Ammonia is a promising carbon-free fuel for internal combustion engines, yet its low flame speed and poor ignitability severely restrict practical application. To address these challenges, this study proposes a novel ammonia combustion mode based on in-cylinder high-pressure hydrogen direct injection coupled with spark-induced jet ignition. Compared with pre-chamber combustion systems, this approach offers superior stability and greater potential for achieving higher combustion efficiency, benefiting from advantages including multi-pulse ignition capability, pre-injection of hydrogen, stratified combustion, and avoidance of cold-start issues. A systematic experimental investigation is performed in a constant volume combustion vessel to comprehensively explore the combustion behaviors of premixed ammonia ignited by high-pressure hydrogen jet. Four key parameters are varied: hydrogen injection pressure (3.0-9.0 MPa), ammonia equivalence ratio (0-1.0), ignition timing (2.5/4.0 ms after start of injection), and ignition position (six axial/radial locations). Key findings reveal that ignition during hydrogen injection promotes flame kernel development over post-injection ignition, though positions away from nozzle/axis are required to avoid jet-induced extinction. Higher injection pressure enhances hydrogen supply and turbulence, accelerating early heat release. Lean ammonia mixtures benefit initial ignition via higher local oxygen, while late-stage performance is governed by ammonia concentration. Stoichiometric ammonia and higher pressure reduce unburned NH3. Notably, this novel mode achieves ~96% combustion efficiency when ammonia energy ratio of 99.3%. Further improvements in ammonia energy ratio and combustion efficiency could be expected in practical engines by adopting strategies such as early hydrogen injection and multi-stage injection.

Improving the thermal-efficiency limit of ammonia-diesel dual-fuel (ADDF) marine engines is critical for low-carbon and practically deployable marine propulsion. This study presents an integrated experimental-numerical framework that combines a calibrated GT-POWER ADDF combustion model with a MATLAB-based Organic Rankine Cycle (ORC) model to quantify waste-heat-recovery effects under explicitly defined energy-boundary assumptions. The combustion model was calibrated using SJTU95 ADDF data and then transfer-guided and checked against V240 engine constraints over a wide range of ammonia energy ratios (0–83% AER); The SJTU175 platform was used for engine-ORC calibration. The engine outputs supplied to the ORC model include exhaust mass flow, exhaust temperature/enthalpy, jacket-water heat transfer and coolant temperature under different engine loads and AERs. The calibrated ADDF model predicts indicated thermal efficiency and peak cylinder pressure within ±5% of the available V240 measurements across the investigated AER range, while the ORC unit delivers 105.3–145.5 kW of net recovered power, corresponding to cycle efficiencies of 8.8%–9.3% and total system output gains of 5.2%–7.2%. At 65% AER, the combined engine-ORC system reaches a thermal efficiency of 37.6%, corresponding to 91.0% of the pure-diesel baseline, while at 83% AER, it retains 86.8% of the baseline performance. These results define a realistic thermal-efficiency benchmark for ORC-assisted ADDF marine engines and show that waste heat recovery can substantially mitigate, though not fully eliminate, the efficiency penalty associated with high ammonia substitution.

Given ammonia's low chemical reactivity and high resistance to ignition, strategies to enhance global mixture reactivity hold promise for improving ammonia combustion performance and reducing emissions. This study explores a novel approach where ammonia and diesel are premixed during the early intake stroke and undergo auto-ignition near top dead center (TDC), marking the first systematic experimental investigation of ammonia-diesel homogeneous charge compression ignition (HCCI) combustion on a 4-cylinder engine test bench. The results demonstrate that in ammonia-diesel HCCI mode, combustion stability governs the achievable ammonia energy ratio limit, while in-cylinder pressure rise rate defines the engine load boundary. A significant limit of ammonia energy ratio at mid-to-high engine loads was found, probably characterized by the reactivity balance of ammonia and diesel. At higher engine loads, substantial energy release occurs well before TDC, elevating combustion temperatures. This effectively suppresses unburned NH3 and N2O emissions but simultaneously increases heat loss, NOx emissions, and pressure rise rates. Increasing ammonia fraction reduces global chemical reactivity, thereby delaying low-temperature reactions. Thermal efficiency improves with ammonia addition owing to reduced heat transfer losses. When ammonia exceeds 50% of the energy share, the low-temperature reaction (LTR) diminishes significantly, becoming negligible and eliminating the characteristic two-stage heat release pattern. In addition, a pronounced negative temperature coefficient (NTC) effect is observed at ammonia energy ratios of 20% and 30%. Emissions analysis reveals a strong proportional correlation between unburned NH3 and N2O. Optimizing NOx-N₂O trade-offs requires precise control of engine load while maintaining an intermediate level of ammonia energy share. At around 50% ammonia substitution, the thermal efficiency of ammonia–diesel dual-fuel HCCI combustion becomes comparable to, or may even exceed, that of conventional dual-fuel compression ignition (CI) modes. This study reveals favorable thermal efficiency at 1000 rpm over an ammonia energy ratio range of 40% ∼ 60%, corresponding to an engine load of approximately 35% ∼ 60%.

The interface between the proton exchange membrane (PEM) and the catalyst layer (CL) plays a critical role in determining the electrochemical kinetics and overall efficiency of proton exchange membrane fuel cells (PEMFCs). This study presents a facile and scalable "etched glass templating" strategy to fabricate surface-textured, ePTFE-reinforced PEMs, aiming to optimize this critical interface. Diverging from conventional planar casting methods, perfluorosulfonic acid (PFSA) dispersion is cast onto etched glass substrates to engineer membranes with controlled surface roughness. Morphological analysis confirms the successful transfer of distinct microstructures from the templates to the membranes, effectively transforming the cathode interface from a conventional 2D plane into a complex 3D surface. Crucially, mechanical and swelling analyses confirm that this templating approach preserves the intrinsic tensile strength and dimensional stability of the composite membranes, thereby ensuring their robustness for practical applications. Electrochemical evaluations reveal that the textured PEMs significantly outperform their planar counterparts (P-PEM). Specifically, the membrane electrode assembly (MEA) incorporating the optimized surface-textured PEM (T1-PEM) exhibits a substantial 35.6% increase in electrochemically active surface area (ECSA) and a 29% enhancement in peak power density. These significant improvements are attributed to the formation of a 3D interlocking PEM/CL interface, which enhances the utilization of catalyst active sites and significantly promotes charge transfer kinetics. In essence, this research validates the etched glass substrate coating method as a scalable and cost-effective PEM preparation strategy to significantly enhance MEA electrochemical performance, thereby offering a robust pathway for the development of high-performance fuel cell technologies based on next-generation PEMs with optimized 3D interfaces.

Turbulent jet ignition technology demonstrates great potential in achieving future emissions standards. The key prechamber structural parameters, including volume, orifice diameter, and orifice number, play critical roles in jet ignition and combustion. However, most researches considered these parameters individually, overlooking comprehensive effects on turbulent jet ignition. This study employs a constant volume combustion chamber with a prechamber system to investigate the co-effect of structure parameters (volume, orifice diameter and number) on jet ignition and combustion performance. The results show that the combustion duration increases with increasing orifice number at same orifice diameter but decreases when maintaining same total orifice crosssectional area, which is associated with the ignition behavior (jet ignition or flame ignition) in this study. Overall, both ignition delay and combustion duration are influenced by multiple prechamber structure parameters, exhibiting a linear relationship with the ratio of volume to total orifice cross-sectional area (V/A), however, there is some variation in the combustion characteristics for different prechamber structures at the same V/A. Furthermore, Damko & uml;hler number (Da) is introduced to further elucidate the relationship between structure parameters and combustion characteristics. The ignition delay shows a negative correlation with Da, however, a positive correlation is observed between combustion duration and Da. Overall, Da decreases with increasing V/A. In addition, the effect of prechamber volume on ignition delay and combustion duration remains independent of Da, however, the effect of volume is not evident on the combustion duration for larger orifice. These results highlight the importance of the co-effects of multiple structural parameters on ignition and combustion characteristics, providing guidance for structure optimization to achieve more efficient prechamber system.

Understanding the transient heat transfer of wall-impinging spray flames is essential for improving thermal efficiency in compression-ignition (CI) engines. This study presents simultaneous measurements and analyses of radiative and convective heat transfer under engine-like conditions using a constant-volume combustion chamber. High-speed two-color pyrometry and fast-response coaxial thermocouples captured flame-temperature distributions and localized wall heat fluxes. A view-factor-based radiation reconstruction is introduced to derive wall-averaged radiative heat fluxes. The experimental matrix varied ambient density and temperature, injection pressure, and injection duration. Results reveal that radiation contributes 5-11 % of the total wall heat flux depending on operating conditions. Higher ambient density and longer injection durations enhanced radiative contributions due to prolonged high-temperature soot-core formation, whereas higher injection pressures intensified convection and reduced the relative radiation fraction.

Prechamber structure has significant influence on the jet ignition phenomenon, thereby directly affecting the overall performance and efficiency of natural gas engines. However, the correlation between the prechamber structural features and ignition phenomenon has not yet been clearly elucidated. This work employs a combined experimental and computational approach in a constant volume combustion chamber (CVCC) with single-hole prechamber system to investigate the ignition phenomena and transition process of jet ignition and jet flame ignition. The investigation reveals the prechamber structural control mechanisms governing the ignition phenomenon in lean methane/air ambience, where both the orifice diameter and prechamber volume were varied. The results show that the smaller hole diameter produces a turbulent jet with lower temperature and active radical concentration, promoting a transition from flame to jet ignition and consequently delaying combustion. However, the volume has less effect on the ignition phenomenon. Moreover, the investigation of both ignition phenomena reveals that the initial flame formation consistently occurs in the upstream region of the jet, with subsequent propagation toward the downstream region driven by the jet development. Finally, the turbulent jet ignition coefficient (k) is proposed to characterize the transition of the jet ignition to jet flame ignition. The k represents the ratio of the characteristic reaction time t to the ignition delay tau T, which is related to initial jet velocity, ambient temperature of main chamber and adiabatic flame temperature of prechamber mixture. In the initial stage of the jet, the k increases faster as the prechamber hole diameter increases, in favour of forming the jet flame ignition. This work provides guidance for the design of prechamber structure, which contributes to the development of more efficient and reliable prechamber engines.

Low-carbon gas engines show great potential in future marine propulsion and power generation applications. Passive pre-chamber jet ignition technology is one of an important approach to enhance ignition reliability and enabling stable lean combustion. Although extensive research has been conducted on pre-chamber jet ignition, most studies have focused on main chamber ignition experiments without decoupling the dynamics of hot turbulent jets from subsequent combustion processes. As a result, the influence of early jet evolution on later flame development remains insufficiently understood. In this work, a novel decoupling methodology is proposed to isolate and analyze early jet evolution independently of subsequent ignition and combustion phases. The optical diagnostics for high-speed photography of the Z-type schlieren image system is established. By comparing jet penetration and morphological characteristics under different ambient densities, which demonstrates that jet development remains consistent between inert and flammable main chamber environments when densities are similar. This finding validates the feasibility of decoupling early jet development from later combustion events. Based on this, the study systematically investigates jet characteristics and flame development, revealing a strong correlation between ignition delay and early jet morphology. Notably, under lean conditions (λ = 1.4), the propagation rate of the jet area significantly influences the axial flame development rate, while under near-stoichiometric conditions (λ = 1.2), radial flame propagation is more closely governed by early jet behavior. These insights provide a foundational understanding of the coupled mechanisms governing turbulent jet ignition and offer new perspectives for optimizing pre-chamber designs in next generation low-carbon engines.

The growing pursuit of carbon-neutral fuels has spotlighted liquid ammonia as a zero-carbon energy carrier for engines and gas turbines. However, its low boiling point causes intense flash boiling upon injection, drastically altering spray behavior compared to conventional non-flashing fuels like methanol. This study experimentally examines how a crossflow air stream affects a superheated liquid ammonia jet, in comparison to a nonsuperheated methanol jet. High-speed diffused back-illumination in a wind tunnel was used to capture spray morphology while varying injection pressure and crossflow velocity over a broad range of momentum flux ratios (q) and aerodynamic Weber number (We). The results reveal that under intense superheat (high Rp), ammonia sprays exhibit a radically different trajectory: the jet undergoes near-instantaneous explosive breakup into a vapor-liquid plume, yielding much shorter penetration and greater lateral deflection than the methanol sprays. Flash boiling enhances atomization (producing finer, more uniformly distributed droplets) but also weakens the jet's resistance to the crossflow, making the ammonia spray penetration relatively insensitive to increased injection momentum. In contrast, methanol jets follow classical shear-driven behavior, with higher q yielding deeper penetration and less deflection. Mechanistically, the ammonia jet's fragmentation is governed by internal vapor generation (superheat-driven) rather than external aerodynamic forces, decoupling its atomization from crossflow effects. These insights provide guidance for ammonia-fueled combustors: controlling the fuel's thermal conditions (e.g. injection temperature or ambient pressure) to moderate superheat is key to optimizing spray penetration and dispersion in crossflow environments, which is crucial for efficient and stable combustion in engines and turbines.

Ammonia-diesel dual-fuel compression ignition engines have been recognized as a promising solution for achieving low or zero carbon emission targets in the maritime industry. However, the exhaust gas emissions of ammonia fueled engines including unburned NH3, N2O and NOx are unusually higher than traditional diesel engines, which should be controlled. Unfortunately, the formation mechanisms of these emissions in ammonia fueled engines remain unclear. This study employs the combined approach of using numerical simulations and experiments, as well as the dynamic phi-T map analysis to investigate the in-cylinder distributions and formation mechanisms of unburned NH3, N2O, and NOx with respect to mixture reactivity, equivalence ratio, ambient temperature and pressure. The result shows that the nitrogenous emissions are primarily associated with the turbulent flame propagation in the premixed mixture of 100 % ammonia energetic ratio (AER) regions. While N2O primarily forms at the NH3 flame front, and resides at the unburned NH3 boundary zones, particularly in regions like the center of combustion chamber and near the cylinder walls. Unburned NH3 is found to accumulate in areas inaccessible to the flame, such as piston ring crevices and vicinity of combustion chamber walls, while NOx emissions predominantly occur in the burned zone behind the flame front of ammonia combustion and 72.5 % of the NOx emissions comes from the fuel ammonia. The formation and consumption of NOx, N2O, and unburned NH3 emissions are determined by local variations of equivalence ratio, temperature, and AER in the cylinder. By combining the dynamic phi-T maps with CFD simulations, this study captures the emission formation processes and identifies the specific equivalence ratio-temperature generation regions of NOx, N2O, and unburned NH3.

Ammonia is expected to become an alternative fuel for internal combustion engines. Compared with hydrocarbon fuels, ammonia has a higher latent heat of evaporation, and its high-pressure direct injection causes severe cooling that may induce condensation and affect unsteady combustion. This study aims to elucidate the cooling characteristics of liquid ammonia spray and compare them with those of diesel under engine-like conditions. First, an efficient Eulerian-Lagrangian framework was developed, in which a tabulated real-fluid equilibrium solver was developed to check the phase stability and compositions, and a condensation model was introduced to compute the condensation process. Both the condensation and spray models were validated against experimental data. Subsequently, extensive spray simulations for diesel and liquid ammonia were performed under various conditions, followed by a detailed thermodynamic analysis. Ammonia, with a lower critical point, makes its spray less prone to condensation compared with diesel spray. However, once the ammonia vapor is supercooled, its condensed mass is significantly greater than that of diesel spray owing to its significant vapor gradient potential. Moreover, owing to ammonia’s low critical temperature, its spray remains in “hot-state”, and its condensation is “pressure-dominated”, which is less sensitive to the variation of ambient temperature. In contrast, diesel, being in “cold-state”, needs to be heated to a higher temperature to trigger rapid evaporation, so its condensation is “temperature-dominated” and occurs at a location far from the nozzle exit. The difference in the thermodynamic properties of the two fuels results in markedly distinct spray condensation characteristics.

For air-independent propulsion (AIP) systems, hydrogen combustion with noble gases offers zero emissions and high efficiency, making understanding hydrogen injection into noble gases essential. This study visualizes highpressure (3-9 MPa) hydrogen jets in air, Ar-O-2, and He-O-2 environments (0/10/21% O-2) using schlieren imaging. Jet penetration is found to be governed by injection-to-ambient pressure ratio (nPR), mean relative molecular mass of the ambient gas mixture (M-a), and time. Conversely, jet angle (30 degrees-36 degrees) is observed to be primarily influenced by nPR, not ambient gas type. In He-21%O-2, penetration is over 15% greater than in air and Ar-21%O-2, peaking above 18%. This difference amplifies with decreasing oxygen concentration. A physics-based correlation, jet penetration proportional to (nPR - 1)/Ma)(0.25) & sdot;t (0.5)(SOI), accurately captures the data trends. Furthermore, an Auto-ML model is trained and predicts penetration for unseen conditions with R-2 > 0.99. These insights guide injection and mixture optimization for hydrogen-fueled AIP systems with noble gases.

Hydrogen-assisted pre-chamber jet ignition effectively overcomes the low reactivity challenge that hinders the practical application of carbon-free ammonia in engines. However, fundamental optical studies on the effects of pre-chamber mixture reactivity and nozzle orifice orientation on hydrogen-assisted ammonia combustion remain limited. In this study, the effects of the pre-chamber mixture reactivity, i.e., the excess air ratio and oxygen concentration, and the configuration of radial and tangential orifice nozzles on the ignition and combustion behaviors of ammonia initiated by jet ignition are optically investigated using Schlieren imaging. The findings reveal that the excess air ratio of the pre-chamber mixture markedly influences the evolution of hot jets and ignition behavior of secondary hot jets. Increasing the excess air ratio from 1.50 to 2.00 results in a delayed hot jet ejection and a reduced penetration velocity. The balance between jet velocity and reactivity enables localized re-ignition in the main chamber at an excess air ratio of 1.75. Moreover, enriching the pre-chamber with oxygen significantly affects both jet penetration and ignition behavior. Oxygen enrichment markedly advances the ignition timing initiated by secondary hot jets. The combustion duration exhibits a decrease-to-increase pattern due to the hot jet characteristics. Although oxygen-enriched combustion leads to increased NOx formation, it can effectively reduce unburned NH3 and N2O emissions. Regarding nozzle orifice orientation, the tangential-orifice design promotes earlier ignition than the radial configuration. However, hot jets from the tangential orifices result in longer combustion duration and higher unburned NH3 emissions owing to their reduced penetration velocity.

Current combustion chamber designs are rarely tailored for ammonia-fueled engines and rely heavily on empirical experience, while existing simulation studies are seldom conducted under optimal compression ratios and lack precise modeling for complex geometries as well as efficient optimization methods. To enhance the performance and emission characteristics of an ammonia-diesel dual-fuel (ADDF) engine, this study established a parameterized combustion chamber model. Under the previously determined optimal compression ratio, the combustion chamber was efficiently modified via a MATLAB program for 3-D simulation. Subsequently, the simulation data were utilized to train predictive regression models, followed by the application of the NSGA-III optimization algorithm to further optimize the combustion chamber. The results demonstrate that the maximum cylinder pressure constraints necessitate a retarded injection timing at high compression ratios, thereby increasing the greenhouse gas emissions. However, by adjusting the valve timing and throttle angle to reduce the excess air ratio, the potential of combustion chamber optimization can be fully realized. By appropriately guiding the development of the diesel spray, the optimized combustion chamber enhances in-cylinder mixture quality and mitigates pollutant accumulation. At the maximum brake torque condition, the indicated thermal efficiency is improved by 5% compared to the base engine. Simultaneously, unburned ammonia and N2O emissions are reduced by 62% and 31% respectively, with a more than 50% reduction in NOX emissions. Additionally, this study reveals the pathways through which combustion chamber geometry dictates in-cylinder combustion and pollutant emissions by modulating the flow intensity distribution across different cross-sections. Building upon previous findings, this study also validated the feasibility of integrating cross-platform simulations with predictive regression models to co-optimize the compression ratio and combustion chamber geometry of ADDF engines under maximum cylinder pressure constraints.

Ammonia/diesel dual-fuel (ADDF) engines that use port injection of low-pressure ammonia typically suffer from low combustion efficiency and poor unburned ammonia (unburned NH3) emissions when operating at high ammonia energy ratio (AER). An intake pre-injection diesel strategy was introduced to enhance the chemical reactivity of the premixed ammonia mixture, while pilot diesel injection near the top dead center (TDC) was employed to complete the ignition process. The effects of diesel pre-injection timing (DPI) and start of pilot injection timing (SOPI), pre-injection diesel energy ratio (PDER) in total diesel fuel, and AER on combustion and emission characteristics were systematically investigated. The high-load operating boundary, limited by pressure rise rate (PRR) and combustion stability, was further explored. The results show that a DPI timing of-342 degrees crank angle after top dead center (degrees CA ATDC) has the most favorable thermal efficiency. There is a sudden change in the pressure rise rate during the variation of AER (from 60% to 70% AER) which may be related to changes in the combustion mode. Advancing SOPI timing not only increases thermal efficiency but also reduces unburned ammonia and greenhouse gas (GHG) emissions. However, excessive advancement of SOPI timing (e.g.-14 degrees CA ATDC) leads to unstable combustion. Additionally, advancing SOPI timing at low AER is also limited by the maximum PRR. A PDER of 40% results in the lowest unburned NH3 emissions and the highest thermal efficiency. Compared with compression stroke pre-injection diesel strategy, the intake pre-injection diesel strategy improves thermal efficiency by approximately 1.8%, while reducing GHG emissions and unburned NH3 about 7.1% and 16.7%.

Investigation on the correlation between wall heat transfer and near-wall quenching contributes to the understanding of the generation mechanism of methane slip in lean LNG engines. Near-wall quenching characteristics of lean to ultra-lean methane flames (phi = 0.55, 0.50, 0.44) with hydrogen enrichment (the hydrogen ratio alpha H2 = 0-50% for phi = 0.50, 0.55 and alpha H2 = 15-75% for phi = 0.44) are experimentally studied at elevated ambient pressures (Pa = 2.0 MPa). Quenching distances of lean methane-hydrogen-air flames are measured, and limits of quenching distances are determined. The quenching distance reaches its limit when alpha H2 = 15% for phi = 0.55 and alpha H2 = 15-30% for phi = 0.50, while the trend for phi = 0.44 is different. The correlation between wall heat transfer and near-wall flame quenching is discussed based on dimensionless analysis of the wall heat transfer characteristic number, the normalized wall heat loss, psi, and the quenching Peclet number, Peq, and a novel psi-Peq correlation is proposed for lean methane-hydrogen-air flames as psi = 10.864 center dot Peq+0.109 , where f1 (phi, alpha H2 ) is a function of the equivalence ratio, phi, and the hydrogen ratio, alpha H2 , and f2 ( Leeff) is a function of the effective Lewis number of the gas mixture. The laminar flame speed (ul) of methane-hydrogen-air flames increases at elevated ambient pressures, and the dependence of ul on phi and alpha H2 is described by a physical model. The pressure effect on the laminar flame speed (ul) and the quenching distance (delta q) is investigated at varied ambient pressures (Pa = 0.5, 1.0, 2.0, 3.0 MPa) to better characterize the near-wall flame quenching at elevated ambient pressures.

This article provides a comprehensive synthesis of the current state of research on fault detection and diagnosis of marine diesel engines, organized along three main threads: fault taxonomy, data acquisition and generation pathways, and diagnostic approaches. Firstly, guided by the combustion functional chain and key components, a hierarchical modeling of fault modes is developed to construct a “system–component–fault” fault taxonomy. Secondly, fault-injection and parameter-equivalent simulation strategies are systematically consolidated, revealing a structural landscape in which bench-test data with controllable fault injection constitutes the primary evidence base, simulation-generated data serves as important complements, while in-service data remains underrepresented and public datasets are scarce. Subsequently, the methodological landscape of marine diesel engine diagnosis and its evolutionary trends are summarized, showing that data-driven approaches dominate, model-based and hybrid approaches provide critical support, and knowledge-based approaches play a supplementary role. Finally, future research directions for this field are discussed.

Pre-chamber turbulent jet ignition holds significant potential for improving ammonia combustion. Hydrogenand oxygen-enriched combustion in the pre-chamber can further enhance ammonia combustion. However, fundamental research on this enhanced combustion mechanism remains limited. This study systematically examines the individual impacts of oxygen-enriched combustion in a hydrogen-enriched multi-orifice pre-chamber and nozzle geometry specifications on jet behavior in a nonreactive environment, as well as the impact of the jet behavior on ammonia ignition behavior, combustion processes, and emissions in a reactive environment. The results show that oxygen enrichment significantly enhances ignition characteristics, with three ignition modes observed as the oxygen concentration increases from 30 % to 70 % by volume: localized re-ignition, jet-induced secondary ignition, and jet flame ignition. With increasing oxygen concentration, combustion duration decreases initially and then increases. Unburned NH3 and N2O emissions decrease, while NOx emissions slightly increase with oxygen enrichment. In the pre-chamber nozzle design, two nozzles with identical cumulative orifice areas, i. e., six 1.50 mm orifices and three 2.12 mm orifices, demonstrate that larger orifices improve ammonia ignition. On the other hand, when both nozzles have an identical orifice diameter, the nozzle featuring a smaller cumulative orifice area enhances ignition behavior as it generates higher-velocity hot jets. For two nozzles featuring the same number of orifices and similar hot jet velocities, the nozzle featuring the larger orifice demonstrates superior ignition behavior. These findings could offer valuable insights into enhancing ammonia combustion and optimizing nozzle design.

Under the 2050 net-zero emissions target, the maritime sector considers ammonia, a carbon-free fuel, to be a highly promising alternative. Scaled-model experiments following similarity theory play a crucial role in minimizing cost, energy, and time in the development of new engines. Unfortunately, although it has been conducted on different-sized diesel engines, no information is available on the accelerated development of ammonia engines. In this work, the single-valued condition and similarity law for scaled-model experiments of ammonia engines are summarized, primarily focusing on the high-pressure direct-injection mode. After confirming the accuracy of numerical simulations using experimental data from liquid ammonia sprays and ammonia engines, the potential of scaled-model experiment is studied using two ammonia engines with bore diameters of 175 mm (i.e. the large engine) and 95 mm (i.e. the small engine) under various engine speeds and 90 % ammonia energetic ratio. The results show that spray development, heat release rate, in-cylinder pressure and temperature, indicated thermal efficiency, NOx and N2O emissions are highly similar between different-sized engines, indicating the effectiveness of similarity theory in facilitating new ammonia engine development. In particular, based on the similarity law summarized in this study, the differences in peak in-cylinder pressure and temperature between the large and small engines are smaller than 2 %, while the difference in indicated thermal efficiency is smaller than 5 %. However, due to the inherent difference in surface area-to-volume ratio between the large and small engines, the small engine experiences increased heat transfer losses and reduced temperatures near the wall, leading to a higher proportion of unburned ammonia residue in the near-wall region. The above results are deemed invaluable for the intensive development of ammonia engines of different sizes.