The Effect of Combustion Phase According to the Premixed Ethanol Ratio Based on the Same Total Lower Heating Value on the Formation and Oxidation of Exhaust Emissions in a Reactivity-Controlled Compression Ignition Engine

Abstract

A compression ignition engine generates power by using the auto-ignition characteristics of fuel injected into the cylinder. Although it has high fuel efficiency, it discharges a lot of exhaust emissions such as NO_X and PM. Therefore, there is much ongoing research aiming to reduce the exhaust emissions by using the technologies applied in this regard, such as PCCI, HCCI, etc. However, these methods still discharge large exhaust emissions. The RCCI method, which combines the spark ignition method and compression ignition method, is attracting attention. So, in this work, the objective of this study is to numerically investigate the effect of combustion phase according to the premixed ethanol ratio based on the same total heating value in-cylinder by changing the initial air composition on the formation and oxidation of exhaust emissions in the RCCI engine. The heating value of the premixed ethanol ratio varied from 0% to 40% based on the same total lower heating value in-cylinder in steps of 10%. It was assumed that the ethanol introduced into the cylinder through the premixing chamber was evaporated, and the initial air composition in the cylinder was changed and set. It was revealed that when the premixed ratio based on the same total lower heating value was increased, the introduced fuel amount into the crevice volume with advancing the start of energizing timing was decreased, which increased the peak cylinder pressure. In addition, the ignition delay was also longer due to the low cylinder temperature by the evaporation latent heat of the ethanol, which reduced the compression loss, so the IMEP value was increased. The rich equivalence ratio had a narrow distribution in the cylinder, which caused a reduction in cylinder temperature, so the NO formation amount was reduced. The ISCO value increased the increase in premixed ethanol ratio based on the same total lower heating value in-cylinder because the flame propagation of ethanol by combustion of diesel did not work well, and the CO formed by combustion was slowly oxidized due to the cylinder’s low temperature as a result of the evaporation latent heat of ethanol. From these results, the optimal operating conditions for simultaneously reducing the exhaust emissions and improving the combustion performance were judged such that the start of energizing timing was BTDC 23 deg, and the premixed ethanol ratio based on the same total lower heating value in-cylinder was 40%.

1. Introduction

A compression ignition (CI) engine generates power by using the auto-ignition characteristics of fuel injected into the cylinder. The auto-ignition characteristics of fuel and the direct injection of fuel into the cylinder make for high fuel efficiency. However, the fuel directly injected into the cylinder forms a rich air–fuel mixture region, inducing incomplete combustion, which generates matter harmful to humans, such as PM, CO, and soot. In addition, when combustion occurs in the rich fuel region, the combustion temperature is raised to more than 2000 K, and it forms several environmentally polluting substances, such as NO_X, which is formed at high temperatures [1].

As mentioned above, although the CI engine has high fuel efficiency, many exhaust emissions are discharged. So, research to satisfy strict regulations on vehicle exhaust emissions are actively underway. To solve these problems, there is a method wherein the rich air–fuel mixture region is reduced and made into a homogeneous air–fuel mixture region. Representatively, it is a method of promoting injected fuel evaporation performance by increasing the fuel injection pressure to improve the fuel atomization. Emiroglu [2] proceeded to investigate the effect of injection pressure on the characteristics of combustion and exhaust emission powered by a butanol–diesel blend and presented interesting results. When the fuel injection pressure increased, the atomization performance of the fuel improved; it induced a homogeneous air–fuel mixture, and the exhaust emissions by incomplete combustion were reduced. However, the injection pressure is already high, and there is a technical limitation to increasing it beyond that. Lee et al. [3] reported that spray targeting, in which the inclined spray angle and the injection timing are changed, induced homogeneous air–fuel mixture distributions, which could reduce exhaust emissions by incomplete combustion.

In another way, there is low-temperature combustion (LTC), a method of reducing exhaust emissions by reducing the combustion temperature. To reduce the combustion temperature, the method of inducing a homogeneous air–fuel mixture by advancing the fuel injection timing is used to locally avoid the rich air–fuel mixture in the CI engine. Typical methods are premixed charge compression ignition (PCCI) and homogeneous charge compression ignition (HCCI). Rohani et al. [4] announced that PCCI combustion, which is generated by incomplete combustion, discharges a few exhaust emissions due to various auto-ignition points of the air–fuel mixture in a whole cylinder. Although the PCCI method secures a relatively greater time for air and fuel mixing than the general CI method, it still discharges a lot of exhaust emissions because of the insufficient time to mix air and fuel [5]. So, Cha et al. [6] and Min et al. [7] conducted studies to simultaneously reduce the exhaust emissions by using a simulated EGR and varying the start of energizing timing under PCCI mode. It was found that the emissions of NO_X and soot were simultaneously reduced under a certain start of energizing timing. However, the IMEP value was decreased because a part of the injected fuel entered into the crevice volume, and it did not burn. To supplement the demerits of the PCCI method, the HCCI method uses a more homogeneous air–fuel mixture. The HCCI method uses low-temperature reactions within a homogeneous mixture in the cylinder [8]. The HCCI method has a serious weakness, which is that it is difficult to control the auto-ignition timing under high-load operation because the in-cylinder temperature is raised by the combustion, which induces the early start of combustion [9,10].

The reactivity-controlled compression ignition (RCCI) method is used to improve the defects of the PCCI method and HCCI method. In general, the RCCI method uses two fuels to make the homogeneous air–fuel mixture and to control the auto-ignition timing. The first fuel has a high octane number and is injected into the intake port. The fuel first injected into the intake port homogeneously mixes with air and is introduced into the cylinder. The second fuel has a high cetane number and is directly injected into the cylinder for ignition [11,12,13]. As is well known, the RCCI method forms a small NO_X emission because the homogeneous air–fuel mixture by high-octane fuel induces the low-temperature reaction. In addition, the ignition timing can be easily controlled because the ignition timing is determined by the auto-ignition of fuel directly injected into the cylinder. However, Singh et al. [14] reported that substances resulting from incomplete combustion such as carbon monoxide (CO) and hydrocarbon (HC) are formed at a low combustion temperature. To solve these problems, Jo et al. [15] conducted an experimental study to investigate the effect of the ethanol ratio on combustion improvement and emission reduction by using the RCCI method. When the ethanol ratio in total heating value increased, although the peak cylinder pressure was decreased, the IMEP value was maintained because the negative work was reduced by lengthening the ignition delay period. Furthermore, the NO_X emission was reduced; however, the substances resulting from incomplete combustion, such as HC and CO, were increased due to the low combustion temperature. To improve the combustion and to reduce the exhaust emissions, the comparison and analysis of distributions such as equivalence ratio and oxidation/formation of exhaust emissions are very important in RCCI engines because they can be modified to have a more homogeneous air–fuel mixture by identifying a rich or lean mixture region through the comparison and analysis of equivalence ratio distribution characteristics. Moreover, it is possible to build a database to promote the oxidation/formation of exhaust emissions through the comparison and analysis of exhaust emission distributions.

Although the experimental research has high confidence in the results, it is difficult to observe the formation and oxidation of exhaust emissions and their distributions. However, in numerical research, the detailed formation and oxidation processes of exhaust emissions and their distributions according to the premixed ethanol ratio can be confirmed. Therefore, this study was carried out to numerically investigate the effect of the combustion phase according to the premixed ethanol ratio based on the same total lower heating value in-cylinder by changing the initial air composition in the cylinder on the formation and oxidation of exhaust emissions in the RCCI engine. At the same time, peak cylinder pressure, IMEP, ISNO, and ISCO were compared and analyzed under the various premixed ethanol ratios, and the formation and oxidation of exhaust emissions were analyzed through the distributions of exhaust emission, and cylinder temperature.

2. Experimental and Numerical Descriptions

2.1. Experimental Descriptions

Experimental data for validation of the numerical analysis reliability were obtained from a previous study [15]. The schematics of RCCI engine experimental setup and the specifications of the engine are shown in Figure 1 and

Table 1. Detailed specifications of test engine.

Item		Specification
Engine	Displacement [cc]	498
	Bore × Stroke [mm]	83 × 92
	Compression ratio [-]	17.7
Injector	Number of nozzle holes [ea]	5
	Hole diameter [mm]	0.168
	Inclined spray angle [deg]	154

[15]. The system of fuel injection has been divided into two parts to implement the RCCI combustion. First, to supply high-octane fuel in the cylinder, ethanol was injected using a gasoline direct injection (GDI) system. The injected ethanol in the premixing chamber was homogeneously mixed with air and flowed into the engine cylinder. Next, to ignite and combust the supplied ethanol in the cylinder, diesel, which is a high cetane fuel, was directly injected into the cylinder using a common rail direct injection (CRDI) system. Therefore, to implement the RCCI combustion, there were two fuel injection systems used in this study. The GDI injector for ethanol injection was installed in the premixing chamber at an angle of approximately 45°, and the diesel injector was placed in the center of the cylinder head. The diesel injector has a 5-hole nozzle and 0.168 mm hole size.

The GDI system for injecting the ethanol fuel used a pneumatic pump (Haskel, DSF-300, Burbank, CA, USA) capable of producing up to 3000 bar, and the CRDI system for injecting the diesel fuel used a CRDI pump, and it was controlled by an injector driver (National Instruments, NI-9751, Sydney, NSW, Australia). To measure the cylinder pressure, a piezo-electric pressure sensor (Kistler, 6052C, Winterthur, Switzerland) was employed, and it was installed in place of the glow plug. A rotary encoder was used to measure the crank angle, and the signal of the cam was measured by a proximity sensor. The exhaust gas from the test engine was measured by a portable emissions measurement system (PEMS). The PEMS can measure NO_X and CO through the chemiluminescence detector (CLD) method and the non-dispersive infrared (NDIR) method, respectively. The detailed measurement methods and specifications of PEMS are listed in

Table 2. Detailed measurement methods and specifications of PEMS.

Emission	Method	Range	Span
NOX	CLD (CLA-150)	0~5000 ppm	NO/N2-bal 1494 ppm
CO	NDIR (AIA-110)	0~3000 ppm	CO/N2-bal 7404 ppm

. The measured NO and CO values were changed to indicated specific NO and CO values (ISNO and ISCO) using Equation (1) below.

$$ISexhaustemissions={\displaystyle \frac{m_{exhaustemission}·n_{cycle}}{P_i}}$$

(1)

where $P_i$ means indicated power, and $m_{exhaustemission}$ and $n_{cycle}$ mean the mass of exhaust emission and the number of cycles.

The flow rate of the intake air and the start of energizing timing were fixed at 366 L/min and BTDC 12 deg, respectively. The engine operating speed was maintained at 1800 rpm. The injection pressures of ethanol and diesel were 10 MPa and 100 MPa, respectively. In this work, the ratio of diesel injection amount in the cylinder to ethanol injection amount in the premixing chamber was changed. As listed in

Table 3. Properties of test fuels [16].

Item	Diesel	Ethanol
LHV; Lower Heating Value [MJ/kg]	42.5	26.8
Latent heat of Evaporation [kJ/kg]	250	846
Density@20 °C [kg/m3]	838.2	789.4
Carbon Content [% mass]	86.7	52.14
Hydrogen content [% mass]	12.71	13.13
Sulfur content [% mass]	0.041	-
Oxygen content [% mass]	-	34.73
Flash point [°C]	67	13
Kinematic viscosity@40 °C [mm2/s]	2.8271	1.056
Typical formula	C14.09H24.78	C2H6O
Cetane number	42.6	8.5
Lubricity, HFRR@60 °C [μm]	Max. 520	Max. 605

[16], the lower heating values (LHVs) of ethanol and diesel ethanol are 26.8 MJ/L and 42.5 MJ/L, respectively. In all conditions, the total heating value was fixed based on the heating value of diesel 14 mg, which is 595 J. The ethanol injection mass was changed from 0% to 40% of the same total lower heating value in steps of 10%. The detailed experimental conditions and the LHV ratio of diesel and ethanol are listed in

Table 4. Initial air composition in the cylinder according to the heating value ratio of diesel and ethanol.

Fuel Injection Ratio [Main/Premixed]	Fuel Amount [mg]	Heating Value [J]	O2 Mass Fraction [-]	N2 Mass Fraction [-]	Ethanol Mass Fraction [-]
D100/E0	14/0	595/0	0.23200	0.76800	0.00000
D90/E10	12.6/2.2	535.5/59.5	0.23081	0.76407	0.00512
D80/E20	11.2/4.4	476/119	0.22984	0.76084	0.00932
D70/E30	9.8/6.7	416.5/178.5	0.22876	0.75273	0.01396
D60/E40	8.4/8.9	357/238	0.22781	0.75413	0.01806

.

2.2. Numerical Descriptions

To confirm the formation and oxidation of exhaust emissions and their distributions in the cylinder, a numerical analysis was conducted in this study.

As listed in

Table 5. Sub-model for numerical analysis [17,18,19].

Phenomenon	Model
Turbulence	k-zeta-f
Break-up	Wave
Evaporation	Dukowicz
Wall interaction	Mundo Tropea Sommerfeld
Combustion	ECFM-3Z
NO	Extended Zel’dovich
Soot	Kennedy-Hiroyasu-Magnussen

, the sub-models were employed in the numerical analysis to express the physical and chemical phenomena. The sub-models were applied from the reference of the library in AVL Fire [17,18,19]. In this work, the species C₂H₆OH was added, corresponding to fuel for RCCI combustion. Its governing equation is given by the following Equation (2).

$${\displaystyle \frac{𝜕\overline{\rho }{\overline{Y}}_x}{𝜕t}}+{\displaystyle \frac{𝜕\overline{\rho }{\overline{u}}_i{\overline{Y}}_x}{𝜕x_i}}={\displaystyle \frac{𝜕}{𝜕x_i}}\left(\left({\displaystyle \frac{\mu }{Sc}}+{\displaystyle \frac{{\mu }_t}{{Sc}_t}}\right){\displaystyle \frac{𝜕{\overline{Y}}_x}{𝜕x_i}}\right)+{\overline{\dot{\omega }}}_x$$

(2)

where the average mass fraction of chemical species $x$ is represented by ${\overline{Y}}_x$. The viscosity coefficient and the Schmidt number are indicated by $\mu$ and $Sc$, respectively. The subscript t means turbulent flow, and without subscript, it means laminar flow. ${\overline{\dot{\omega }}}_x$ is the generation term that causes the evaporation and combustion of the fuel.

The mixing rate of the fuel region (F) and the air region (A) is determined by the turbulence model-specific time scale (τ_m) of the following Equation (3).

$${\tau }_m^{-1}={\beta }_m{\displaystyle \frac{\epsilon }{k}}$$

(3)

where ε and κ represent the turbulent dissipation rate and the turbulent kinetic energy, respectively. β_m is a constant that determines the amount of fuel and air mixing.

To reduce the simulation time and improve efficiency, the mesh geometry was generated for only 1/5 of the whole engine geometry. Like the experiment, the introduced total heating value was constant at 595 J in the cylinder, which is based on the lower heating value of diesel 14 mg, and the heating value of premixed ethanol ratio was varied from 0% to 40% based on the same total lower heating value in-cylinder in steps of 10%. In numerical analysis, the fuel mass was set to 1/5 of the whole fuel mass. It was assumed that the ethanol introduced into the cylinder through the premixing chamber was evaporated, and the initial air composition in the cylinder was changed and set.

The start of energizing timing was changed from BTDC 12 deg to BTDC 30 deg. Diesel-D1 (C₁₃H₂₃) fuel and ethanol (C₂H₆OH) were applied from the reference of the library in AVL Fire [19]. The detailed numerical analysis conditions and the initial air composition in the cylinder according to the lower heating value ratio of diesel and ethanol are listed in Table 4 and

Table 6. Detailed experimental and numerical analysis conditions.

Contents	Experiment and Numerical Analysis
RPM	1800
Injection pressure [MPa]	Diesel: 100, Ethanol: 10
Total heating value [J]	595
Start of energizing timing [ATDC deg]	−30~−12
Heating value ratio of diesel and ethanol	D100/E0, D90/E10, D80/E20, D70/E30, D60/E40

. The required injection rate data for numerical analysis was measured by using Bosch’s suggestion [20]. As described above, the injection mass was varied from 8.4 mg to 14 mg due to the ethanol fuel addition, and the ethanol mass was also varied from 0 mg to 8.88 mg according to the premixed ethanol ratio based on the same total lower heating value in-cylinder, as shown in Figure 2.

3. Results and Discussion

3.1. Validation of Numerical Models

To secure the reliability of the used sub-model in this study, the numerical results according to the premixed ethanol ratio based on the same total LHV in-cylinder were compared with the experimental results as shown in Figure 3. Figure 3a shows the effect of the premixed ethanol ratio based on the same total LHV in-cylinder on the cylinder pressure and the rate of heat release (ROHR). It was confirmed that the cylinder pressure results of numerical analysis agreed well with the experimental results, and the error rate of peak cylinder pressure value was a maximum of 3.79% and a minimum of 1.43%. The validation results of exhaust emission characteristics are shown in Figure 3b. The experimental and numerical results for the indicated specific nitric oxide (ISNO) and the indicated specific carbon monoxide (ISCO) matched well with the experimental results by error rate within 2.07% and within 1.39%, respectively. Thus, it can be said that the applied sub-models for combustion and exhaust emission in numerical analysis were secured reliability.

3.2. Peak Cylinder Pressure and IMEP Characteristics

Figure 4 shows the effect of premixed ethanol ratio based on the same total LHV in-cylinder on the peak cylinder pressure. It was found that as the start of emerging timing was advanced, the peak cylinder pressure was increased because, as shown in Figure 5, the ignition delay was longer due to the low temperature for ignition, which secured the time for mixing air and fuel. However, it was not the highest value, at BTDC 30 deg, because, as shown in Figure 6, the fuel injected too quickly did not flow into the cylinder bowl but into the crevice volume, and the combustion was not actively caused due to the small air in the crevice volume. In addition, when the premixed ethanol ratio based on the same total LHV in-cylinder was increased, the ignition delay became longer because the intake air temperature was decreased due to the latent heat of evaporation by the evaporation of the ethanol injected into the premixing chamber [15]. It was observed that when the premixed ethanol ratio based on the same total LHV in-cylinder was increased, a small amount of the injected diesel fuel was introduced into the crevice volume because the spray tip penetration was short due to the short injection duration.

As discussed by Maes et al. [21], it was expected that a large amount of the injected diesel fuel was not introduced into the crevice volume and it was introduced into the piston bowl by using things such as the controlled inclined spray angle, which improved the combustion performance.

However, when the premixed ethanol ratio based on the same total LHV in-cylinder was 40%, the peak cylinder pressure was the lowest because, as reported by Aronsson et al. [22], the intake air temperature was too reduced due to the latent heat of evaporation by the introduced ethanol in the cylinder, which deteriorated the combustion performance.

Figure 7 shows the effect of premixed ethanol ratio based on the same total LHV in-cylinder on the IMEP characteristics. When the premixed ethanol ratio based on total heating value was 40% under early injection conditions, although its conditions had the lowest peak cylinder pressure value, as shown in Figure 4, the IMEP value was the highest because the negative work, such as compression loss, was small due to the long ignition delay, and the positive work was increased by the long combustion duration, as shown in Figure 8.

Figure 8 shows the effect of premixed ethanol ratio based on the same total LHV in-cylinder on the combustion phase characteristics. The combustion phase was divided into CA 10, CA 50, and CA 90. CA 10, CA 50, and CA 90 were denoted as corresponding to 10%, 50%, and 90% accumulated ROHR. CA 10 and CA 90 mean the combustion start timing and the combustion end timing. As reported by Dhole [23], CA 50 is a critical parameter for RCCI combustion because advanced CA 50 causes rapid combustion, high heat release, and knocking. On the other hand, too retarded CA 50 causes many HC and CO emissions due to incomplete combustion. Under the early injection conditions, the premixed combustion proportion was increased because the time was secured to form the homogeneous air–fuel mixture due to the long ignition delay, so the combustion duration was shorter with advancing start of energizing timing. In the case of D60/E40, although the peak cylinder pressure was the lowest value, the IMEP value was the highest value under the early injection conditions because the combustion phase was placed after TDC, which decreased the compression loss.

On the other hand, the combustion duration was longer when delaying the start of energizing timing because the premixed combustion proportion was reduced due to the short ignition delay. Furthermore, it was expected that the NO emission was larger, and the CO emission was smaller when delaying the start of energizing timing and increasing the premixed ethanol ratio. The reason is that when the start of energizing timing was retarded, the cylinder temperature was high enough to ignite, so the ignition delay was short, which produced the long combustion duration. When the premixed ethanol ratio based on the same total LHV was increased, the flame propagation was slow with ethanol, which caused the long combustion duration [14].

3.3. ISNO and ISCO Characteristics

The effect of the premixed ethanol ratio based on the same total LHV on the ISNO characteristics under a different start of energizing timing is shown in Figure 9. The ISNO value was increased when advancing the start of energizing timing because, as described in 12, the premixed combustion proportion was increased due to the long ignition delay, which increased the cylinder temperature, as shown in Figure 10. As is well known, NO formation is a function of temperature, so a lot of NO was formed when advancing the increase in the cylinder temperature. Moreover, the difference in ISNO value was larger than the difference in cylinder temperature because when the start of energizing timing was advanced, the IMEP value was also decreased, as shown in Figure 11, which caused a large difference in ISNO value. The ISNO value was reduced by about −31% when increasing the premixed ethanol ratio based on the same total LHV in-cylinder in steps of 10% because the relative equivalence ratio was homogeneously distributed in the cylinder due to the homogeneously distributed ethanol, and when the ethanol injected into the premixing chamber for the homogeneous air–fuel mixture evaporated, the cylinder temperature was reduced due to the latent heat of vaporization by the ethanol evaporation, as shown in Figure 9.

Furthermore, as discussed by Einecke et al. [24], when the premixed ethanol ratio based on the same total LHV in-cylinder was increased, the rich equivalence ratio was distributed smaller, which did not actively cause the combustion. From these results, it was found that when the premixed ethanol ratio was increased, the formation amount of NO was decreased because the cylinder temperature was reduced due to the reduction in the intake air temperature by the latent heat of evaporation in ethanol, and the rich equivalence ratio was less distributed due to the small amount of diesel fuel.

These results are also able to confirm the distributions of NO mass fraction and cylinder temperature, as shown in Figure 11 and Figure 12. Figure 11 and Figure 12 show the effect of premixed ethanol ratio based on the same total LHV in-cylinder on the distributions of NO mass fraction and cylinder temperature under the different start of energizing timing, respectively. It was observed that the high-NO-mass-fraction region was widely distributed when advancing the start of energizing timing because the rich equivalence ratio was distributed, which formed a high-temperature region, as shown in Figure 12. It was known that when the premixed ethanol ratio was increased, the NO formation region was reduced because, as shown in Figure 6 and Figure 12, the cylinder temperature was decreased, and the rise in combustion temperature was suppressed due to the higher specific heat of ethanol (2.460 kJ/kg·K) than air (1.012 kJ/kg·K) [25].

Therefore, it is expected that the injection amount of diesel fuel is reduced when increasing the premixed ethanol ratio based on the same total LHV in-cylinder, which forms a small amount of NO.

The effect of premixed ethanol ratio based on the same total LHV in-cylinder on the ISCO characteristics under the different start of energizing timing is shown in Figure 13. When the premixed ethanol ratio was 0% and the start of energizing timing was advanced, although the ISCO slightly increased, it had a low value of less than 5 g/kWh. However, when the start of energizing timing was BTDC 30 deg, the ISCO was significantly increased because the injected fuel was introduced into the crevice volume, which had a small amount of air, and the CO oxidation deteriorated. Moreover, in the other case, which was when the premixed ethanol ratio was increased, the ISCO value was decreased until BTDC 24 deg and then increased again. The reason is that when the start of energizing timing was advanced until BTDC 24 deg, the cylinder temperature was increased proportionately to the rise in premixed combustion. However, when the start of energizing timing was advanced more, a part of the injected diesel fuel was introduced into the crevice volume, which formed a lot of CO, and the flame propagation by diesel fuel combustion did not work well [26], which caused the deterioration of the CO oxidation.

When the premixed ethanol ratio based on the same total LHV was increased in steps of 10%, the ISCO value was also increased by about 40% because, as described in Figure 10 and Figure 12, the cylinder temperature was decreased when increasing the premixed ethanol ratio due to the ethanol latent heat of evaporation; the oxidation of the CO formed by combustion was suppressed. In addition, as shown in Figure 14, although the high-CO-mass-fraction region was similarly distributed regardless of the premixed ethanol ratio at the timing of ATDC 10 deg, when the premixed ethanol ratio was low, CO was rapidly oxidized and disappeared.

However, when the premixed ethanol ratio was increased, the formed CO was not oxidized, and a high-CO-mass-fraction region was continuously observed due to the low combustion temperature. From these results, it was found that the amount of formed CO was similar regardless of the premixed ethanol ratio due to the injected diesel fuel combustion; however, when the premixed ethanol ratio was increased, the formed CO was not oxidized well due to the low combustion temperature.

3.4. Optimal Operating Conditions in RCCI Engine

To find the optimal operating conditions in the RCCI method, the comparison of ISNO and ISCO according to the start of energizing timing is shown in Figure 15. It was found that when the start of energizing timing was advanced, the ISNO value and ISCO value were simultaneously increased. A lot of the injected diesel fuel was introduced into the crevice volume, which formed a lot of CO due to the small amount of air, and the premixed combustion proportion was increased due to the long ignition delay, so the combustion temperature was increased, which formed a lot of NO. So, it is judged that the advanced start of energizing timing of BTDC 27 deg and BTDC 30 deg is not the optimal start of energizing timing. Moreover, when the start of energizing timing was retarded, at BTDC 12 deg, BTDC 15 deg, and BTDC 18 deg, although its conditions had a lower ISNO value than 5 g/kWh, the ISCO value was mostly high. The reason is that the combustion temperature was reduced because the premixed combustion proportion was decreased due to the short ignition delay, which caused the formation of a small amount of NO; however, the formed CO was not oxidized well due to the low combustion temperature. Therefore, it is judged that the optimal start of energizing timing is teng = BTDC 21 deg and teng = BTDC 24 deg because the ISNO value and the ISCO value are simultaneously low.

In order to find the exact optimal operating conditions, the comparison of ISNO and ISCO according to the premixed ethanol ratio based on the same total LHV in-cylinder is shown in Figure 16. It was observed that when the premixed ethanol ratio based on the same total LHV in-cylinder was increased, the ISCO value was increased, and when the premixed ethanol ratio was decreased, the ISNO value was increased. The reason is that when the premixed ethanol ratio was increased, the combustion temperature was decreased due to the vaporization latent heat of ethanol, which reduced the NO formation amount, and the oxidation of the formed CO was suppressed. When the optimal start of energizing timing found in Figure 15 was compared by projecting to Figure 16, the conditions of simultaneously low ISNO and ISCO values were that the premixed ethanol ratio based on the same total LHV in-cylinder was 40%.

As found in the optimal operating conditions above, although the simultaneous reduction in exhaust emissions is important, combustion performance is also very important. Therefore, the IMEP value was compared in terms of ISNO and ISCO to find the simultaneous reduction in exhaust emission and the improvement of combustion performance, as shown in Figure 17. Figure 17 shows the comparison of ISNO and ISCO according to the IMEP characteristics to find the optimal operating conditions in the RCCI method. It was found that where the high IMEP values were distributed, the ISNO value was high or an ISCO value was high because a place with high peak cylinder pressure due to the improved combustion performance had a high ISNO value due to the high combustion temperature, and a place with a low compression loss due to the long ignition delay had the high ISCO value due to the low combustion temperature. However, the optimal starts of energizing timing, at BTDC 21 deg and BTDC 24 deg, were found, as shown in Figure 10, with the optimal premixed ethanol ratio based on the same total LHV in-cylinder at 40%, as shown in Figure 16; compared with projecting to Figure 17, these conditions have a high IMEP value while satisfying the simultaneous reduction in exhaust emissions. Therefore, when the start of energizing timing was BTDC 21 deg and BTDC 24 deg and the premixed ethanol ratio based on the same total LHV in-cylinder was 40%, these conditions were judged as optimal operating conditions by simultaneously reducing exhaust emissions while improving combustion performance.

4. Conclusions

This study aimed to investigate the effect of premixed ethanol ratio based on the same total LHV in-cylinder on the simultaneous reduction in exhaust emissions and the improvement of combustion performance. The following conclusions could be obtained.

• When the premixed ethanol ratio based on the same total LHV in-cylinder was increased, the amount of fuel introduced into the crevice volume when advancing the start of energizing timing was decreased, which increased the peak cylinder pressure.
• The ignition delay was longer when increasing the premixed ethanol ratio due to the low cylinder temperature due to the evaporation latent heat of ethanol, which reduced the compression loss. For this reason, the IMEP value was increased.
• When the premixed ethanol ratio based on the same total LHV in-cylinder was increased, the rich equivalence ratio was less distributed in the cylinder, which caused less increase in the cylinder temperature. So, the NO formation amount was reduced.
• The ISCO value was increased when increasing the premixed ethanol ratio based on the same total LHV in-cylinder because the flame propagation of ethanol by combustion of diesel did not work well, and the CO formed by combustion was slowly oxidized due to the low cylinder temperature due to the evaporation latent heat of ethanol.
• From these results, the optimal operating conditions for simultaneously reducing the exhaust emissions and improving the combustion performance were judged such that the start of energizing timing was BTDC 23 deg, and the premixed ethanol ratio based on the same total LHV in-cylinder was 40%.