engine configuration. The engine control system in this study applied the ECD-U2 common rail system manufactured by NIPPON DENSO (Tokyo, Japan). To input the engine revolutions and crank positions into the engine control unit (ECU), Ne pulse (crank position pulse) for injection characteristics experiments and Ge pulse (cylinder detect pulse) embedded in the supply pump were used. For the measurement of exhaust gases, non-dispersive infrared spectrometry (using MEXA 324JK from HORIBA) was performed for CO and THC, the chemiluminescence method (using the 10 AR model of the thermo environmental instrument) for NO x , and the reflection photometer method (AFT-2000 from World Environment) for smoke. The data obtained from these instruments were used as a representative value using the average value for the three measurement results.

Experimental Methodology
Fuel injection conditions were changed by controlling the injection timing, injection duration, injection pressure, pilot injection rate, and injector needle lift using the ECU program.
The injection timing (10°, 14°, 18°, and 22° before top dead center (BTDC)) was adjusted using the common rail injection device manual controller based on the injection timing (22° BTDC) of the mechanical injection pump in accordance with the specifications of the single-cylinder horizontal diesel engine that was used as the test engine. In this instance, the injection lift (0.14, 0.21, 0.24, and 0.36 mm) was changed based on the injection nozzle (ψ = 0.22 mm × 5 ea), and the injection characteristics, engine performance, and exhaust gas characteristics were investigated according to the pilot and main injections. The engine revolutions were 1800 rpm, load ratio was 75%, and common rail injection pressure was 50 MPa. The braking thermal efficiency was focused on maintaining the current common rail engine and reducing NOx among the exhaust gases [10][11][12]. Figure 3 shows the effect of high-pressure injection pressure on PM and NOx emissions. Many studies have been conducted by active researchers on the high-pressure engine; however, the problem of reducing PM and NOx emissions is still not solved. In the existing diesel (compression ignition) engines, the common rail high-pressure injection system has limitations in reducing PM and NOx emissions. Therefore, to solve the exhaust gas emission problem of the compression ignition engine, a compression ignition engine using an alternative fuel was studied [21][22][23][24]. In this study, the exhaust gas problem, which cannot be solved by the existing common rail diesel compression ignition engine, was solved with the optimization of a DME engine system that can satisfy both efficiency and exhaust gas emission.

Experimental Methodology
Fuel injection conditions were changed by controlling the injection timing, injection duration, injection pressure, pilot injection rate, and injector needle lift using the ECU program.
The injection timing (10 • , 14 • , 18 • , and 22 • before top dead center (BTDC)) was adjusted using the common rail injection device manual controller based on the injection timing (22 • BTDC) of the mechanical injection pump in accordance with the specifications of the single-cylinder horizontal diesel engine that was used as the test engine. In this instance, the injection lift (0.14, 0.21, 0.24, and 0.36 mm) was changed based on the injection nozzle (ψ = 0.22 mm × 5 ea), and the injection characteristics, engine performance, and exhaust gas characteristics were investigated according to the pilot and main injections. The engine revolutions were 1800 rpm, load ratio was 75%, and common rail injection pressure was 50 MPa. The braking thermal efficiency was focused on maintaining the current common rail engine and reducing NO x among the exhaust gases [10][11][12]. Figure 3 shows the effect of high-pressure injection pressure on PM and NO x emissions. Many studies have been conducted by active researchers on the high-pressure engine; however, the problem of reducing PM and NO x emissions is still not solved. In the existing diesel (compression ignition) engines, the common rail high-pressure injection system has limitations in reducing PM and NO x emissions. Therefore, to solve the exhaust gas emission problem of the compression ignition engine, a compression ignition engine using an alternative fuel was studied [21][22][23][24]. In this study, the exhaust gas problem, which cannot be solved by the existing common rail diesel compression ignition engine, was solved with the optimization of a DME engine system that can satisfy both efficiency and exhaust gas emission. Appl. Sci. 2019, 9, x FOR PEER REVIEW 5 of 21  Table 2 presents the causes of PM and NOx generation and the mitigation measures. To solve the problem of exhaust gas emission from conventional fossil fuels of compression ignition engines, DME, which is a natural gas reforming fuel, was used as an alternative fuel. The thermal efficiency of the compression ignition engines using the DME fuel was studied to determine the optimal injection conditions to maintain the current level of common rail diesel engines and simultaneously reduce the PM and NOx emissions.

Emission
Methods

NOx
Injection timing delay Lowering of combustion temperature [35] Exhaust gas recirculation Lowering of combustion temperature [36] Lowering of oxygen thickness [37] Water fulmination Lowering of combustion temperature [38] Fuel-water injection

Pilot injection
Lowering of combustion rate in the first period [39] Improving cetane number Lowering of combustion rate in the first period [40] After treatment NOx recharge and resolve [41][42][43][44][45] PM Spray improvement Air inhalation increase in spray [46] Mixing increase [47] Optimization in combustion chamber Mixing increase [48] Increase in air usage rate [49] Confusion combustion Mixing increase [50] Increase of smoke oxidation [51] Table 2 presents the causes of PM and NO x generation and the mitigation measures. To solve the problem of exhaust gas emission from conventional fossil fuels of compression ignition engines, DME, which is a natural gas reforming fuel, was used as an alternative fuel. The thermal efficiency of the compression ignition engines using the DME fuel was studied to determine the optimal injection conditions to maintain the current level of common rail diesel engines and simultaneously reduce the PM and NO x emissions.

Emission
Methods

NO x
Injection timing delay Lowering of combustion temperature [35] Exhaust gas recirculation Lowering of combustion temperature [36] Lowering of oxygen thickness [37] Water fulmination Lowering of combustion temperature [38] Fuel-water injection

Effects of Injection Pressure on Exhaust Gas Emission Characteristics of Diesel and DME Engines
1. NO x emission characteristics of diesel and DME engines Figure 4 shows a comparison of the exhaust gas emission characteristics of DME and diesel engines depending on the injection speed and pressure of the engines at an injection timing of 14 • BTDC and 25% load factor. Figure 4 shows a comparison of the exhaust gas emission characteristics of DME and diesel engines depending on the injection speed and pressure of the engines at an injection timing of 14° BTDC and 25% load factor. NOx emissions in DME engines are not significantly different from those in diesel engines; however, the emission characteristics of the fuel injection pressure are different.
In particular, NOx emissions are lowest at 2100 rpm in DME engines, unlike those in diesel engines. This suggests that the DME engine used in this study has an unstable combustion state because of the non-optimized fuel injection conditions, such as injection timing and amount according to the injection pressure of the fuel. Thus, the combustion state of the DME fuel, which is an oxygen fuel, is generally improved, thereby increasing the combustion temperature and NOx emission concentration [58,59]. This is consistent with the emission characteristics of other internal combustion engines. Therefore, it suggests that it is necessary to investigate the possibility of reducing NOx emissions in DME engines using the fuel injection pressure characteristics.  . NO x emission characteristics of diesel and DME engines without pilot injection as a function of injection pressure and engine speed. NO x emissions in DME engines are not significantly different from those in diesel engines; however, the emission characteristics of the fuel injection pressure are different.
In particular, NO x emissions are lowest at 2100 rpm in DME engines, unlike those in diesel engines. This suggests that the DME engine used in this study has an unstable combustion state because of the non-optimized fuel injection conditions, such as injection timing and amount according to the injection pressure of the fuel. Thus, the combustion state of the DME fuel, which is an oxygen fuel, is generally improved, thereby increasing the combustion temperature and NO x emission concentration [58,59]. This is consistent with the emission characteristics of other internal combustion engines. Therefore, it suggests that it is necessary to investigate the possibility of reducing NO x emissions in DME engines using the fuel injection pressure characteristics.
2. HC emission characteristics of diesel and DME engines Figure 5 shows the emission characteristics of the unburned HC in DME and diesel engines depending on the engine operating speed and injection pressure with light load and injection timing of 14 • BTDC.
At the engine operating speed of 1200-1800 rpm, the HC emissions were slightly lower than the diesel engine emissions. However, at the operating speed of 2100 rpm, the HC emissions of the DME engines were about 2.5 times higher than those of the diesel engines. This indicates that the combustion characteristics of oxygenated fuel appeared in the low-speed range but in the case of 2100 rpm the emission of HC was increased compared with 1200-1800 rpm by incomplete combustion. The excessive injection amount because of the high-pressure injection and the inconsistency of the injection timing was shown in the high-speed operation condition. In DME engines, the effect of fuel injection pressure on the HC emission characteristics were not significant at low speeds; however, at high speeds, the higher was the fuel injection pressure, the lower was the HC emission. To optimize the DME engines, it was necessary to optimize the fuel injection pressure and injection amount for the engine operating conditions. of 14° BTDC.
At the engine operating speed of 1200-1800 rpm, the HC emissions were slightly lower than the diesel engine emissions. However, at the operating speed of 2100 rpm, the HC emissions of the DME engines were about 2.5 times higher than those of the diesel engines. This indicates that the combustion characteristics of oxygenated fuel appeared in the low-speed range but in the case of 2100 rpm the emission of HC was increased compared with 1200-1800 rpm by incomplete combustion. The excessive injection amount because of the high-pressure injection and the inconsistency of the injection timing was shown in the high-speed operation condition. In DME engines, the effect of fuel injection pressure on the HC emission characteristics were not significant at low speeds; however, at high speeds, the higher was the fuel injection pressure, the lower was the HC emission. To optimize the DME engines, it was necessary to optimize the fuel injection pressure and injection amount for the engine operating conditions. 3. CO emission characteristics of diesel and DME engines Figure 6 shows the CO emission characteristics of DME and diesel engines depending on the engine operating speed and injection pressure with light load and injection timing of 14° BTDC. The emission characteristics of CO were similar to those of HC. The CO emission concentration of the DME engine was slightly lower than that of the diesel engine at all operating conditions and not significantly affected by the fuel injection pressure. This was because of the combustion characteristics of the oxygenated fuel of the DME, as described in HC. 3. CO emission characteristics of diesel and DME engines Figure 6 shows the CO emission characteristics of DME and diesel engines depending on the engine operating speed and injection pressure with light load and injection timing of 14 • BTDC. The emission characteristics of CO were similar to those of HC. The CO emission concentration of the DME engine was slightly lower than that of the diesel engine at all operating conditions and not significantly affected by the fuel injection pressure. This was because of the combustion characteristics of the oxygenated fuel of the DME, as described in HC. Figure 7 shows a comparison of the soot concentration with the injection pressure for operating speed of DME and diesel engines at the engine operating speed of 1800 rpm, light load condition, and injection timing of 14 • BTDC. The exhaust gas emission characteristics of DME engines were not discharged as compared to those of the diesel engine [33,34,60]. As a result of injecting high-pressure liquid fuel DME, which is a compressible oxygen fuel, the mixing ratio with air in the injection characteristic improved as compared to that with diesel, thereby enhancing the diffusive combustion state due to inject the high-pressure fuel. Appl. Sci. 2019, 9, x FOR PEER REVIEW 8 of 21 shows a comparison of the soot concentration with the injection pressure for operating speed of DME and diesel engines at the engine operating speed of 1800 rpm, light load condition, and injection timing of 14° BTDC. The exhaust gas emission characteristics of DME engines were not discharged as compared to those of the diesel engine [33,34,60]. As a result of injecting high-pressure liquid fuel DME, which is a compressible oxygen fuel, the mixing ratio with air in the injection characteristic improved as compared to that with diesel, thereby enhancing the diffusive combustion state due to inject the high-pressure fuel.

Effects of Injection Pressure and Pilot Injection Period on Exhaust Gas Emission Characteristics of DME Engines
1. NOx emission characteristics of diesel and DME engines depending on the injection pressure and pilot injection period Figure 8 shows that the NOx emissions of DME as a diesel alternative fuel was more than those of diesel near the top dead center (TDC) at the same diesel injection timing of the experimental engine. As the injection pressure increased, the amount of NOx increased, and pilot injection showed that the diesel could be reduced to 20% (100 ppm) and DME to 10% (50 ppm). In this experiment, based on improving the diesel injection timing as much as possible and the pilot injection about 15% of the total injection amount, it was confirmed that NOx could be reduced to about 10-20% of the total Load25%/BTDC14˚/NZ0.22×6

Effects of Injection Pressure and Pilot Injection Period on Exhaust Gas Emission Characteristics of DME Engines
1. NO x emission characteristics of diesel and DME engines depending on the injection pressure and pilot injection period Figure 8 shows that the NO x emissions of DME as a diesel alternative fuel was more than those of diesel near the top dead center (TDC) at the same diesel injection timing of the experimental engine.
As the injection pressure increased, the amount of NO x increased, and pilot injection showed that the diesel could be reduced to 20% (100 ppm) and DME to 10% (50 ppm). In this experiment, based on improving the diesel injection timing as much as possible and the pilot injection about 15% of the total injection amount, it was confirmed that NO x could be reduced to about 10-20% of the total emission rate by high-pressure fuel injection and pilot injection. For pilot injection at 6 • before the 14 • BTDC, where combustion occurred, the thermal efficiency decreased with increasing fuel consumption rate; however, it was expected that the NO x generated in a high-temperature region would be reduced because the temperature of the combustion chamber decreases. 2. HC emission concentration in diesel and DME engines depending on the injection pressure and pilot injection period Figure 9 shows the HC emission with pressure during high-pressure pilot and main injections. HC was reduced by more than 50% in this study as compared to those in other studies (injection pressure 20-35 MPa) [24,26] for high-pressure pilot injection. HC was generated in the combustion chamber wall or in a state where the air-fuel ratio was rich, and it was generated in a state where the fuel was vaporized and not oxidized. Moreover, DME, as an oxygenated fuel, generated almost no HC, and generated higher fuel density per unit volume. Furthermore, unlike the emission characteristics of NOx, the emission rate of HC was reduced by the increase of pressure. It could be expected that, as the injection pressure increases, the injection amount in the diesel and DME engines for the injection period is optimized in the laboratory, and incomplete combustion is less likely to occur.  2. HC emission concentration in diesel and DME engines depending on the injection pressure and pilot injection period Figure 9 shows the HC emission with pressure during high-pressure pilot and main injections. HC was reduced by more than 50% in this study as compared to those in other studies (injection pressure 20-35 MPa) [24,26] for high-pressure pilot injection. HC was generated in the combustion chamber wall or in a state where the air-fuel ratio was rich, and it was generated in a state where the fuel was vaporized and not oxidized. Moreover, DME, as an oxygenated fuel, generated almost no HC, and generated higher fuel density per unit volume. Furthermore, unlike the emission characteristics of NO x , the emission rate of HC was reduced by the increase of pressure. It could be expected that, as the injection pressure increases, the injection amount in the diesel and DME engines for the injection period is optimized in the laboratory, and incomplete combustion is less likely to occur.
3. CO emission concentration in diesel and DME engines depending on the injection pressure and pilot injection period Figure 10 shows the CO emission rate as a function of pressure when performing high-pressure pilot and main injections. CO is generally generated in an incomplete combustion environment because of lack of oxygen in an excess fuel state, which was not a problem because DME is an oxygenated fuel, and compression ignition engines are not a big problem because they operate in a lean air-fuel ratio region. In this study, the CO emission rate was lower than that of the diesel engine.

CO emission concentration in diesel and DME engines depending on the injection pressure and
pilot injection period Figure 10 shows the CO emission rate as a function of pressure when performing high-pressure pilot and main injections. CO is generally generated in an incomplete combustion environment because of lack of oxygen in an excess fuel state, which was not a problem because DME is an oxygenated fuel, and compression ignition engines are not a big problem because they operate in a lean air-fuel ratio region. In this study, the CO emission rate was lower than that of the diesel engine.   3. CO emission concentration in diesel and DME engines depending on the injection pressure and pilot injection period Figure 10 shows the CO emission rate as a function of pressure when performing high-pressure pilot and main injections. CO is generally generated in an incomplete combustion environment because of lack of oxygen in an excess fuel state, which was not a problem because DME is an oxygenated fuel, and compression ignition engines are not a big problem because they operate in a lean air-fuel ratio region. In this study, the CO emission rate was lower than that of the diesel engine.

Effects of Pilot Injection Rate on Exhaust Gas According to Injection Time in DME Engines
In this study, the emission characteristics of HC, NO x , CO, and carbon dioxide (CO 2 ) were analyzed for the engine performance tests. DME showed unleaded combustion and no special combustion method was required for exhaust gas reduction. However, NO x and HC vary depending on the operating conditions and fuel characteristics; thus, the reduction rate was investigated for each pilot injection rate, injection timing, needle lift, and injection timing in this experiment. 1. NO x emission characteristics of DME engines depending on the injection pressure and pilot injection period Figure 11 shows the effect of pilot injection period on NO x emissions with injection timing; other studies [35,39,46] have shown that the NO x emission rate of DME as a diesel substitute fuel decreases as it approaches the TDC. Therefore, in this experiment, the pilot injection rate was examined from 0% to 30% of the total injection amount with the injection pressure of 50 MPa and the injection timing at the lowest was 10 • BTDC and at the highest was 22 • BTDC. As a result, NO x decreased as the pilot injection rate increased. This was because the pilot injection before the main injections slowed the diffusion combustion because of the injection period delay. Furthermore, it reduced the combustion chamber temperature and average effective pressure because the injection delay decreased, thereby reducing the emission of NO x generated at high temperatures. NO x was mostly composed of nitrogen monoxide (NO) and nitrogen dioxide (NO 2 ), which were produced by the combination of nitrogen molecules in the air and oxygen at high temperatures. NO x in the compression ignition engine were divided into thermal NO and fuel NO, oxidized by the nitrogen contained in the fuel, most of which was thermal NO. The generation mechanism was based on the Zeldovich mechanism and had a mechanism (N 2 + O → NO + N) in which NO molecules dissociated during combustion in an excess air atmosphere to produce NO. The generation of NO by this dissociation was a strong endothermic reaction, which exponentially increased at high temperatures and was discharged during the expansion stroke. That is, the concentration of NO was greatly dominated by the combustion temperature, and the reaction time and oxygen concentration were also important factors.

Effects of Pilot Injection Rate on Exhaust Gas According to Injection Time in DME Engines
In this study, the emission characteristics of HC, NOx, CO, and carbon dioxide (CO2) were analyzed for the engine performance tests. DME showed unleaded combustion and no special combustion method was required for exhaust gas reduction. However, NOx and HC vary depending on the operating conditions and fuel characteristics; thus, the reduction rate was investigated for each pilot injection rate, injection timing, needle lift, and injection timing in this experiment.
1. NOx emission characteristics of DME engines depending on the injection pressure and pilot injection period Figure 11 shows the effect of pilot injection period on NOx emissions with injection timing; other studies [35,39,46] have shown that the NOx emission rate of DME as a diesel substitute fuel decreases as it approaches the TDC. Therefore, in this experiment, the pilot injection rate was examined from 0% to 30% of the total injection amount with the injection pressure of 50 MPa and the injection timing at the lowest was 10° BTDC and at the highest was 22° BTDC. As a result, NOx decreased as the pilot injection rate increased. This was because the pilot injection before the main injections slowed the diffusion combustion because of the injection period delay. Furthermore, it reduced the combustion chamber temperature and average effective pressure because the injection delay decreased, thereby reducing the emission of NOx generated at high temperatures. NOx was mostly composed of nitrogen monoxide (NO) and nitrogen dioxide (NO2), which were produced by the combination of nitrogen molecules in the air and oxygen at high temperatures. NOx in the compression ignition engine were divided into thermal NO and fuel NO, oxidized by the nitrogen contained in the fuel, most of which was thermal NO. The generation mechanism was based on the Zeldovich mechanism and had a mechanism (N2 + O → NO + N) in which NO molecules dissociated during combustion in an excess air atmosphere to produce NO. The generation of NO by this dissociation was a strong endothermic reaction, which exponentially increased at high temperatures and was discharged during the expansion stroke. That is, the concentration of NO was greatly dominated by the combustion temperature, and the reaction time and oxygen concentration were also important factors. Pilot 0% Pilot 10% Pilot 20% Pilot 30% Figure 11. NO x emission characteristics of DME engines with the pilot injection duration rate as a function of injection timing.
2. HC emission characteristics of DME engines depending on the injection timing and pilot injection rate Figure 12 shows the effect of pilot injection rate on HC generation with injection timing. The comparison in Figure 12 shows that NO x and HC were opposite to each other. Furthermore, the pilot injection quantity increased and, as the TDC was approached, the HC emission amount increased [47,49]. This was because, when injecting more than 10% of the pilot injection, the injection period was delayed, and the amount of HC emission increased because of unstable combustion.
Moreover, HC were generated by unstable combustion of fuel, generated because of lack of oxygen in a state where the combustion chamber wall surface or air-fuel ratio was rich, and discharged without being oxidized. It also occurred when the fuel was evaporated during fueling or when the fuel was burning unsafely, similar to CO. The ignition of the fuel sprayed from the fuel injection nozzle in the combustion process of the compression ignition engine occurred at a point slightly leaner than the stoichiometric air-fuel ratio (λ = 1), and spontaneous ignition could not occur outside the lean burn limit (A/F = 40) [49]. In such ultra-lean region, unreacted fuel did not end well, producing unburned fuel (HC). The concentration of HC generated in the ultra-lean region was significantly affected by the amount of fuel injected during the ignition delay period and the mixing ratio of the injected fuel with air. That is, a large amount of fuel injected during the ignition delay period or a delay in the ignition delay period might cause an increase in the amount of generated HC. Moreover, when the injected fuel became rich, there could be two scenarios: (1) the fuel remaining in the sack volume or sack hole was injected at a low speed in the latter stage of the combustion process; and (2) when the fuel was excessively supplied to the combustion chamber from the beginning. In particular, the sack volume and the volume of HC tended to be almost proportional.
2. HC emission characteristics of DME engines depending on the injection timing and pilot injection rate Figure 12 shows the effect of pilot injection rate on HC generation with injection timing. The comparison in Figure 12 shows that NOx and HC were opposite to each other. Furthermore, the pilot injection quantity increased and, as the TDC was approached, the HC emission amount increased [47,49]. This was because, when injecting more than 10% of the pilot injection, the injection period was delayed, and the amount of HC emission increased because of unstable combustion. Moreover, HC were generated by unstable combustion of fuel, generated because of lack of oxygen in a state where the combustion chamber wall surface or air-fuel ratio was rich, and discharged without being oxidized. It also occurred when the fuel was evaporated during fueling or when the fuel was burning unsafely, similar to CO. The ignition of the fuel sprayed from the fuel injection nozzle in the combustion process of the compression ignition engine occurred at a point slightly leaner than the stoichiometric air-fuel ratio (λ = 1), and spontaneous ignition could not occur outside the lean burn limit (A/F = 40) [49]. In such ultra-lean region, unreacted fuel did not end well, producing unburned fuel (HC). The concentration of HC generated in the ultra-lean region was significantly affected by the amount of fuel injected during the ignition delay period and the mixing ratio of the injected fuel with air. That is, a large amount of fuel injected during the ignition delay period or a delay in the ignition delay period might cause an increase in the amount of generated HC. Moreover, when the injected fuel became rich, there could be two scenarios: (1) the fuel remaining in the sack volume or sack hole was injected at a low speed in the latter stage of the combustion process; and (2) when the fuel was excessively supplied to the combustion chamber from the beginning. In particular, the sack volume and the volume of HC tended to be almost proportional. 3. CO emission characteristics of DME engines depending on the injection timing and pilot injection rate Figure 13 shows the CO emission from the injection timing and the pilot injection amount. It also shows that the amount of CO decreased with the increasing injection delay and pilot injection amount. CO is a colorless, odorless, toxic gas that was generated in the engine when operating at rich air-fuel ratios. When there was not enough oxygen to convert all the carbon in the HC-based fuel to CO 2 , some of the fuel would not burn and remained as CO. Maximum emission of CO occurred when the engine's fuel was operating in a rich state, that is, at start-up or under accelerated load. Even when the intake fuel-air mixture was at the theoretical mixture ratio or the lean mixture ratio, CO was also generated in the engine. The fuel-air mixture was poor or there was a local rich mixture, resulting in a knock in incomplete combustion.
air-fuel ratios. When there was not enough oxygen to convert all the carbon in the HC-based fuel to CO2, some of the fuel would not burn and remained as CO. Maximum emission of CO occurred when the engine's fuel was operating in a rich state, that is, at start-up or under accelerated load. Even when the intake fuel-air mixture was at the theoretical mixture ratio or the lean mixture ratio, CO was also generated in the engine. The fuel-air mixture was poor or there was a local rich mixture, resulting in a knock in incomplete combustion. 4. CO2 emission characteristics of DME engines depending on the injection time and pilot injection rate Figure 14 shows CO2 emission characteristics of DME engines according to the injection timing and pilot injection rate. CO2 was checked as it is identified as an emission causing global warming, and, similar to the CO generation shown in Figure 13, from these results, the reduction of CO2 through DME engine also showed a tendency to decrease as the pilot injection amount increased [39]. Pilot 0% Pilot 10% Pilot 20% Pilot 30% Figure 13. CO emission characteristics of DME engines with the pilot injection duration rate as a function of injection timing. 4. CO 2 emission characteristics of DME engines depending on the injection time and pilot injection rate Figure 14 shows CO 2 emission characteristics of DME engines according to the injection timing and pilot injection rate. CO 2 was checked as it is identified as an emission causing global warming, and, similar to the CO generation shown in Figure 13, from these results, the reduction of CO 2 through DME engine also showed a tendency to decrease as the pilot injection amount increased [39].  Figures 15-17 show the exhaust gas rates (HC, NO x , and CO) for the engine performance tests. In this experiment, the exhaust gas characteristics were examined by applying the DLL-AP type needle lift of 0.14, 0.21, 0.24, and 0.36 mm nozzle (DL-AP type), which was one of the injector improvements, different from the examination result by the condition change of the ECU program.

Effects of Needle Lift on Exhaust Gas According to Injection Timing in DME Engines
1. NO x emission characteristics of DME engines depending on the injection timing and needle lift Figure 15 shows the effect of needle lift on NO x emissions with injection timing at a loading rate of 75% at an injection pressure of 50 MPa. In this case, the effect of the needle lift on the NO x emission was found to increase as the injection timing was delayed and decreased as the injection angle was advanced. Further, the needle lifts showed low values at 0.36 and 0.14 mm, while at 0.21 and 0.24 mm showed an increasing trend. First, the increase in the injection timing was caused by the temperature rise of the combustion chamber because of diffusion combustion from the delay of the injection period. The decrease of 0.14 and 0.36 mm of needle lift at 22 • BTDC caused incomplete combustion at 0.14 mm due to the long supply period because of the common rail characteristics for the required fuel supply when injected with the required injection period. 2. HC emission characteristics of DME engines depending on the injection timing and needle lift Figure 16 shows the effect of needle lift on HC emissions with injection timing at a loading rate of 75% at an injection pressure of 50 MPa. In this case, the HC emission fluctuation was not large with different injection timing delays. The 0.36 mm needle lift showed a HC emission concentration lower than 0.21 mm, suggesting that the DME supply per unit time was oversupplied to the engine. Thus, the needle lift of 0.21 mm was useful for engine gas; however, it was not good in terms of exhaust gas.  2. HC emission characteristics of DME engines depending on the injection timing and needle lift Figure 16 shows the effect of needle lift on HC emissions with injection timing at a loading rate of 75% at an injection pressure of 50 MPa. In this case, the HC emission fluctuation was not large with different injection timing delays. The 0.36 mm needle lift showed a HC emission concentration lower than 0.21 mm, suggesting that the DME supply per unit time was oversupplied to the engine. Thus, the needle lift of 0.21 mm was useful for engine gas; however, it was not good in terms of exhaust gas. Figure 16 shows the effect of needle lift on HC emissions with injection timing at a loading rate of 75% at an injection pressure of 50 MPa. In this case, the HC emission fluctuation was not large with different injection timing delays. The 0.36 mm needle lift showed a HC emission concentration lower than 0.21 mm, suggesting that the DME supply per unit time was oversupplied to the engine. Thus, the needle lift of 0.21 mm was useful for engine gas; however, it was not good in terms of exhaust gas.  3. CO emission characteristics of DME engines depending on the injection timing and needle lift Figure 17 shows the effect of needle lift on CO emissions with injection timing at a loading rate of 75% at an injection pressure of 50 MPa. The CO emission slightly fluctuated at 0.36 mm because of the injection timing delay; however, there was no large fluctuation. This was expected to be incomplete combustion because of the longer injection period caused by the injection timing delay, as in the case of HC. 3. CO emission characteristics of DME engines depending on the injection timing and needle lift Figure 17 shows the effect of needle lift on CO emissions with injection timing at a loading rate of 75% at an injection pressure of 50 MPa. The CO emission slightly fluctuated at 0.36 mm because of the injection timing delay; however, there was no large fluctuation. This was expected to be incomplete combustion because of the longer injection period caused by the injection timing delay, as in the case of HC.

Conclusions
Based on the results of this study on diesel and DME engines with common rail conditions applied, we examined the engine performance changes and emission characteristics of exhaust gases

Conclusions
Based on the results of this study on diesel and DME engines with common rail conditions applied, we examined the engine performance changes and emission characteristics of exhaust gases with varying injection pressure and injection rate. The emission characteristics of NO x , HC, and CO were investigated for different injection pressures of the pilot injection. The exhaust gas emission characteristics for the pilot injection period and needle lift were investigated as follows: 1.
For DME engines, the effect of fuel injection pressure on HC emissions was not significant at low speeds; however, at high speeds, higher fuel injection pressures caused lower HC emissions.
To optimize the DME engine, it is necessary to optimize the fuel injection pressure and injection amount depending on the engine operating conditions. 2.
The emission characteristics of CO were similar to those of HC. The CO emission concentration of DME engines was slightly lower than that of diesel engines. The fuel injection pressure did not have a significant effect at all operating conditions. This was because the combustion characteristics of the oxygenate fuel of DME described in HC. As the liquid fuel DME, which is a compressible oxygen fuel, was injected at high pressure, the mixing ratio with air was improved in the injection characteristic as compared to that with diesel. Further, it was expected to enhance into a diffusive combustion state.

3.
The HC emission can be extremely low, as it was generated by oxygen deficiency in a state where the combustion chamber wall surface or the air-fuel ratio was rich, and because DME was generated in a state where the fuel was not vaporized and the oxidized fuel was an oxygen fuel. Moreover, as the injection pressure increased, the emission rate of NO x reduced. This was because the injection rate in diesel and DME engines was optimized according to the injection period, and incomplete combustion is not expected. 4.
The NO x emission was reduced at the injection pressure of 50 MPa, with pilot injection rate increasing from 0% to 30% of the total injected amount, and the minimum and maximum injection timing being 10 • and 22 • BTDC, respectively. The NO x emission decreased with the increasing pilot injection rate because the pilot injection before the main injection delays the rapid diffusion combustion. This was because the injection period delay lowered the combustion chamber temperature; and it was considered that NO x emissions generated at a high temperature were reduced because the average effective pressure caused by the injection delay decreased.

5.
It can be seen from this experiment that NO x and HC were opposite to each other. Further, as the pilot injection amount increased, or as it got closer to TDC, the HC emission amount increased. When pilot injection was applied in this study, it was expected that the injection period would be delayed if more than 10% was applied, so that the HC emissions would increase because of unstable combustion. 6.
The increase in injection timing was caused by the delay in injection period, occurring with the temperature rise in the combustion chamber because of diffusive combustion. The decrease of 0.14 mm and 0.36 mm of needle lift at 22 • BTDC was caused by incomplete combustion because of the longer supply period of the characteristics of common rail for supplying the required fuel when injected with the required injection period at 0.14 mm.
Author Contributions: S.Y. and C.L. conceived and designed the experiments; S.Y. performed the experiments; C.L. analyzed the data; S.Y. contributed reagents/materials/analysis tools; C.L. wrote the paper.
Funding: This study was supported by the development of key fusion technology of the industry-academia research cooperation cluster support project.