Computational Investigation of Combustion, Performance, and Emissions of a Diesel-Hydrogen Dual-Fuel Engine

Abstract

This paper aims to expose the effect of hydrogen on the combustion, performance, and emissions of a high-speed diesel engine. For this purpose, a three-dimensional dynamic simulation model was developed using a reasonable turbulence model, and a simplified reaction kinetic mechanism was chosen based on experimental data. The results show that in the hydrogen enrichment conditions, hydrogen causes complete combustion of diesel fuel and results in a 17.7% increase in work capacity. However, the increase in combustion temperature resulted in higher NOx emissions. In the hydrogen substitution condition, the combustion phases are significantly earlier with the increased hydrogen substitution ratio (HSR), which is not conducive to power output. However, when the HSR is 30%, the CO, soot, and THC reach near-zero emissions. The effect of the injection timing is also studied at an HSR of 90%. When delayed by 10°, IMEP improves by 3.4% compared with diesel mode and 2.4% compared with dual-fuel mode. The NOx is reduced by 53% compared with the original dual-fuel mode. This study provides theoretical guidance for the application of hydrogen in rail transportation.

1. Introduction

Adjusting the industrial and energy structures is inevitable to realize an emission peak and carbon neutrality [1,2]. The gaseous and particle emissions from internal combustion engines (ICEs) contribute an important proportion of total atmospheric pollutants [3,4]. Therefore, searching for low-carbon or zero-carbon fuels has become an important research direction for developing ICEs [5,6,7]. Hydrogen is regarded as the clean energy with the most development potential in the 21st century because of many advantages, such as diverse sources, being clean and low-carbon, being flexibile and efficient, and having various application scenarios [8,9,10]. Hydrogen energy has become the preferred direction for the new round of carbon emission reduction and carbon neutrality worldwide [11,12]. It has been incorporated into the energy strategy deployment by many countries [13,14]. From the strategic point of view of energy security and sustainable development, China has considered hydrogen energy a new strategic industry for development [15,16].

There are two main ways to utilize hydrogen energy: fuel cells [17] and ICEs [18]. Fuel cells have the advantages of high efficiency and zero-emissions, but they are technically complex, costly, and dependent on the construction of supporting systems [19]. The hydrogen-fueled ICEs can use industrial by-product hydrogen to convert energy by the combustion method to achieve similar thermal efficiency as fuel cells, which has the significant advantage of low cost [20]. Hydrogen-fueled ICEs retain the main structure and system of traditional ICEs [21,22,23]. Based on traditional ICEs, hydrogen-fueled ICEs can be realized by simply replacing the hydrogen supply and injection system, hydrogen-specific cold spark plugs, matching a new turbocharger, and adapting the lubrication and crankcase ventilation accordingly [24,25,26]. Therefore, hydrogen-fueled ICEs are an essential technology direction to promote the upgrading and transformation of various application fields of traditional ICEs, which help achieve peak emissions and carbon neutrality [27,28,29,30].

For the spark-ignited hydrogen-fueled ICEs, the higher laminar flame speed and larger dilute combustion limits of hydrogen allow the engine to operate over a wide range of equivalence ratios, in which the thermal efficiency typically equals or exceeds that of gasoline-fueled engines [31]. It has been shown that spark-ignited hydrogen-fueled ICEs can achieve near-zero NOx emissions under lean-burn operating conditions [32]. However, hydrogen-fueled ICEs face problems such as backfire, premature ignition, and detonation at high loads, which severely limit their rapid development [33]. The heat transfer loss through the wall increases rapidly as the equivalent ratio increases [34]. The problems of detonation and premature ignition are challenging for heavy engines, which has prompted research on hydrogen compression ignition (CI) engines [35]. Therefore, dual-fuel technology was developed for engines [36]. The diesel fuel is designed to assist the ignition of the hydrogen fuel, known as the diesel pilot ignition mode [37,38]. In fact, this concept has long been widely used in natural gas-fueled ICEs [39] and extensively used in the marine field [40,41]. Tripathi et al. [42] investigated the performance and emissions of a diesel-hydrogen dual-fuel engine using numerical simulations. The results showed that the combination of two injection strategies can simultaneously reduce NOx emissions and improve IMEP. Sharma et al. [43] investigated the performance and emissions of dual-fuel engines with different compression ratios and hydrogen fractions through a similar methodology. The results show that compression-ignition mode is not suitable for compression ratios less than 14.5. Köse et al. [44] experimentally investigated that hydrogen enrichment reduced pollutant emissions except for NOx and increased brake thermal efficiency (BTE) and exhaust temperature. When operated at 1750 rpm, 40.4% BTE was achieved at 2.5% hydrogen enrichment, while in diesel mode, the BTE was 33%. Ramsay et al. [45] studied the effect of the constant volume combustion phase on the performance and emissions of a dual-fuel engine under various load and hydrogen energy share conditions. The results demonstrated that this method could improve thermal efficiency with far lower carbon-based emissions under all conditions. Taghavifar et al. [46], through a 1-D model, parametrically investigated the effects of levels of diesel and hydrogen, compressor pressure ratio, and combustion duration on energy, exergy, and performance in a diesel-hydrogen dual-fuel engine. The results indicated that supercharging can significantly improve thermal efficiency and reduce fuel consumption. Wu et al. [47] optimized the operation parameters of a dual-fuel engine based on the Taguchi method. The results revealed that for NOx, using EGR technology reduces more than 60.5% at various loads, and BSFC can reduce it by 14.52%.

From the above literature, it is clear that hydrogen can significantly improve thermal efficiency and performance. It is essential to improve thermal efficiency and reduce carbon emissions for rail transportation. Therefore, in this paper, hydrogen enrichment and hydrogen substitution are investigated separately by using numerical simulation. In addition, injection timing studies are carried out to optimize performance and emissions. This paper explored the potential of hydrogen in ICEs for rail transportation and provided a theoretical basis for practical applications.

2. Materials and Methods

2.1. Numerical Methodology

In this work, the prototype engine is a high-speed diesel engine with a total displacement of 87.54 L. The cylinder bore, stroke, and compression ratio were 180 mm, 215 mm, and 17, respectively. The engine is designed to meet future emissions regulations by taking into account compactness, power-to-weight ratio, economy, and reliability. The CONVERGE code was applied to calculate the flow motion and combustion phenomena in the combustion chamber [48,49,50]. The SolidWorks software was used to establish the 3-D geometric model. The model (*.stl) was then imported into CONVERGE to calculate the combustion. To simulate the turbulence, spray, and combustion, the mathematical models adopted in CFD calculations are summarized in

Table 1. Mathematical models adopted in this research.

Region	Type
Turbulence	RNG $k-\epsilon$ model
Wall heat transfer	O’Rourke and Amsden model
Spray breakup	KH-RT model
Evaporation	Frossling model
Droplet collision	O’Rourke’s model
Combustion	SAGE
NOx formation	Extended Zel’dovich mechanism
Soot formation	Hiroyasu model

[51]. The dual-fuel reaction mechanism with 76 species and 464 reactions was chosen to simulate the combustion of the hydrogen and diesel mixture [52]. This dual-fuel reaction mechanism was coupled with the GRI3.0 mechanism, which can accurately simulate hydrogen combustion. This mechanism was used in the simulation of pure diesel and dual-fuel, which avoids the influence of mechanism differences on the results [53]. In addition, at 1800 rpm, the temperatures of the piston, cylinder wall, and cylinder head were set to 553, 433, and 523 K, respectively.

2.2. Model Validation

To reduce calculation time, only 1/8 of the domain was recorded and analyzed since the injection is located in the center of the chamber with eight nozzles. As illustrated in Figure 1a, the cylindrical domain was classified into angular sectors such that one injector falls at the center of each sector. Figure 1b shows the computational mesh at TDC, in which the adaptive mesh refinement and fixed embedding were activated to guarantee the calculation accuracy. Figure 2 shows the predicted pressure profile under different meshes. It can be seen from Figure 2 that the 4 mm basic grid can meet the calculation accuracy. As the dual-fuel mode is still in the development stage, only the diesel mode was tested. To validate the combustion and turbulence model, the model was verified at speeds of 600, 1400, and 1800 rpm. Figure 3 shows the comparison between experimental and simulated cylinder pressure. The calculation accuracy was high and met the engineering requirements for the next step of research. Since the reaction kinetics mechanism used includes hydrogen and diesel, the verified model was used for the next study. In addition, Tripathi et al. [42] was validated using the same turbulence model and mechanism in diesel-hydrogen conditions, which is another indication that the accuracy of the model in this paper can be studied in the next part [54].

2.3. Research Schemes

The present work aims to evaluate the potential of hydrogen in diesel engines and to explore the maximum hydrogen substitution ratio for the purpose of reducing carbon emissions. In all the simulations, the speed was kept at 1800 rpms, and the total ejected energy was same as with the 18-bar BMEP. In the hydrogen enrichment test, the source diesel was kept the same, and the hydrogen was the added energy. The hydrogen enrichment ratio ($HER$) was between 10 and 20%. In the hydrogen substitution test, the total energy was kept the same as with the origin diesel, and the substitution ratios ($HSR$) were 30, 60, and 90%, respectively. In the relative injection timing test, the high $HSR$ was tested with varing injections. The summary of the test conditions is listed in

Table 2. Summary of different calculated schemes.

Case Group	Hydrogen Energy Ratio/%	Relative Injection Timing/°CA
Hydrogen enrichment	0% 1	0
	10% 2	0
	20% 2	0
Hydrogen substitution	30%	0
	60%	0
	90%	0
Relative injection timing	0%	0
	90%	0
	90%	5
	90%	10
	90%	15

¹ pure diesel, ² $HER$.

.

The definition of $HER$ and $HSR$ is as follows:

$$HER=\frac{m_{{\mathrm{H}}_2}\cdot {\mathrm{LHV}}_{{\mathrm{H}}_2}}{m_{\mathrm{Diesel}}\cdot {\mathrm{LHV}}_{\mathrm{Diesel}}}$$

(1)

$$HSR=\frac{m_{{\mathrm{H}}_2}\cdot {\mathrm{LHV}}_{{\mathrm{H}}_2}}{m_{\mathrm{Diesel}}\cdot {\mathrm{LHV}}_{\mathrm{Diesel}}+m_{{\mathrm{H}}_2}\cdot {\mathrm{LHV}}_{{\mathrm{H}}_2}}$$

(2)

3. Results and Discussion

3.1. Effect of Hydrogen Enrichment

The in-cylinder pressure is an important parameter to deeply understand the combustion process and directly affects performance and emissions. The comparison of in-cylinder pressures is shown in Figure 4. In this section, D and H represent diesel and hydrogen, respectively. D100 means the diesel energy fraction is 100%, and H10 represents the hydrogen energy fraction at 10%. As shown in Figure 4, a larger $HER$ leads to a higher in-cylinder pressure [42]. The corresponding CA position (PFP_CA) is delayed with the increasing $HER$, as shown in

Table 3. Comparison of PFP, corresponding CA position, and IMEP at hydrogen additional enrichment.

Parameter	PFP (bar)	PFP_CA (°CA, aTDC)	IMEP (bar)
D100	21.75	7.26	21.53
D100 + H10	22.49	7.43	23.40
D100 + H20	23.51	7.73	25.33

. The IMEP of D100 + H10 and D100 + H20 increased by 8.7 and 17.7%, respectively. The temperature contour is shown in

Table 4. Contour of temperature distribution at hydrogen additional enrichment.

CAD	D100 (K)	D100 + H10 (K)	D100 + H20 (K)
−4 °CA aTDC
TDC
4 °CA aTDC
8 °CA aTDC
16 °CA aTDC
24 °CA aTDC

. It is obvious that the temperature increases with the $HER$. This is because diesel has a dominant effect on combustion, while hydrogen only promotes combustion. The comparison of HRR and combustion phases is shown in Figure 5. The profiles of HRR and combustion phases are similar to each other. The phases are delayed with increasing $HER$, but the level is minor. This is because the overall energy of the additional hydrogen enrichment condition is higher than the original engine. Although the hydrogen enrichment could enhance the flame speed, the ratio is relatively small.

The comparison of emissions is shown (Figure 6) at hydrogen additional enrichment, where the value is recorded at the exhaust opening timing. The data are scaled so that it can be displayed on a single figure. In general, all pollutant emissions except NOx and CO₂ decrease with the increase of $HER$. This is because soot, THC, and CO can all be further oxidized and burn more completely as the $HER$ increases. The higher $HAR$ results in a higher in-cylinder temperature, which is mainly responsible for the complete oxidation of $CO+OH\iff C{O}_2+H$ and $CO+O+M\iff C{O}_2+M$, and improving the degree of complete combustion consequently [55]. The NOx formation is chiefly determined by mean temperature. For the case D100 + H2, the NOx emissions increased by 45%. It is recommended to combine after-treatment equipment for further optimization.

3.2. Effect of Hydrogen Substitution

In order to investigate the effect of hydrogen substitution enrichment on combustion, performance, and emissions, the conditions with $HSR$ of 0, 30, 60, and 90% are selected for further simulation. In all the simulations, the total energy is kept the same, where D and H represent diesel and hydrogen, respectively. The D10-H90 means the diesel energy fraction is 10% and the hydrogen energy fraction is 90%. The in-cylinder pressure is shown in Figure 7. The comparison of PFP, corresponding CA, and IMEP is listed in

Table 5. Comparison of PFP, corresponding CA position, and IMEP at hydrogen additional enrichment.

Parameter	PFP (bar)	PFP_CA (°CA, aTDC)	IMEP (bar)
D100-H0	21.75	7.26	21.53
D70-H30	23.93	5.92	22.00
D40-H60	28.33	1.85	22.01
D10-H90	30.13	−0.48	21.76

. The PFP increases with the increase of $HSR$. As the $HSR$ increases, the corresponding CA advances. This is because the higher burning velocity of hydrogen accelerates the flame speed and results in a higher PFP and an advanced CA. The IMEP of D10-H90 is lower than D40-H60, although the PFP is higher. However, too high a PFP is also not conducive to the modification of the original engine and even requires a redesign of the overall strength.

As shown in

Table 6. Contour of temperature distribution at hydrogen substitution enrichment.

CAD	D100-H0 (K)	D70-H30 (K)	D40-H60 (K)	D10-H90 (K)
−4 °CA aTDC
TDC
4 °CA aTDC
12 °CA aTDC
24 °CA aTDC

, the mean temperature is raised with the increase of $HSR$. The high-temperature zone appears earlier as the $HSR$ increases. When the mixing ratio is 90%, the high-temperature zone appears inside the combustion chamber, while the high-temperature zone at other ratios is beyond the combustion chamber near the cylinder head. Therefore, increasing $HSR$ is beneficial to reduce the heat load on the cylinder head. The high temperature zone inside the cylinder is more uniform with the increase of $HSR$. This is due to the fact that as the $HSR$ increases, the combustion mode gradually shifts from diffusion combustion to premixed combustion mode. This transition can be visualized from the HRR, as shown in Figure 8a. As the $HSR$ increases, the HRR gradually shows a dual trend. The first peak is caused by diesel igniting hydrogen, and the second peak is caused by the simultaneous ignition of hydrogen at multiple points of the combustion flame. When the $HSR$ is 60%, the two values are similar, while the peak of hydrogen combustion exotherm is much higher than that of diesel when the $HSR$ is 90%. It can also be seen from Table 6 that diffusion combustion generally starts from the tail, and the high temperature area is small. Because the laminar flame of hydrogen is faster, when premixed, the flame quickly spreads throughout the combustion chamber. Therefore, when the $HSR$ is higher, the second peak of heat release is also higher. The comparison of combustion phases is also shown in Figure 8b. Due to the faster burning rate of hydrogen and the shift in the combustion mode, the combustion phases are advanced. However, when the $HSR$ is 90%, the CA50 is already before the TDC, which is not conducive to power output.

For the emissions, as shown in Figure 9, all pollutant emissions except NOx are reduced, especially THC and CO, which have become small in order of magnitude at $HSR$ >30%. It can be analyzed in two ways: first, the increased $HSR$ reduces the diesel fuel, resulting in lower carbon for THC and CO, i.e., the total fuel carbon content. Secondly, due to the higher combustion temperature of hydrogen enrichment, hydrogen increases the free radical content of the reaction, which promotes the combustion process and makes the combustion more adequate. For soot, when the $HSR$ is higher, it leads to more homogeneous fuel mixing due to the lower mixture density. In addition, the hydrogen raises the temperature of the compression process, resulting in a lower in-cylinder temperature gradient, which deteriorates the soot generation environment. The increase in combustion temperature provides a good environment for soot oxidation. The synergistic effect of the two leads to extremely low soot. For CO₂, because hydrogen is a carbon-free fuel, CO₂ emissions are dramatically reduced at higher $HSR$ [56].

3.3. Effect of Pilot Injection Timing

From the above study, it can be concluded that hydrogen enrichment improves the combustion process and reduces pollutant emissions except NOx. For performance, the optimization of the injection strategy is necessary because the combustion phase at high $HSR$ conditions is too advanced [57]. In addition, at high $HSR$, the PFP is too high, resulting in excessive mechanical load. In addition, considering the goal of carbon reduction and the common operating conditions of rail internal combustion engines, this section investigates the effect of the pilot fuel injection timing on the performance, combustion, and emissions of dual-fuel engines.

The “I” represents the relative injection timing, i.e., H0-I0 is the injection timing of the original engine in diesel mode (the relative injection timing is 0 °CA). The H90-I5 represents 90% hydrogen energy and 10% diesel energy, and the diesel injection timing is delayed by 5 °CA. Figure 10 shows the comparison of in-cylinder pressure at different injection timings. In all cases, the PFP is higher than the diesel mode. The pressure rise rate is also increased in dual-fuel mode. As listed in

Table 7. Comparison of PFP, corresponding CA position, and IMEP with various pilot injection timings.

Parameter	PFP (bar)	PFP_CA (°CA, aTDC)	IMEP (bar)
H0-I0	21.75	7.26	21.53
H90-I0	30.13	−0.48	21.76
H90-I5	29.57	2.52	22.00
H90-I10	25.55	9.32	22.27
H90-I15	15.37	17.03	21.20

, the CA position of PFP decreases with the delayed injection timing. According to the above study, in dual-fuel mode, when at the maximum $HSR$, the delayed injection timing facilitates an increase in engine power output. The IMEP exceeds the original diesel condition. When delayed by 10 °CA, IMEP improves by 3.4% compared with diesel mode and 2.4% compared with dual-fuel mode.

The contour of temperature distribution is illustrated in

Table 8. Contour of temperature distribution with various pilot injection timings.

CAD	H90-I0 (K)	H90-I5 (K)	H90-I10 (K)	H90-I15 (K)
−4 °CA aTDC
TDC
8 °CA aTDC
16 °CA aTDC
24 °CA aTDC

. The appearance of the high temperature zone is delayed with the delay in injection timing. The mean temperature in the dual-fuel mode is higher than the diesel model, which undoubtedly causes higher NOx emissions. The trend of peak temperature is the same as the injection timing. When the injection timing is early, two separate high-temperature zones appear at the beginning of combustion. This is due to lower temperatures and incomplete fuel atomization. Therefore, the fuel in the combustion chamber and near the injector ignited hydrogen gas separately. When in the dual-fuel mode, the diesel is already atomized and broken at the end of the injection compared with the pure-diesel mode. This is due to the reduced in-cylinder ambient density, which facilitates fuel atomization. The HRR and combustion phases are shown in Figure 11. The combustion center is advanced with the early injection timing. This is because as the injection timing is delayed, the in-cylinder pressure and temperature decrease, which delay the combustion phases [58]. As the injection timing is delayed, the second peak of heat release rises initially before decreasing. When the injection timing is delayed by 15 °CA, the second peak of heat release dissipates, resulting in a shift to premixed combustion.

The effect of the injection timing on the emissions is shown in Figure 12. The soot, THC, CO, and CO₂ are much smaller than the original engine after hydrogen enrichment conditions. Alternatively, the carbon content of the fuel is reduced to 10% of the original engine because the pilot diesel is 1/10 of the original engine. For CO₂, it is reduced by nearly 90%. Conversely, due to the higher combustion temperature in dual-fuel mode, it leads to a dramatic reduction in the production of unburned emissions. At high $HSR$, the lower in-cylinder temperature gradient due to less diesel injection mass and more homogeneous mixing is not conducive to soot production. In addition, due to the higher temperature, a good environment for soot oxidation is provided. When the injection timing was delayed by 15 °CA, a small amount of THC appeared. For NOx generation, the NOx generation decreases with the delayed injection timing. This is because the delayed injection timing reduces the in-cylinder combustion temperature, which is not conducive to NOx generation [42]. Compared with the original injection timing, NOx is reduced by 53% when delaying 10 °CA.

4. Conclusions

In this paper, the effect of hydrogen on diesel combustion is investigated using a numerical method. The effects of additional hydrogen enrichment, hydrogen substitution, and pilot injection timing on combustion, performance, and emissions are investigated. The main findings are as follows:

(1)

The hydrogen enrichment was used for the study at very high loads. With the increase of $HER$, the diesel fuel atomized better and burned more fully. When in a small $HER$, the combustion phase of the engine had a small range of variation and a consistent shape of the HRR. Due to the increase in temperature after hydrogen enrichment, it leads to higher NOx. However, CO, soot, and THC emissions were reduced due to more complete combustion.

(2)

The hydrogen substitution was also studied in order to reduce carbon emissions and increase the $HSR$. As $HSR$ increases, the peak cylinder pressure increases, and the combustion phase advances. The higher combustion temperature of hydrogen leads to more NOx. When the $HSR$ was 90%, the center of combustion was located unfavorably in front of the upper stop, resulting in a lower IMEP.

(3)

Since the center of gravity of combustion is too advanced at a higher $HSR$, the study of injection timing was carried out. With the delay in injection timing, the in-cylinder pressure decreases and the combustion temperature decreases. In addition, the work capacity increases and the NOx decreases.

(4)

The implementation of hydrogen injectors necessitates increased control standards, as the low energy density of hydrogen may cause limitations in larger-scale applications. In addition, hydrogen safety must be taken into account.

In conclusion, this paper investigates the effect of hydrogen on diesel engine combustion and emissions and provides theoretical guidance for practical design optimization. Subsequent work will couple EGR with an injection strategy to further increase power and reduce NOx emissions.