1 s2.0 S0959652624000696 Main
1 s2.0 S0959652624000696 Main
1 s2.0 S0959652624000696 Main
A R T I C L E I N F O A B S T R A C T
Handling Editor: Panos Seferlis In pursuit of unleashing ammonia’s potential in internal combustion engines, an extensive computational study
was conducted to explore the effects of two advanced combustion techniques: pre-chamber combustion (PCC)
Keywords: and dual-fuel compression ignition (DF− CI) on a heavy-duty ammonia engine. In the PCC mode, hydrogen was
Ammonia introduced into the pre-chamber (PC) to generate distributed reacting jets, with a small energy fraction of 0.5%.
Pre-chamber combustion
The case featuring a slightly narrower PC throat (diameter of 5.4 mm) yielded the highest indicated thermal
Dual-fuel
efficiency (ITE) of 50.4%, owing to the rapid distribution of reacting jets and advanced combustion phasing.
Spark ignition
Hydrogen Further increase in hydrogen energy fraction and reduction in throat diameter resulted in a lower ITE because of
the enhanced wall heat transfer loss. The DF− CI mode also exhibited significant potential for achieving high
efficiency. Through a systemic optimization of parameters such as diesel injection timing, spray included angle,
injection pressure, swirl ratio, and piston shape, a significantly enhanced ITE of 50.3% was attained, 7.2% higher
than the baseline scenario. The improvement was primarily attributed to the promoted mixing of diesel with air.
Among the design parameters, the swirl ratio and spray included angle exhibited the most significant impact on
the engine performance. Overall, both the PCC and DF− CI modes were found to yield high ITE and low ammonia
emissions. However, because 20% of the diesel energy fraction was utilized to mitigate ammonia slip in the
DF− CI mode, notably higher greenhouse gas emissions (carbon dioxide of 96 g/kW-h) were generated than in the
PCC mode. In this regard, the ammonia-hydrogen PCC mode should be a more promising solution to achieve
zero-carbon emissions for future ammonia engines.
2023) and ammonia (Uddeen et al., 2023; Wang et al., 2023a; Xu et al.,
2023a; Zhou et al., 2023) are receiving significant attention in the field
1. Introduction
of engine combustion due to their carbon-neutral nature. However,
hydrogen requires extremely low temperatures to be liquefied and is
The increasing greenhouse gas (GHG) emissions are considered a
typically stored and transported as a highly compressed gas, leading to
major cause of global warming. In response to this critical issue, in 2015,
substantially higher costs. In contrast, ammonia can be easily liquefied
the Paris Agreement was forged through the collaborative efforts of 196
under relatively low pressure (approximately 9 bar) at room tempera
countries. The primary aim of this accord is to regulate GHG emissions
ture and it has a volumetric energy density 1.6 times greater than
and limit the global temperature increase to within 2 ◦ C above pre-
hydrogen. Therefore, ammonia is considered a good hydrogen carrier.
industrial levels. To fulfill the target, it is critical to transform the
Moreover, the well-established industrial infrastructure for the pro
transportation sector which is one of the major contributors to carbon
duction, transportation, and storage of ammonia makes its extensive use
dioxide (CO2) emissions. While the battery electric and fuel cell vehicles
in transportation and power generation a feasible prospect in the near
are rapidly growing, conventional internal combustion engines (ICEs)
future.
powered by renewable fuels such as hydrogen (H2) (Matthias et al.,
Ammonia can be utilized in both compression ignition (CI) and spark
2012) and ammonia (NH3) (Valera-Medina et al., 2018), are being
ignition (SI) engines. However, due to its high autoignition temperature,
explored as competitive and sustainable alternatives for future trans
low flame velocity, and limited flammability, ammonia requires a high
portation, particularly in heavy-duty applications.
compression ratio or elevated intake temperatures for successful ignition
Both hydrogen (Babayev et al., 2022; Gao et al., 2022; Lee et al.,
* Corresponding author.
E-mail address: [email protected] (X. Liu).
https://doi.org/10.1016/j.jclepro.2024.140622
Received 27 October 2023; Received in revised form 19 December 2023; Accepted 4 January 2024
Available online 9 January 2024
0959-6526/© 2024 Elsevier Ltd. All rights reserved.
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
in SI engines (Gray et al., 1967). Even in such cases, pure ammonia The pre-chamber combustion (PCC) mode (Zhou et al., 2023; Liu
combustion in SI engines still suffers from reduced fuel efficiency and et al., 2023a; Dong et al., 2023; Cao et al., 2023; Wei et al., 2023)
increased incomplete combustion compared to conventional gasoline SI emerges as a potential solution for achieving high efficiency and low
engines. To enhance engine performance, hydrogen or gasoline is usu GHG emissions in ammonia engines. Different from the DF− CI mode,
ally used to blend with ammonia to enhance the reactivity (Pandey which depends on high-reactivity hydrocarbon fuels such as diesel to
et al., 2023). Additionally, the implementation of multiple spark plugs initiate high-temperature combustion, the PCC mode generates distrib
can expedite the flame propagation process, thereby significantly uted reacting jets by capitalizing on the pressure differential between
boosting the engine’s thermal efficiency. In contrast, in CI engines, an the pre-chamber (PC) and main chamber (MC) (Zhao et al., 2023). In
auxiliary high-reactivity fuel, such as diesel or dimethyl ether, is used to comparison to the passive PCC mode, which involves no auxiliary fuel
initiate high-temperature ignition, a mode referred to as dual-fuel (DF)− delivery through the PC, the active PCC mode extends the lean-burn
CI (Xu and Bai, 2023). Given that CI engines typically possess higher limit and ensures enhanced combustion stability (Hlaing et al., 2019;
compression ratios than SI engines, they often yield greater thermal Liu et al., 2023b). Extensive research has explored the impact of various
efficiency and lower incomplete combustion losses in the DF− CI mode. PC and piston geometries on the combustion characteristics of PC en
Consequently, CI engines are the preferred choice for heavy-duty ap gines (Shah et al., 2015; Silva et al., 2020, 2022; Hlaing et al., 2022). The
plications such as marine ships, trucks, and power generators. throat diameter was identified as having a more pronounced effect on PC
Early research into the ammonia dual-fuel compression ignition buildup than the nozzle’s diameter and length (Silva et al., 2020). Given
(DF− CI) mode was pioneered by Pearsall and Garabedian (1967). the substantial interaction between reacting jets and the piston wall,
Through an examination of the impact of various pilot fuels on power co-optimization of the PC and piston is crucial for achieving optimal
output and fuel consumption, their work revealed that fuels with high engine performance (Liu et al., 2023c). Despite extensive studies on
cetane numbers contributed to superior engine performance. Optimal gasoline (Attard and Blaxill, 2012), natural gas (Shah et al., 2014),
fuel economy was achieved within an overall equivalence ratio (φ) range methanol (Liu et al., 2023b; Wang et al., 2022), and hydrogen PCC (Liu
of 0.7–0.75, while the maximum power output was attained at an overall et al., 2023d), research on ammonia PCC remains relatively limited.
φ of approximately 0.9. Reiter and Kong (2011) explored the effects of Note that the active ammonia PCC mode, when combined with auxiliary
pilot energy fraction on combustion and emission characteristics in a fuel such as hydrogen supplied to the PC, significantly accelerates the
four-cylinder ammonia DF− CI engine. Their findings indicated that the combustion of the ammonia-air mixture. While this approach has the
ideal fuel economy demanded the utilization of 40–60% of the pilot fuel potential to achieve zero CO2 emissions, prior studies on the NH3–H2
energy fraction, with a notable increase in oxides of nitrogen (NOx) PCC mode (Wang et al., 2023a; Liu et al., 2023e) in light-duty applica
emissions when this fraction exceeded 50%. Niki et al. (2019) reported tions reported relatively high incomplete combustion losses with a
that ammonia emissions were promoted with the addition of ammonia single-hole PC design. Consequently, there is a pressing need for further
to the intake. While the use of pilot or post-injection could reduce fundamental research and optimization studies to enhance the efficacy
ammonia emissions, it unexpectedly increased nitrous oxide (N2O) of this innovative combustion mode.
emissions. More recent work by Jin et al. (2023) introduced a split in In the present work, a comprehensive investigation of the NH3–H2
jection strategy that effectively reduced N2O emissions with an PCC and NH3-diesel DF− CI combustion modes on a heavy-duty engine
ammonia energy fraction of 50%. Nevertheless, as N2O holds almost 300 was conducted, to identify which combustion mode could be the better
times the global warming effect of CO2, the net reduction in GHG solution for future ammonia engines. The computational fluid dynamics
emissions was only 14.2% when compared to pure diesel combustion. (CFD) modeling method was employed. Under both modes, some of the
Consequently, the adoption of after-treatment systems and advanced key design parameters such as ammonia energy fraction and piston
combustion strategies is imperative for further GHG reduction. shape etc. were varied and optimized. In particular, to attain the best
2
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
engine performance in the PCC mode, an optimized multi-hole PC based including spray angle (SA), swirl ratio (SR), injection pressure (Pinj), and
on our previous study on methane PCC mode (Silva et al., 2022) was nozzle hole diameter (Dnoz), were adjusted to achieve optimal engine
implemented. Extensive analyses were performed by examining the performance. Furthermore, for both combustion modes, three distinct
engine performance and emissions data. The findings of this work are piston shapes were examined. Detailed geometry information for these
expected to guide the design of future ammonia engines. pistons is provided in Fig. 2. P0 represents an ω-type piston used in the
PCC engine experiment at KAUST (Hlaing et al., 2020). P1 is a flat piston
2. Methodology that demonstrated exceptional engine performance in the methanol PCC
mode (Liu et al., 2023b). P2 is an ω-type piston employed in the DF− CI
2.1. Engine configuration experiment at Tianjin University (Jin et al., 2023). Lastly, P2* is a U-type
piston that was found to be an optimal design in the methane PCC mode
The prototype engine under investigation is a single-cylinder heavy- (Liu et al., 2023c).
duty diesel engine. Table 1 provides an overview of the engine and pre-
chamber (PC) specifications. A more in-depth analysis of the engine 2.3. Computational models
configuration can be referred to (Hlaing et al., 2020). The narrow-throat
PC developed at KAUST was designed as a drop-in solution for the Three-dimensional (3D) CFD simulations were conducted using
commercial high-pressure DI injector. This new assembly incorporates CONVERGE/3.0 (Richards et al.). In the PCC mode, full-cycle simula
the spark plug, pressure sensor, and auxiliary gaseous fuel supply sys tions starting from the exhaust valve opening (EVO) timing were per
tem. Previous experimental studies (Hlaing et al., 2020, 2021) have formed to attenuate the influence of internal gas exchange on computing
demonstrated the efficacy of the narrow-throat PC in significantly uncertainties (Liu et al., 2023g). In the DF− CI mode, closed-cycle sim
extending the lean-burn limits for various fuel types. However, when ulations were performed from the intake valve closing (IVC) to EVO
auxiliary hydrogen is introduced into the PC, there’s an elevated risk of timing to save computational expenses. In addition, given the axisym
detonation and knocking (Liu et al., 2023d). Therefore, a larger throat metric piston geometry and 8-hole injector configuration, a 45◦ -sector
design proves more suitable for the hydrogen-assisted PCC mode (Liu computational domain was utilized. Note that the initial pressure and
et al., 2023f). In this study, the optimized large-throat PC selected out of temperature at the IVC timing for the DF-PI cases were calculated from
300 designs (Silva et al., 2022), was utilized as a reference. Furthermore, the full-cycle simulations. For both combustion modes, the intake
it’s important to note that the compression ratio (CR) was increased mixture was assumed homogenous for simplifications. In the PCC mode,
from the original 11.5:1 to 20:1 in order to promote ammonia com an injection rate profile referenced from methane injection (Hlaing
bustion. This led to a reduction in the PC throat height and volume, a et al., 2020) was applied to model the hydrogen injection process. In the
necessary step to mitigate the maximum pressure rise rate (MPRR). For a DF− CI mode, an in-house injector model (Liu et al., 2023h) was used to
visual comparison, the original and optimized PCs are presented in calculate the ROI profile for diesel injection.
Fig. 1. The Redlich-Kwong equation of state (Redlich and Kwong, 1949)
was employed to calculate density. The mixture-averaged diffusion
properties were considered. The renormalization group k-ε model (Han
2.2. Parametric cases and Reitz, 1995) was used for turbulence closure. The detailed chemistry
combustion solver SAGE (Senecal et al., 2003) coupled with the newly
In this study, the PCC and DF− CI modes were investigated. The key developed reduced mechanism was applied to simulate the combustion
details of the engine operating conditions and parametric cases are process. The wall heat transfer (HT) loss was predicted by the O’Rourke
summarized in Table 2. For both combustion modes, the intake tem model (O’Rourke and Amsden, 2000). Furthermore, in the DF− CI mode,
perature and pressure were maintained at room conditions at, 298 K and the Lagrangian method was adopted to simulate the diesel injection
1 atm. Ammonia injection occurred through the intake port at − 360◦ process. The spray breakup, droplet collision, and droplet evaporation
crank angle after top dead center (CA aTDC). In the PCC mode, auxiliary were predicted by the Kelvin-Helmholtz Rayleigh-Taylor model (Beale
hydrogen was introduced into the PC with a start of injection (SOI) and Reitz, 1999), the no-time-counter algorithm (Schmidt and Rutland,
timing also set at − 360◦ CA aTDC. The optimization of engine perfor 2000), and the Frossling correlation (Froessling, 1956), respectively. For
mance involved variations in the PC fuel energy ratio (PCFR), defined as, all the simulation cases, the conservation equations were solved with the
QH2 PISO algorithm and a varying time-step scheme was implemented to
PCFR = (1) control the computational time step. Considering the balance between
QNH3 + QH2
computational accuracy and cost, a unity convective CFL number was
In the DF− CI mode, considering the high CR, a high ammonia energy employed. In particular, in the PCC mode, a peak Mach CFL number of 3
fraction (AEF) of 80% was applied to reduce CO2 emissions. A further was imposed during the jet issuing stage to ensure the computational
increase in AEF would lead to a significantly higher combustion loss. accuracy.
Moreover, various primary engine design and control parameters, Comprehensive details regarding the mesh configuration can be
found in Table S1 in the Supplementary Material. Fig. S1 presents the
Table 1 computational meshes used in both combustion modes. In the simula
Engine and PC specifications. tions involving PCC, fixed embeddings were applied throughout the PC
Type 4-stroke region with a finer grid size of 0.25 mm during the primary combustion
stage. As the simulation transitioned to the SI stage, further refinement
Bore/Stroke (mm) 131/158
Connecting rod length (mm) 255 was employed in the regions around the spark plug and the PC nozzle
Displacement volume (L) 2.1 channel, with a minimum grid size of 0.125 mm. Downstream of the
Compression ratio (bowl) 11.5/20a nozzles, the mesh was refined with a grid size of 0.25 mm. In contrast,
Swirl ratio 0 for the DF− CI simulation cases, specific attention was given to the
PC volume (cm3) 5.07/3.75a
PC throat diameter (mm) 3.3/6.4a
injector downstream, with a minimum grid size of 0.125 mm. This finer
PC nozzle diameter (mm) 1.5/1.42a mesh was implemented to more accurately resolve the flow field in the
PC nozzle layer number 2/1a vicinity of the injector nozzle. In all simulation cases, the adaptive mesh
PC nozzle number 12/8a refinement (AMR) model was activated, which allowed the generation
PC nozzle angle (◦ ) 134/149a
of a minimum mesh size of 0.25 mm in the bulk region during the pri
a
Bold font is the case with the optimized PC. mary combustion process, ensuring a high level of detail and accuracy in
3
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
Fig. 2. Profiles of various piston geometries. Note that P0, P1, and P2 were
used for DF− CI simulations, and P0, P1, and P2* were used for PCC
simulations.
the simulations.
4
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
Table 3 Both the mechanisms of this work and Zhang et al. (2023) gave similar
Summary of the experimental uncertainties. predictions, exhibiting overall good agreement with the experiment. In
Exp. Type Uncertainty particular, the NO and N2O formation was accurately predicted for a
range of temperatures and H2 blending ratios. Fig. S2 compares the
ID NH3 <20%
NH3/H2 <20% experimental (Ichikawa et al., 2015) and predicted laminar flame speeds
NC7H16 Not reported of the NH3/H2/air mixtures at φ = 1.0 and P = 5 atm. Again, the reduced
NH3/NC7H16 <15% mechanism of this work predicted similar results to the Zhang mecha
JSR species NH3/H2 <10% nism (Zhang et al., 2023), both showing a good agreement with the
Flame species NC7H16
measured data with the variation of the H2 blending ratio. Furthermore,
<3%
Flame speed NH3/H2 Not reported
NC7H16 <0.8 cm/s the n-heptane combustion chemistry was validated against the
DF− CI engine Cylinder pressure <0.4 bar measured species data of a premixed NC7H16/nitrogen (N2)/O2 flame
Emissions <1% (Bakali et al., 1998) and laminar flame speeds of NC7H16/air mixtures at
Ammonia mass <0.35%
different φ (van Lipzig et al., 2011), as presented in Fig. S3. The good
Diesel mass <0.12%
predictions of n-heptane combustion indicate that the combustion
chemistry of n-heptane has been well established by the combined
was applied to compute IDs in shock tubes while varying volume profiles reduced mechanism of this work.
were implemented in rapid compression machines. Fig. 4 presents the
measured (Shu et al., 2019; Chen et al., 2021; Yu et al., 2020; Fieweger 2.4.2. Validation of engine combustion
et al., 1997) and predicted IDs for NH3, NH3/H2 mixtures, NC7H16, and The ammonia-diesel DF− CI combustion experiment conducted by
NH3/NC7H16 mixtures. In Fig. 4(a) and (b), the predicted IDs by the Tianjin University (Jin et al., 2023) was employed for the validation of
Zhang mechanism (Zhang et al., 2023) were also included for compar the engine simulations. The engine specifications and experimental
ison. Both mechanisms gave similar predictions, in reasonable agree conditions are summarized in Table S2. The measured uncertainties are
ment with the experiment. In addition, the combined reduced listed in Table 3. Further details of the experiment can be found in (Jin
mechanism was also able to reasonably predict the IDs of the NC7H16 et al., 2023). Briefly, the experiment was conducted on a heavy-duty
and NH3/NC7H16 mixtures, as shown in Fig. 4(c) and (d). Note that when single-cylinder diesel engine, which has a high CR at 18.5:1. The en
compared to the reduced mechanism developed by Xu et al. (2023a), the gine speed was set at 1000 rpm and the IMEP was controlled at about 6
mechanisms developed in this study, as well as by Wang et al. (2023b), bar. The intake pressure was fixed at 1.2 bar. The diesel energy fraction
exhibited slightly better agreement with experimental data, particularly ranged from 0% to 80% and was injected at − 14◦ CA aTDC with a
in the lower-temperature range (650–750 K). This improvement was pressure of 600 bar. Fig. S4 illustrates the computational domain.
primarily attributed to the incorporation of the NH3 and NC7H16 Because an 8-hole injector was utilized in the experiment, a sector
cross-reaction into the mechanism. computational domain was modeled to save computational expenses.
Fig. 5 shows the measured (Zhang et al., 2021) and predicted evo Fig. 6 shows the measured and predicted pressure and apparent heat
lution profiles of NH3, H2O, NO, and N2O for the NH3/N2/O2 mixture release rate (AHRR) profiles at different AEF. Fig. 7 summarizes the
with different H2 blending ratios at pressure (P) of 1 atm and φ of 1.0. comparison of the indicated thermal efficiency (ITE) and NOx emissions.
Fig. 4. Measured and predicted IDs for (a) NH3, (b) NH3/H2, (c) NC7H16, and (d) NH3/NC7H16.
5
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
Fig. 5. Measured and predicted evolution profiles of (a) NH3, (b) H2O, (c) NO, and (d) N2O at the variation of H2 volume fraction (φ = 1.0, P = 1.0 atm, and
residence time at 1.0 s). Solid lines: this work; Dash dot lines: Zhang mechanism.
Fig. 6. Comparison of the measured and predicted P-AHRR traces at different AEF.
Fig. 7. Comparison of the measured and predicted ITE and NOx emissions. To optimize the engine performance at the ammonia-diesel DF− CI
mode, the SA, Pinj, SR, piston shape, and Dnoz of the diesel injector were
An overall reasonable agreement is seen between the simulation and varied and examined. In addition, the pilot diesel SOI timing was swept
experiment. Although the NOx emissions were slightly overpredicted, from − 15◦ to − 35◦ CA aTDC to identify the optimal operating point. To
the declining trend with the growth in AEF was well captured. minimize CO2 emissions, the diesel quantity was kept as small as
For the ammonia-hydrogen PCC mode, no experimental data are possible. Therefore, a 20% energy fraction of diesel was adopted, cor
available for rigorous validation. Therefore, our previously well- responding to the AEF at 80%. Further growth in AEF or advance in SOI
validated computational setup for the methane PCC (Silva et al., 2022; timing would increase the incomplete combustion loss and lower the
ITE. Two injection pressures at 600 and 1000 bar were evaluated, and
6
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
the further increase in injection pressure did not benefit the improve
ment in ITE and yielded a higher MPRR. For the same reason, only two
swirl ratios at 0 and 1.5 were studied.
For quantitative analyses, energy balance accounting for the
contribution of engine work, wall heat transfer loss, incomplete com
bustion loss, and exhaust loss was evaluated by,
QFuel = W + QHT + QIncom + HExh (2)
7
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
Fig. 10. Predicted distributions of NC7H16, T, NH3, and wall temperatures on the piston and head.
Fig. 12. Predicted traces of P-AHRR, energy balance, NOx emissions, and traces
of Tavg and NOx mass.
8
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
Fig. 13. Predicted distributions of NC7H16, PDF of φ, and T on the jet trajectory and central planes as well as piston and cylinder walls.
9
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
and NOx emissions. Among all the cases explored, the operating point
featuring Dnoz = 0.13 mm and SOI = − 30◦ CA aTDC emerges as the
optimal configuration within the DF− CI mode. This configuration yields
the highest ITE at an impressive 50.3% and minimizes combustion loss
to a mere 0.1%.
Fig. 15. Predicted traces of P-AHRR, energy balance, NOx emissions, and traces
of Tavg and NOx mass.
Fig. 16. Predicted distributions of NC7H16, T on the jet trajectory and central planes, piston and cylinder walls, and PDF of φ.
10
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
11
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
Fig. 20. Predicted distributions of λ, T, V, TKE, NO, OH, and isosurface of T = 1600 K.
Fig. 21. Predicted ITE, combustion loss, and NOx emissions for the cases with Fig. 22. Predicted traces of P-AHRR, energy balance, NOx emissions, and traces
different pistons. of Tavg and NOx mass.
P2*, the integrated value across the entire chamber is still higher. ST at − 15◦ CA aTDC.
Fig. 25 compares the predicted P-AHRR traces, energy balance, NOx
3.2.3. Effects of PC throat diameter emissions, and Tavg-NOx mass profiles for the three cases with varying
The effect of Dthroat was also examined to further optimize the Dthroat values at ST = − 15◦ CA aTDC. The case with Dthroat = 6.4 mm
NH3–H2 PCC mode. Three different Dthroat (6.4, 5.4, and 4.4 mm) were initiates the earliest combustion process within the PC. However, as the
used. The previous case with PCFR = 0.5% and the P0 piston was taken combustion process progresses, its combustion phasing is surpassed by
as the baseline case. For cases with Dthroat values of 5.4 and 4.4 mm, the the cases with Dthroat of 5.4 and 4.4 mm, as also shown by the predicted
nozzle included angle was enlarged to 160◦ to further minimize jet CA5, CA50, and CA90 values in Fig. S14. The case with the smaller
flame-piston interaction. Fig. 24 shows the predicted ITE, combustion Dthroat (4.4 mm) yields a higher peak P and AHRR, resulting in a more
loss, and NOx emissions, for all these configurations. Reducing the Dthroat advanced combustion phasing and a reduction in combustion loss, pri
effectively contributes to a reduction in combustion loss. However, an marily due to the higher combustion temperature. However, this in
excessively large Dthroat leads to a slightly lower ITE but higher NOx crease in combustion temperature also promotes wall HT loss and NOx
emissions. The highest ITE was achieved with Dthroat set at 5.4 mm and formation, resulting in a slightly lower ITE but higher NOx emissions for
12
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
Fig. 24. Predicted ITE, combustion loss, and NOx emissions for various
Fig. 25. Predicted traces of P-AHRR, energy balance, NOx emissions, and traces
PCC cases.
of Tavg and NOx mass.
the case with Dthroat = 6.4 mm compared to the case with Dthroat = 5.4 reacting jets and a higher combustion temperature than the case with
mm. Dthroat = 4.4 mm. This is primarily due to the richer mixture distribution
Fig. 26 shows more details of the mixing, combustion, and NO for within the PC before ST. However, as the combustion process continues,
mation processes, shedding light on how throat design influences these the latter case (Dthroat = 4.4 mm) soon catches up and generates a higher
phenomena. Note that throat design played a pivotal role in shaping combustion temperature, and subsequently produces more NO.
both the mixing formation and the jet flame-issuing processes. In
particular, the case with the largest throat (Dthroat = 6.4 mm) induces the
most pronounced jet flame quenching. Furthermore, owing to the 3.3. Comparison of the PCC and DF− CI modes
smaller pressure differential between the PC and MC (Fig. 25), the jet
velocity and TKE are comparatively lower in this case than in the other Two optimized cases operating in the PCC and DF− CI modes were
two cases with larger throat diameters. Consequently, this case experi selected for comparative analysis. The details of these two cases are
ences less flame-wall interaction, leading to reduced wall HT loss. Note summarized in Table 4, and the results are presented in Fig. 27. Detailed
that at − 5◦ CA aTDC, the case with Dthroat = 5.4 mm shows even faster- quantitative comparison was further conducted and shown in Table 5.
Overall, these two cases exhibit similar features in terms of combustion
13
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
Fig. 27. Predicted traces of P-AHRR and T-NOx as well as the major engine metrics for the representative PCC and DF− CI cases.
14
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
from the intense dual-fuel combustion heat release process, Science Inc. for providing the CONVERGE license.
leading to the higher combustion temperature and more NO
formation. Notably, the PCC mode generated no CO2 emissions, Appendix A. Supplementary data
marking a crucial step toward the stringent carbon-neutral target
in future transportation. Supplementary data to this article can be found online at https://doi.
(2) In the DF− CI mode, over-large and small spray angles led to org/10.1016/j.jclepro.2024.140622.
lower ITE, primarily because of the increased combustion heat
release within the squish region. This slowed down the combus References
tion of the bulk ammonia-air mixture in the central piston.
Additionally, increased injection pressure and swirl ratio effec Attard, W.P., Blaxill, H., 2012. A gasoline fueled pre-chamber jet ignition combustion
system at unthrottled conditions. SAE Int. J. Eng. 5, 315–329.
tively enhanced pilot fuel-air mixing and advanced combustion Babayev, R., Im, H.G., Andersson, A., Johansson, B., 2022. Hydrogen double
phasing, resulting in lower combustion losses and higher ITE. compression-expansion engine (H2DCEE): a sustainable internal combustion engine
Moreover, the reduction in nozzle diameter further improved fuel with 60%+ brake thermal efficiency potential at 45 bar BMEP. Energy Convers.
Manag. 264, 115698.
economy and reduced combustion losses due to an enhanced Bakali, A.E., Delfau, J.L., Vovelle, C., 1998. Experimental study of 1 atmosphere, rich,
mixing process and advanced combustion phasing. premixed n-heptane and iso-octane flames. Combust. Sci. Technol. 140, 69–91.
(3) Piston shape played a significant role in both the DF− CI and PCC Beale, J.C., Reitz, R.D., 1999. Modeling spray atomization with the Kelvin-Helmholtz/
Rayleigh-Taylor hybrid model. Atomization Sprays 9, 623–650.
modes. In the DF− CI mode, the flat piston yielded the highest ITE Cao, J., Dong, D., Wei, F., Long, W., Xiao, G., Jiang, L., Li, B., et al., 2023. Investigation
because of the generation of multiple ignition spots and reduced on jet controlled diffusion combustion (JCDC) mode applied on a marine large-bore
flow resistance from the piston bowl. In the PCC mode, the two-stroke engine. J. Clean. Prod. 429, 139546.
Chang, Y., Jia, M., Wang, P., Niu, B., Liu, J., 2022. Construction and derivation of a series
ω-piston generated the highest ITE due to the delayed jet flame-
of skeletal chemical mechanisms for n-alkanes with uniform and decoupling
piston interaction, advancing the combustion phasing. structure based on reaction rate rules. Combust. Flame 236, 111785.
(4) In the ammonia-hydrogen PCC mode, an increase in the hydrogen Chen, J., Jiang, X., Qin, X., Huang, Z., 2021. Effect of hydrogen blending on the high
energy fraction facilitated jet flame formation, advancing com temperature auto-ignition of ammonia at elevated pressure. Fuel 287, 119563.
Dong, S., Wang, B., Jiang, Z., Li, Y., Gao, W., Wang, Z., Cheng, X., et al., 2022. An
bustion phasing, although it also contributed to increased wall experimental and kinetic modeling study of ammonia/n-heptane blends. Combust.
HT losses, resulting in a decrease in ITE. Furthermore, a slightly Flame 246, 112428.
narrower throat design yielded the highest ITE compared to the Dong, D., Wei, F., Long, W., Dong, P., Tian, H., Tian, J., Wang, P., et al., 2023. Optical
investigation of ammonia rich combustion based on methanol jet ignition by means
baseline case, due to a faster jet ejection process that advanced of an ignition chamber. Fuel 345, 128202.
combustion phasing and reduced exhaust losses. Fieweger, K., Blumenthal, R., Adomeit, G., 1997. Self-ignition of S.I. engine model fuels:
(5) In summary, the ammonia-hydrogen PCC mode exhibited a shock tube investigation at high pressure. Combust. Flame 109, 599–619.
Froessling, N., 1956. Evaporation, heat transfer, and velocity distribution in two-
competitive fuel economy and outperformed the ammonia-diesel dimensional and rotationally symmetrical laminar boundary-layer flow. N.A.C.A.
DF− CI mode in terms of NOx and CO2 emissions. However, note 168. AD-B189.
that this conclusion is derived in the context of premixed Gao, J., Wang, X., Song, P., Tian, G., Ma, C., 2022. Review of the backfire occurrences
and control strategies for port hydrogen injection internal combustion engines. Fuel
ammonia combustion. Moreover, since this study primarily 307, 121553.
focused on intermediate engine load (IMEP = 10 bar), addressing Gray Jr., J.T., Dimitroff, E., Meckel, N.T., Quillian Jr., R., 1967. Ammonia fuel-engine
the challenges of extending operations to low- and high-load compatibility and combustion. SAE Trans. 75, 785–807.
Han, Z., Reitz, R.D., 1995. Turbulence modeling of internal combustion engines using
conditions needs further investigation. Needless to say, further
RNG κ-ε models. Combust. Sci. Technol. 106, 267–295.
optimization of engine design and control parameters remains Hlaing, P., Echeverri Marquez, M., Bhavani Shankar, V.S., Ben Houidi, M., Johansson, B.,
crucial for the successful implementation of the ammonia- Cenker, E., 2019. A Study of Lean Burn Pre-chamber Concept in a Heavy Duty
hydrogen PCC mode. Engine. SAE Technical Paper. https://doi.org/10.4271/2019-24-0107, 2019-24-
0107.
Hlaing, P., Echeverri Marquez, M., Singh, E., Almatrafi, F., Ben Houidi, M.,
CRediT authorship contribution statement Johansson, B., Cenker, E., 2020. Effect of Pre-chamber Enrichment on Lean Burn Pre-
chamber Spark Ignition Combustion Concept with a Narrow-Throat Geometry. SAE
Technical Paper. https://doi.org/10.4271/2020-01-0825, 2020-01-0825.
Xinlei Liu: Conceptualization, Data curation, Formal analysis, Hlaing, P., Echeverri Marquez, M., Burgos, P., Cenker, E., Ben Houidi, M., Johansson, B.,
Investigation, Methodology, Software, Validation, Visualization, 2021. Analysis of Fuel Properties on Combustion Characteristics in a Narrow-Throat
Writing – original draft. Qinglong Tang: Funding acquisition, Investi Pre-chamber Engine. SAE Technical Paper. https://doi.org/10.4271/2021-01-0474,
2021-01-0474.
gation, Project administration, Writing – review & editing. Hong G. Im: Hlaing, P., Echeverri Marquez, M., Cenker, E., Im, H.G., Johansson, B., Turner, J.W.G.,
Funding acquisition, Investigation, Project administration, Resources, 2022. Effects of volume and nozzle area in narrow-throat spark-ignited pre-chamber
Supervision, Writing – review & editing. combustion engines. Fuel 313, 123029.
Ichikawa, A., Hayakawa, A., Kitagawa, Y., Kunkuma Amila Somarathne, K.D., Kudo, T.,
Kobayashi, H., 2015. Laminar burning velocity and Markstein length of ammonia/
Declaration of competing interest hydrogen/air premixed flames at elevated pressures. Int. J. Hydrogen Energy 40,
9570–9578.
Jin, S., Wu, B., Zi, Z., Yang, P., Shi, T., Zhang, J., 2023. Effects of fuel injection strategy
The authors declare that they have no known competing financial and ammonia energy ratio on combustion and emissions of ammonia-diesel dual-fuel
interests or personal relationships that could have appeared to influence engine. Fuel 341, 127668.
the work reported in this paper. Kee, R.J., Rupley, F.M., Miller, J.A., 1989. CHEMKIN II: A Fortran Chemical Kinetics
Package for the Analysis of Gas-phase Chemical Kinetics, Sandia Laboratories
Report, S. 89-8009B. Sandia National Laboratories.
Data availability Lee, S., Hwang, J., Bae, C., 2023. Understanding hydrogen jet dynamics for direct
injection hydrogen engines. Int. J. Engine Res., 14680874231169562
Data will be made available on request. Li, R., Konnov, A.A., He, G., Qin, F., Zhang, D., 2019. Chemical mechanism development
and reduction for combustion of NH3/H2/CH4 mixtures. Fuel 257, 116059.
Liu, X., Wang, H., Zheng, Z., Yao, M., 2021. Development of a reduced primary reference
Acknowledgments fuel-PODE3-methanol-ethanol-n-butanol mechanism for dual-fuel engine
simulations. Energy 235, 121439.
Liu, Z., Zhou, L., Wei, H., 2023a. Experimental investigation on the performance of pure
The authors thank the support from King Abdullah University of ammonia engine based on reactivity controlled turbulent jet ignition. Fuel 335,
Science and Technology. Dr. Qinglong Tang is funded by the National 127116.
Natural Science Foundation of China (NSFC) through its project Liu, X., Sharma, P., Silva, M., AlRamadan, A.S., Cenker, E., Magnotti, G., Im, H.G.,
2023b. Computational investigation of methanol pre-chamber combustion in a
52206166. Computational simulations utilized the clusters of the heavy-duty engine. Appl. Energy Combust. Sci. 15, 100192.
KAUST Supercomputing Laboratory. The authors thank Convergent
15
X. Liu et al. Journal of Cleaner Production 436 (2024) 140622
Liu, X., Silva, M., Mohan, B., AlRamadan, A.S., Cenker, E., Im, H.G., 2023c. Shah, A., Tunestal, P., Johansson, B., 2015. Effect of Pre-chamber Volume and Nozzle
Computational optimization of the performance of a heavy-duty natural gas pre- Diameter on Pre-chamber Ignition in Heavy Duty Natural Gas Engines. SAE
chamber engine. Fuel 352, 129075. Technical Paper. https://doi.org/10.4271/2015-01-0867, 2015-01-0867.
Liu, X., Aljabri, H., Silva, M., AlRamadan, A.S., Ben Houidi, M., Cenker, E., Im, H.G., Shu, B., Vallabhuni, S.K., He, X., Issayev, G., Moshammer, K., Farooq, A., Fernandes, R.
2023d. Hydrogen pre-chamber combustion at lean-burn conditions on a heavy-duty X., 2019. A shock tube and modeling study on the autoignition properties of
diesel engine: a computational study. Fuel 335, 127042. ammonia at intermediate temperatures. Proc. Combust. Inst. 37, 205–211.
Liu, Z., Zhou, L., Zhong, L., Wei, H., 2023e. Enhanced combustion of ammonia engine Silva, M., Sanal, S., Hlaing, P., Cenker, E., Johansson, B., Im, H.G., 2020. Effects of
based on novel air-assisted pre-chamber turbulent jet ignition. Energy Convers. Geometry on Passive Pre-chamber Combustion Characteristics. SAE Technical Paper.
Manag. 276, 116526. https://doi.org/10.4271/2020-01-0821, 2020-01-0821.
Liu, X., Aljabri, H., Panthi, N., AlRamadan, A.S., Cenker, E., Alshammari, A.T., Silva, M., Mohan, B., Badra, J., Zhang, A., Hlaing, P., Cenker, E., AlRamadan, A.S., et al.,
Magnotti, G., et al., 2023f. Computational study of hydrogen engine combustion 2022. DoE-ML guided optimization of an active pre-chamber geometry using CFD.
strategies: dual-Fuel compression ignition with Port- and Direct-Injection, Pre- Int. J. Engine Res. 24, 2936–2948.
Chamber Combustion, and Spark-Ignition. Fuel 350, 128801. Uddeen, K., Tang, Q., Shi, H., Magnotti, G., Turner, J., 2023. Multiple Spark Ignition
Liu, X., Echeverri Marquez, M., Sanal, S., Silva, M., AlRamadan, A.S., Cenker, E., Approach to Burn Ammonia in a Spark-Ignition Engine: an Optical Study. SAE
Sharma, P., et al., 2023g. Computational assessment of the effects of pre-chamber Technical Paper. https://doi.org/10.4271/2023-01-0258, 2023-01-0258.
and piston geometries on the combustion characteristics of an optical pre-chamber Valera-Medina, A., Xiao, H., Owen-Jones, M., David, W.I.F., Bowen, P.J., 2018. Ammonia
engine. Fuel 341, 127659. for power. Prog. Energy Combust. Sci. 69, 63–102.
Liu, X., Aljabri, H., Allehaibi, M., AlRamadan, A.S., Badra, J., Im, H.G., 2023h. Numerical van Lipzig, J.P.J., Nilsson, E.J.K., de Goey, L.P.H., Konnov, A.A., 2011. Laminar burning
investigation of the effect of injection strategy on a high-pressure isobaric velocities of n-heptane, iso-octane, ethanol and their binary and tertiary mixtures.
combustion engine. Int. J. Engine Res. 24, 595–609. Fuel 90, 2773–2781.
Matthias, N.S., Wallner, T., Scarcelli, R., 2012. A hydrogen direct injection engine Wang, B., Xie, F., Hong, W., Du, J., Chen, H., Su, Y., 2022. The effect of structural
concept that exceeds U.S. DOE light-duty efficiency targets. SAE Int. J. Eng. 5, parameters of pre-chamber with turbulent jet ignition system on combustion
838–849. characteristics of methanol-air pre-mixture. Energy Convers. Manag. 274, 116473.
Metcalfe, W.K., Burke, S.M., Ahmed, S.S., Curran, H.J., 2013. A hierarchical and Wang, Y., Zhou, X., Liu, L., 2023a. Feasibility study of hydrogen jet flame ignition of
comparative kinetic modeling study of C1-C2 hydrocarbon and oxygenated fuels. Int. ammonia fuel in marine low speed engine. Int. J. Hydrogen Energy 48, 327–336.
J. Chem. Kinet. 45, 638–675. Wang, B., Dong, S., Jiang, Z., Gao, W., Wang, Z., Li, J., Yang, C., et al., 2023b.
Niemeyer, K.E., Sung, C.J., 2011. On the importance of graph search algorithms for Development of a reduced chemical mechanism for ammonia/n-heptane blends. Fuel
DRGEP-based mechanism reduction methods. Combust. Flame 158, 1439–1443. 338, 127358.
Niki, Y., Nitta, Y., Sekiguchi, H., Hirata, K., 2019. Diesel fuel multiple injection effects on Wei, F., Wang, P., Cao, J., Long, W., Dong, D., Tian, H., Tian, J., et al., 2023.
emission characteristics of diesel engine mixed ammonia gas into intake air. J. Eng. Visualization investigation of jet ignition ammonia-methanol by an ignition chamber
Gas Turbines Power 141. fueled H2. Fuel 349, 128658.
O’Rourke, P.J., Amsden, A.A., 2000. A Spray/wall Interaction Submodel for the KIVA-3 Xu, L., Bai, X.-S., 2023. Numerical investigation of engine performance and emission
Wall Film Model. SAE Technical Paper. https://doi.org/10.4271/2000-01-0271, characteristics of an ammonia/hydrogen/n-heptane engine under RCCI operating
2000-2001-0271. conditions. Flow Turbul. Combust. https://doi.org/10.1007/s10494-023-00453-y.
Pandey, J.K., Dinesh, M.H., Kumar, G.N., 2023. A comparative study of NOx mitigating Xu, L., Chang, Y., Treacy, M., Zhou, Y., Jia, M., Bai, X.-S., 2023a. A skeletal chemical
techniques EGR and spark delay on combustion and NOx emission of ammonia/ kinetic mechanism for ammonia/n-heptane combustion. Fuel 331, 125830.
hydrogen and hydrogen fuelled SI engine. Energy 276, 127611. Xu, L., Xu, S., Bai, X.-S., Repo, J.A., Hautala, S., Hyvönen, J., 2023b. Performance and
Pearsall, T.J., Garabedian, C.G., 1967. Combustion of Anhydrous Ammonia in Diesel emission characteristics of an ammonia/diesel dual-fuel marine engine. Renew.
Engines. SAE Technical Paper, 670947. https://doi.org/10.4271/670947. Sustain. Energy Rev. 185, 113631.
Redlich, O., Kwong, J.N., 1949. On the thermodynamics of solutions. V. An equation of Yu, L., Zhou, W., Feng, Y., Wang, W., Zhu, J., Qian, Y., Lu, X., 2020. The effect of
state. Fugacities of gaseous solutions. Chem. Rev. 44, 233–244. ammonia addition on the low-temperature autoignition of n-heptane: an
Reiter, A.J., Kong, S.-C., 2011. Combustion and emissions characteristics of compression- experimental and modeling study. Combust. Flame 217, 4–11.
ignition engine using dual ammonia-diesel fuel. Fuel 90, 87–97. Zhang, X., Moosakutty, S.P., Rajan, R.P., Younes, M., Sarathy, S.M., 2021. Combustion
K. Richards, P. Senecal, E. Pomraning,CONVERGE (v3.0), Madison (WI): Convergent chemistry of ammonia/hydrogen mixtures: jet-stirred reactor measurements and
Science. comprehensive kinetic modeling. Combust. Flame 234, 111653.
Schmidt, D.P., Rutland, C.J., 2000. A new droplet collision algorithm. J. Comput. Phys. Zhang, X., Yalamanchi, K.K., Mani Sarathy, S., 2023. Combustion chemistry of ammonia/
164, 62–80. C1 fuels: a comprehensive kinetic modeling study. Fuel 341, 127676.
Senecal, P.K., Pomraning, E., Richards, K.J., Briggs, T.E., Choi, C.Y., McDavid, R.M., Zhao, D., An, Y., Pei, Y., Shi, H., Wang, K., 2023. Numerical study on the asymmetrical
Patterson, M.A., 2003. Multi-dimensional Modeling of Direct-Injection Diesel Spray jets formation from active pre-chamber under super-lean combustion conditions.
Liquid Length and Flame Lift-Off Length Using CFD and Parallel Detailed Chemistry. Energy 262, 125446.
SAE Technical Paper. https://doi.org/10.4271/2003-01-1043, 2003-01-1043. Zhou, C.-W., Li, Y., Burke, U., Banyon, C., Somers, K.P., Ding, S., Khan, S., et al., 2018. An
Shah, A., Tunestal, P., Johansson, B., 2014. Effect of Relative Mixture Strength on experimental and chemical kinetic modeling study of 1,3-butadiene combustion:
Performance of Divided Chamber ‘Avalanche Activated Combustion’ Ignition ignition delay time and laminar flame speed measurements. Combust. Flame 197,
Technique in a Heavy Duty Natural Gas Engine. SAE Technical Paper. https://doi. 423–438.
org/10.4271/2014-01-1327, 2014-01-1327. Zhou, L., Zhong, L., Liu, Z., Wei, H., 2023. Toward highly-efficient combustion of
ammonia–hydrogen engine: prechamber turbulent jet ignition. Fuel 352, 129009.
16