This comprehensive review examines the application of liquid-ammonia injection and combustion in engine systems, highlighting the potential of liquid ammonia as a carbon-neutral fuel alternative. The study synthesizes recent advancements in liquid-ammonia injection and combustion technologies, addressing critical domains such as fundamental fuel properties, injection and spray dynamics, combustion behavior, and engine performance. Key challenges are identified, including ammonia’s high latent heat of vaporization, slow flame-propagation speed, narrow flammability range, and elevated NOx emissions, while emphasizing the need for optimized injection strategies and nozzle designs to enhance atomization and mixing. The research findings indicate that liquid-ammonia injection can significantly reduce greenhouse gas emissions, with dual-fuel modes (e.g., ammonia-diesel) proving effective in overcoming ammonia’s low reactivity. Studies show that both low-pressure and high-pressure dual fuel-injection modes can achieve substantial emission reductions, with high-pressure injections offering better thermal efficiency and lower NOx emissions. Innovative approaches, such as turbulent jet ignition, stratified fuel injection, and hydrogen co-injection, have been explored to improve ignition efficiency and combustion stability. Future research should prioritize the development of integrated solutions that combine advanced combustion technologies, optimized engine designs, and effective emission-control strategies. Collaboration between academia, industry, and policymakers will be crucial in driving the adoption of ammonia as a sustainable fuel alternative.
Highlights the challenge of developing truly calibration-free models, especially for new fuels like ammonia whose properties differ significantly from those of hydrocarbons
Hao Wu, Fahad Almatrafi, Moez Ben Houidi, Tiegang Fang, William L. Roberts.
A Review on Liquid-Ammonia Injection and Combustion for Engine Applications.
Engineering, 2026, 59(4): 82-117 DOI:10.1016/j.eng.2025.09.008
The adoption of carbon-free energy sources and vectors has emerged as a pivotal strategy to mitigate carbon emissions and address climate-related challenges [1], [2], [3]. In recent years, ammonia (NH3) has garnered significant attention as a potential energy source due to its unique properties and the urgent need for sustainable alternatives in the global energy transition [4,5]. Historically, ammonia has been predominantly used as a fertilizer, with about 70% of global production allocated to this purpose [6]. Traditional ammonia production relies heavily on the Haber-Bosch process, which is highly energy-intensive, accounting for about 2% of global energy consumption and approximately 450 Mt of carbon dioxide (CO2) emissions in 2020 [7]. The focus on ammonia has driven a shift toward “green ammonia” production, which utilizes renewable electricity, water, and air to produce hydrogen for synthesizing ammonia, thereby reducing the latter’s carbon footprint [8], [9], [10], [11].
Fig. 1 highlights the role of ammonia as an energy carrier in a future energy system. Ammonia has a high volumetric (108 kg·m−3 H2 in liquid NH3 at 20 °C and 8.6 bar) and gravimetric (17.8 wt%) energy density [12], making it an efficient energy-storage medium. It also offers a lower cost per unit of energy stored (0.54 USD·kg−1 H2, 182 days) [13], compared with hydrogen (14.95 USD·kg−1 H2). Unlike hydrogen, ammonia can be stored and transported relatively easily at reasonable temperatures and pressures, making it particularly attractive for maritime shipping and long-term energy storage [14,15]. Additionally, ammonia can be catalytically decomposed into COx-free hydrogen [16], [17], [18] or directly used in power-generation devices such as engines, turbines, and fuel cells [19,20]. Furthermore, ammonia has been manufactured and transported globally for more than a century, benefiting from an existing high-capacity infrastructure [21]. The shipping industry, in particular, views ammonia as a promising alternative to traditional marine fuels, given its potential to reduce emissions [22].
The use of ammonia as a fuel traces its origins to the early 20th century, with notable experiments conducted during the 1940s and later periods [23], [24], [25], [26], [27]. However, recent advancements in renewable energy technologies and the growing emphasis on sustainability have rekindled interest in ammonia as an energy source. By 2050, it is projected that over 50% of global ammonia manufacturing capacity will be dedicated to energy decarbonization [28]. Despite its potential, ammonia’s widespread adoption still presents challenges, including high green-production costs, the need for technology to address renewable-energy intermittency [29], and safety concerns regarding its hazardous nature [6]. Ongoing research and development and robust policy support are essential to overcome these barriers and realize ammonia’s potential as a sustainable energy carrier.
1.2. Ammonia as a sustainable fuel
Ammonia presents several key advantages as a fuel in transportation and industrial processes [30]. As a carbon-free compound composed only of nitrogen and hydrogen, ammonia produces no direct CO2 emissions when combusted, making it an attractive option for reducing greenhouse gas (GHG) emissions [31]. Its liquefiability at relatively modest pressures and temperatures makes it practical for storage and transportation, offering superior energy density and manageable infrastructure requirements [32]. Additionally, ammonia’s versatility allows it to be utilized across multiple sectors, including maritime transport, power generation, and industrial processes, with the potential for “green ammonia” production using renewable electricity [33]. Its versatility in application methods, ranging from direct use in modified gas turbines to co-firing with traditional fuels, offers flexible pathways for adoption while leveraging existing infrastructure [21].
Ammonia shows significant promise as a fuel across various sectors, with particular applicability in maritime transportation [34,35]. Its low reactivity and slower burn rate suit low-speed engines, enabling ocean vessels to transition from carbon-intensive bunker fuel to ammonia power, thereby reducing shipping’s environmental impact. Several major shipping companies and engine manufacturers have already begun developing and testing ammonia-powered marine engines [36], [37], [38], [39], [40]. In the power-generation sector, ammonia can be utilized in modified gas turbines or fuel cells. Some power companies are synchronously exploring co-firing ammonia with coal or natural gas in existing power plants as a transitional solution [41], [42], [43]. Industrial processes are another key application, where ammonia serves as both fuel and chemical feedstock. The fertilizer industry’s extensive ammonia-handling experience provides a natural expansion starting point. Research continues into stationary power applications and specialized vehicles, although vehicular use presents greater technical and safety challenges due to ammonia’s toxic nature [5,13].
Ammonia combustion is a complex process with unique characteristics that make it both promising and challenging. Table 1 [44], [45], [46], [47] summarizes the fundamental physical and chemical properties of ammonia compared with those of other commonly used fuels. Ammonia is characterized by its well-known low reactivity, which results in poor ignition characteristics and a narrow flammability range (15%-28% by volume in air) [48,49]. Its minimum ignition energy is significantly greater than that of hydrogen and other hydrocarbon fuels, making it difficult to ignite under normal conditions. This poses a particular challenge for internal-combustion engines, where rapid and reliable ignition is essential for stable operation [50,51]. However, ammonia has a high-octane number of 110, indicating its superior anti-knock properties compared with gasoline [44]. The narrow flammability range means that ammonia combustion is more sensitive to variations in the air-to-fuel ratio; thus, maintaining stable combustion becomes more challenging [52]. As a result, precise control of the fuel injection and mixing processes is crucial for efficient ammonia combustion. Another significant challenge associated with ammonia combustion is its slow flame-propagation speed [53], which makes it difficult for the flame to propagate efficiently through the fuel mixture. This results in longer combustion duration and lower thermal efficiency. Additionally, the adiabatic flame temperature of ammonia is only approximately 1800 °C, lower than those of hydrogen, natural gas, and other traditional liquid fuels [54]. This lower temperature, combined with the absence of CO2 in the exhaust, reduces radiative heat transfer and causes combustion delay [55].
In addition to these technical challenges, ammonia combustion produces unique emissions requiring careful management. Ammonia’s high nitrogen content leads to increased NOx emissions during combustion [44,48]. Specifically, both nitric oxide (NO) and nitrous oxide (N2O) are formed during ammonia combustion, with N2O being particularly problematic due to its high global warming potential (GWP) (265-298 times that of CO2) [56]. Moreover, incomplete ammonia combustion can result in significant unburned NH3 emissions, posing environmental and health risks [57]. To mitigate these emissions, various strategies have been explored, including optimizing combustion conditions, using exhaust-gas recirculation (EGR), and employing selective catalytic reduction (SCR) systems [58]. While the characteristics of ammonia combustion pose significant challenges for its practical application, ongoing research and technological advancements are gradually overcoming these obstacles. Given its potential to contribute to a low-carbon future, ammonia combustion remains an area of active investigation and development.
1.3. Overview of liquid-ammonia injections and applications
Liquid ammonia presents numerous advantages as a combustion fuel, making it a practical option for a wide range of applications [59], [60], [61]:
Higher volumetric energy density. Liquid ammonia has a significantly higher volumetric energy density than gaseous ammonia. Thus, liquid ammonia can store more energy in a given volume, making it more efficient for transportation and storage. This is particularly advantageous in applications with limited space and infrastructure.
Simplified supply system. Using liquid-phase ammonia at terminals eliminates the need for additional devices (i.e., vaporizers, electric heaters, and accumulators) for gaseous conversion. This full-chain liquid-ammonia path from supply to application reduces evaporation-related energy consumption. Liquid ammonia also enables more accurate fuel metering and flow control than gaseous ammonia.
Compatibility with direct-injection systems. Direct injection (DI) of liquid fuels is well-established in internal-combustion engines and gas turbines. High-pressure liquid ammonia can utilize existing injectors, enabling liquid-ammonia DI. This method is a primary approach for investigating liquid-ammonia spray and combustion, given the absence of mature ammonia-specific injectors. However, the use of conventional gasoline and diesel direct injectors for liquid-ammonia injections poses reliability concerns.
Cost efficiency. Liquid ammonia can be transported using existing infrastructure—such as pipelines, trucks, and ships—with minimal modifications. Storage and transportation costs for liquid ammonia are lower than those for gaseous ammonia. This can make liquid ammonia more economically viable as a fuel, particularly in applications where fuel must be transported over long distances.
Fig. 2 maps liquid-ammonia injection test conditions from the current literature onto an ambient pressure-temperature phase diagram, with colors indicating research methods. Current research focuses on three aspects: liquid-ammonia injection into low-pressure, low-temperature environments, with an emphasis on flash-boiling spray characteristics; injection into high-pressure, low-temperature environments that characterize non-reactive spray behavior; and injection into high-temperature, high-pressure environments to examine spray combustion characteristics that simulate engine operating conditions and reactive states. Table 2 summarizes practical liquid-ammonia injection applications within specific ambient pressure and temperature ranges based on these test conditions.
Prior to 2021, relatively few studies were conducted on liquid-ammonia injections. However, in recent years, this area has seen exponential growth. The trends shown in Fig. 3 indicate that flash-boiling studies—in which ammonia is injected into a low-pressure ambient—have gained more momentum than high-pressure injection studies, likely due to the relative ease of experimental setup. However, in 2024, flash-boiling studies were fewer than elevated-pressure studies, likely due to the additional time required to establish safer high-pressure experimental setups. Additionally, the consensus on ammonia being more suitable as a retrofit fuel for dual-fuel, diesel-based engines may have contributed to this shift.
In addition to the fundamental challenges associated with burning ammonia as a fuel, another non-negligible issue related to the DI and combustion of liquid ammonia pertains to its high latent heat of vaporization [23,58]. The high latent heat of vaporization of liquid-ammonia spray significantly impacts the combustion process by creating a strong local cooling effect as droplets evaporate [62]. This cooling effect substantially reduces the temperature in the spray region, inhibiting flame propagation and slowing chemical reaction rates [63]. In extreme cases, it may even lead to flame extinction [64]. To address these challenges, ammonia combustion systems often require additional energy input to maintain stable combustion. Strategies such as increasing compression ratios (CRs) [65,66], preheating air [67,68], or using pilot fuels [69,70], have been explored to counteract the strong evaporative cooling effect and ensure efficient and stable combustion.
1.4. Motivation and outline
This paper reviews recent research progress on the injection of liquid ammonia as a fuel and its application in combustion engines. It does not further discuss the topics of where, when, or whether liquid ammonia should be used as an energy carrier. The preceding sections have summarized the attractiveness of using ammonia as a fuel, the growing interest in ammonia combustion, and potential technical challenges. Previous articles have already extensively covered gaseous ammonia combustion, with excellent reviews and research work on chemical reaction mechanisms [71], [72], [73], [74], basic combustion characteristics [75], [76], [77], combustion instability [78], [79], [80], [81], combustion emissions and control [44,51,82], combustion applications [83], [84], [85], and combustion modeling [86], [87], [88], [89].
This paper synthesizes recent liquid-ammonia research from fundamental spray characteristics to engine implementation. Section 2 examines liquid-ammonia injections in low-pressure environments, emphasizing flash-boiling spray behavior. Section 3 explores spray dynamics under high-pressure, high-temperature conditions, covering both non-reacting sprays and combustion phenomena. Section 4 analyzes practical liquid-ammonia DI applications in engine systems. Section 5 identifies literature knowledge gaps, highlights key contemporary findings, and provides future research directions in this rapidly evolving field.
2. Liquid-ammonia injection into low pressure
Spray flash boiling occurs when pressurized liquid is rapidly depressurized below its saturated vapor pressure, causing instantaneous, explosive liquid-to-vapor phase change [90]. In this process, the pressurized liquid (or typically preheated to near boiling point) is injected through a nozzle into a lower-pressure environment. The sudden pressure drop creates superheat and thermodynamic instability [91]. This triggers rapid vapor bubble nucleation throughout the liquid, disrupting its structure and causing explosive vaporization. The rapid expansion significantly alters the spray characteristics, typically resulting in wider initial spray angles, smaller droplet sizes, enhanced atomization, and increased rapid mixing and heat transfer [92,93]. Flash boiling in sprays is still an active area of research, with relevant research progress summarized in the latest reviews [94], [95], [96].
Fig. 4 shows the ammonia phase diagram with the gas-liquid saturation line. Two pathways induce liquid-ammonia boiling: increasing the temperature at constant pressure above the saturation temperature (A → S1 → C) or decreasing the pressure at constant temperature below the saturation pressure (B → S2 → C).
The extent of flash boiling is characterized using two methods: pressure/temperature difference or pressure ratio [96]. The first uses the superheat degree (SD), which is defined as the difference between the saturated vapor pressure and the ambient pressure (PS2 − Pc) or between the fuel temperature and the boiling temperature (Ta − TS1). The second uses the superheat index, a dimensionless ratio expressed as either PS2/Pa or Pa/PS2 [97]. For an injected superheated liquid, Pa/PS2 < 1 [92,98], with values near 0.3 marking the onset of flash boiling [99]. For an injected subcooled liquid, Pa/PS2 ˃ 1. To maintain consistency, this review adjusts all reported values to match the first definition. Additionally, the term “superheat degree” is sometimes used interchangeably to describe both parameters.
2.1. Liquid-ammonia injection using single-hole injectors
Single-hole injectors provide a simplified, controlled environment for spray analysis by eliminating plume interactions, enabling precise measurements of spray evolution, penetration, and atomization. They minimize pattern variability, allowing the isolation of parameters such as injection pressure and nozzle geometry. Table 3 [100], [101], [102], [103], [104], [105], [106], [107], [108], [109], [110], [111] summarizes extensive recent experimental research on liquid-ammonia spray.
Li et al. [100] (Fig. 5) reported superheated ammonia spray characteristics across different saturation-ambient pressure ratios (Rp), identifying three flashing regimes: initial-flashing (1.0 < Rp ≤ 2.0), transitional-flashing (2.0 < Rp ≤ 5.0), and flare-flashing (Rp > 5.0). Near-field analysis showed that the bubble explosion intensity depends on the SD and fuel viscosity. Far-field penetration exhibited three stages—flashing-dominated, transitional, and quasi-steady—with time-dependence exponents deviating from traditional spray models. While ammonia spray penetration resembles that of diesel under critical- and initial-flashing conditions, bubble behavior increasingly dominates penetration evolution under transitional- and flare-flashing conditions.
The team’s subsequent investigation [101] of near-field characteristics and two-phase distribution in superheated ammonia sprays revealed exponential spray-volume expansion when the non-dimensional SD fell below 0.5. At a fuel temperature of 338 K, the spray volume increased by 200% within 1 mm below the nozzle. The maximum gasification ratio reached 3.5% at high temperatures, demonstrating the significant impact of bubble puffing and micro-explosions in the near-field region.
Ma et al. [102] studied flash-boiling spray characteristics and jet fluctuation in high-pressure liquid-ammonia injection, identifying three axial-development stages dominated sequentially by flash boiling, unbroken liquid-core acceleration, and broken liquid-core inertia (Fig. 6). The first and third stages negatively correlated with the SD, while the second correlated positively with the initial velocity. High-speed shadowgraphy revealed a decreasing spray area with increasing ambient temperature, disappearing swirling vortices, and increased asymmetry. Near-nozzle regions exhibited significant instability under flare flash-boiling conditions, with fluctuations intensifying up to 59% at elevated temperatures due to thermal disturbances.
Fang et al. [103] experimentally investigated high-pressure liquid-ammonia injection under non-flash and flash-boiling conditions, revealing a “spray-resistance phenomenon” characterized by suppressed liquid penetration and spray tip velocity during initial injection, particularly below 0.7 MPa ambient pressure. This phenomenon, attributed to needle-lift-induced cavitation and subsequent vapor formation, obstructed the nozzle hole, reducing the mass flow rate and spray tip velocity. Flash-boiling conditions caused near-nozzle spray expansion and bubble explosion, rapidly increasing the spray width and altering aerodynamic behavior. Under non-flash conditions, the Hiroyasu and Arai's model [112] predicted liquid penetration within 3% accuracy at 4 MPa ambient pressure, outperforming the Naber and Siebers' model [113]. Additionally, this team [104] measured the droplet diameter and velocity of high-pressure liquid-ammonia sprays up to 75 MPa using a phase Doppler particle analyzer, which determines particle velocity and size by measuring phase differences between Doppler signals from intersecting laser beams [114]. Compared with that of diesel at 1 MPa ambient pressure, ammonia’s Sauter mean diameter (SMD) was 50%-70% smaller across injection pressures of 25-75 MPa, attributed to its lower viscosity and higher evaporation rate. Above an injection pressure of 25 MPa, ammonia’s SMD remained below 12 µm.
Liu et al. [105] employed particle-droplet image analysis based on shadowgraphy to measure liquid-ammonia spray droplets and characterize their microscopic properties, although this method is limited to localized dilution zones and requires accurate individual droplet identification. Microscopic statistics revealed that the most probable droplet size increased from about 13 µm at Rp= 0.12 to approximately 26 µm at Rp = 1.06 (ambient to saturated pressure ratio), with corresponding SMD growth from about 20 to about 60 µm and significant reduction in droplet number density. While ambient temperature minimally affected the macroscopic spray morphology, it significantly increased evaporation, particularly at 373 K, where droplet density reached minimum values.
Li et al. [106] employed high-resolution microscopic drop-size imaging to measure liquid-ammonia spray droplet-size distributions (Fig. 7). Under flare-flashing conditions (Pa = 0.1 MPa), the distribution exhibited an M-shaped cumulative volume curve, best fitted by the BIDoseResp function. This bimodal distribution resulted from vertex recirculation, with numerous small droplets dominating the probability density, while fewer large droplets contributed significantly to the cumulative volume, confirmed by dual volume density peaks. The bimodal pattern reflects complex spray-air interactions under flare-flashing conditions. As the ambient pressure increased from 0.2 to 0.8 MPa, flash boiling transitioned from transitional to initial flashing regions, with the probability densities showing corresponding rightward shifts in droplet-size distribution.
Zhong et al. [107] examined nozzle diameter effects on the macroscopic and microscopic characteristics of flash-boiling liquid-ammonia sprays. High-speed shadowgraphy revealed that smaller diameters (0.12 mm) promoted intense radial expansion and vortex formation through enhanced thermodynamic fragmentation under low SDs, while larger nozzles (0.52 mm) produced asymmetric spray structures and larger droplets due to higher flow rates and weaker bubble nucleation. Microscopic analysis revealed non-monotonic droplet-size distribution trends, with the most probable droplet size increasing marginally for the largest diameter (0.52 mm). Spray SMD increased with nozzle diameter, particularly during non-flash boiling, where it rose up to 35%. Optimal spray performance balancing atomization, penetration, and radial dispersion was achieved using a 0.32-mm nozzle. Shen and Leach [108] also investigated liquid-ammonia spray atomization characteristics through droplet-diameter analysis. Flash-boiling conditions at low ambient pressure produced significant radial expansion, shortened penetration lengths, and reduced droplet sizes, with an SMD reaching 16 μm at 3 bar ambient pressure. Higher ambient pressures suppressed flash boiling, yielding longer penetration lengths and larger droplets (SMD of ∼30 μm at 15 bar). A bimodal droplet-size distribution emerged, with smaller droplets predominating under flash-boiling conditions.
Colson et al. [109] studied injection temperature and nozzle geometry effects on liquid-ammonia spray flashing transitions using single-hole nozzles. The transition between non-flash and flash boiling strongly depended on the nozzle aspect ratio L/D, with a critical threshold of L/D = 5 identified for flash-boiling onset, attributed to increased heterogeneous nucleation near nozzle walls. Jacob and Weber number transition criteria were validated but showed deviations, due to ammonia’s low viscosity and surface tension promoting bubble nucleation. Glass nozzle observations revealed that minor geometry variations, such as orifice tapering, could trigger internal flashing and enhance atomization.
Desclaux et al. [110] investigated the behavior of superheated ammonia sprays under flash-boiling conditions. With Schlieren imaging and thermocouples, they demonstrated that reducing Rp significantly amplifies radial spray expansion, with a notable change observed at Rp > 8.85, resulting in distinct flaring of the liquid plume. Spray penetration exhibited a non-linear relationship with Rp, initially increasing before stabilizing due to increased bubble formation and droplet fragmentation near the nozzle. The cooling effect induced by flash boiling led to temperature reductions of up to −60 °C near the spray core, indicating a correlation with the latent heat vaporization.
Bjørgen et al. [111] further investigated the flash-boiling dynamics of superheated liquid-ammonia sprays. Their results showed that higher SDs led to significant radial spray expansion and reduced penetration, primarily due to increased bubble nucleation within droplets. At Rp > 4, radial expansion dominated over penetration, effectively decoupling spray morphology from ambient density effects. Near-field analysis revealed that, for high Rp, the liquid core became less dense and exhibited shock-wave formation, with shocks becoming more pronounced as Rp increased beyond 65, as presented in Fig. 8. In-spray temperature measurements highlighted substantial cooling within the spray core, with minimum temperatures reaching as low as −66 °C at low ambient pressures. This significant cooling was attributed to ammonia’s high latent heat of vaporization.
In addition to the experimental investigations discussed above, several researchers utilized numerical methods to study the behavior of liquid-ammonia sprays from single-hole nozzles. A summary of these studies is presented in Table 4 [115], [116], [117], [118].
Huang et al. [115] investigated high-pressure liquid-ammonia injection under flash-boiling and non-flash-boiling conditions. Lower ambient pressure produced longer spray penetration and higher gas-phase velocity, while the ammonia vapor mass fraction decreased with increasing ambient pressure, intensifying heat transfer and lowering the gas-phase temperature. Flash-boiling sprays exhibited high initial droplet velocity, significant radial expansion shortly after injection, and larger evaporation source terms compared with non-flash-boiling sprays, which showed lower droplet velocity, moderate velocity variation over time, and weaker radial expansion.
An et al. [116] numerically evaluated five phase-change models for liquid-ammonia flash sprays under various temperatures and pressures. The combined Zuo and Langmuir-Knudsen (LK) model (Eq. (1)) proved most suitable for superheated conditions, matching experimental penetration length and spray characteristics data (Fig. 9). Evaporation rate increased with decreasing pressure, significantly reducing droplet and gas-field temperatures. The ammonia vapor mass fraction showed complex temperature relationships: initially decreasing then increasing under low pressure, while positively correlating at high ambient pressure. These findings highlight vaporization cooling effects and pressure’s influence on droplet-gas phase interactions, which are critical for liquid-ammonia combustion applications.
where χneq,s and χeq,s are vapor molar fraction at the droplet surface contains non-equilibrium and equilibrium term, respectively; LK is the Knudsen layer thickness; d is the droplet diameter; β is the non-dimensional evaporation parameter.
Wang et al. [117] utilized 2D numerical simulations to investigate flash-boiling liquid-ammonia spray characteristics, focusing on orifice aspect ratios and SDs. They validated the Eulerian framework and homogeneous relaxation model (HRM) for simulating flash-boiling ammonia sprays through a comparison with experimental data from Colson et al. [109]. The rate of phase change is modeled by
where dYl is the change in liquid mass fraction after timestep dt, and θ is the relaxation timescale over which the phase change is expected to occur.
Higher SDs significantly enhanced bubble growth and phase transitions, improving atomization and fuel-distribution uniformity. An aspect ratio (L/D) of 10 produced a unique gas-phase distribution with a lower void fraction near walls and a higher fraction near the center axis, indicating intense phase transitions at the nozzle exit. Fig. 10 [117] shows axial and radial temperature distributions for four sprays. Flash boiling significantly accelerates atomization while notably reducing the spray-area temperature, with the temperature decrease intensifying as the SD increases. A higher superheat causes internal flash boiling and rapid vaporization, creating sharp temperature drops downstream of the injection nozzle that gradually flatten. The radial temperature distribution exhibits an M-shaped profile from the competing effects of droplet evaporation and convective heat transfer with ambient gas. While flash boiling enhances atomization, it substantially reduces spray temperature.
In addition, Zhou et al. [118] conducted theoretical and numerical investigations on high-pressure DI liquid-ammonia spray similarity across different engine sizes and nozzle-hole dimensions. They established single-value conditions and two similarity laws for scaling liquid-ammonia sprays, demonstrating that both the speed law and pressure law achieve similarity in fuel-injection rate, spray breakup, spray tip penetration evolution, and maximum liquid length.
2.2. Liquid-ammonia injection using multi-hole injectors
Multi-plume sprays from multi-hole injectors closely represent actual engine applications, as these injectors ensure sufficient fuel delivery while synergizing with combustion chamber geometry to optimize mixing. Table 5 [119], [120], [121], [122], [123], [124] summarizes recent studies on liquid-ammonia spray characteristics from multi-hole injectors.
Pelé et al. [119] conducted the first experimental study on liquid-ammonia spray characteristics using Schlieren imaging (Fig. 11). Ammonia sprays showed greater sensitivity to air density and temperature than conventional fuels, producing longer, narrower sprays under low-air-density conditions. Flash-boiling conditions near saturation pressure (8.8 bar at 20 °C) yielded wider spray angles at half penetration length, with notable collapse effects from vapor condensation. The researchers developed a semi-empirical correlation for spray-penetration prediction based on thermodynamic parameters, achieving high accuracy. Temperature-sensitivity quantification revealed exponential coefficient values of 0.46 for liquid ammonia, 0.30 for gasoline, and 0.21 for methanol, confirming ammonia’s heightened temperature responsiveness.
Using a Bosch six-hole injector, Akram et al. [120] experimentally studied ammonia spray flash-boiling behavior. Ammonia exhibited significantly greater vaporization than methanol, ethanol, or gasoline, particularly as the injector tip temperature increased to 300 °C, which was consistent across chamber pressures of 10 and 30 bar. At a chamber pressure of 10 bar, ammonia showed the lowest penetration rates and spray areas, while other fuels demonstrated comparable values. Generally, increasing the injector tip temperature at a lower chamber pressure decreased the penetration rates and spray areas for all fuels. At a chamber pressure of 30 bar, ammonia and other fuels exhibited similar penetration rates and spray area ranges.
Scharl et al. [121] investigated fuel-temperature and injection-timing effects on ammonia DI, innovatively measuring injection mass and evaluating discharge coefficients through exhaust-gas composition analysis during engine operation with ammonia injection but without combustion. Increasing the fuel temperature decreased the ammonia injection mass up to 12.8%, while minimally impacting discharge coefficients during late injection timings and high in-cylinder pressures. However, early injection timings showed discharge coefficient decreases of up to 17.4% at elevated fuel temperatures. The study confirmed that temperature increases notably reduce the injection mass, primarily because ammonia’s pronounced temperature sensitivity significantly affects its density.
Zembi et al. [122,125] conducted numerical studies based on Pelé’s multi-hole ammonia spray experiments [119]; they successfully compared numerical results with experimental data for liquid and vapor tip penetration, local SMD, and overall spray morphology, establishing a predictive simulation framework for ammonia injections. Their study highlighted challenges in capturing local spray details across different regimes—particularly flash-boiling phenomena crucial for understanding ammonia’s unique properties. Standard breakup model constants for conventional fuels proved inadequate for ammonia, requiring significant tuning and specific calibration. The researchers investigated liquid-ammonia spray behavior under varying ambient pressures, focusing on transitions between flash-boiling and non-flashing regimes, and specifically calibrated Kelvin-Helmholtz and Rayleigh-Taylor (KH-RT) model constants for both conditions. The results demonstrated good agreement between the simulations and experiments for liquid spray tip penetration and local SMD (Fig. 12) [122].
Based on previous experimental and simulation studies, Hu et al. [123] conducted a numerical investigation of multi-plume liquid-ammonia spray. They developed a novel variable discharge coefficient Cd accounting for spray resistance from ammonia vaporization within the nozzle, achieving prediction errors below 1.5% for spray penetration at a fuel temperature of 308 K across all flashing ratios. A breakup model constant B1 that strongly influences spray penetration was designed as a variable constant formula with an identical functional form to variable Cd (Eq. (3)), improving predictions at higher fuel temperatures. Using the variable Cd-B1 model at a fuel temperature of 338 K, the spray penetration prediction error at an injection time of 1 ms remained below 5% for all ambient pressures and approached 0 during the flare flash-boiling stage, as shown in Fig. 13 [123].
TTrefwhere Fr is the flashing ratio and Tref is the reference temperature of 338 K.
Rachakonda et al. [124] compared the internal flow and near-nozzle behavior of iso-octane, methanol, and ammonia under flash-boiling and subcooled conditions. Flash-boiling sprays demonstrated greater steadiness than their subcooled counterparts and behaved as either under-expanded or over-expanded jets, depending on the operating conditions and fuel thermophysical properties. High Rp values (saturation to ambient pressure ratio) produced a “bell-shaped” morphology with significant spray collapse, characterized by strong under-expanded jets that generated shock fronts truncated by Mach disks perpendicular to the spray axes. The Mach disk appeared as an abrupt pressure rise, visible as darker regions in superheated ammonia spray pressure contours (Fig. 14 [124]). Rapid vaporization and expansion accelerated sprays beyond sonic velocities, creating low-pressure regions relative to ambient pressure. Shock cell boundaries formed through barrel shocks or constant pressure streamlines, followed by intercepting shocks as the flow turned inward. Triple point regions were clearly visible where the Mach disks intersected intercepting shocks, producing reflected shocks and slipstreams from both sides, as shown in the log-scaled density gradients in Fig. 14.
2.3. Summary
This section provides a comprehensive review of research on liquid-ammonia injection and spray characteristics, with a particular focus on flash-boiling phenomena and the behavior of ammonia sprays from both multi-hole and single-hole injectors. Key parameters influencing flash boiling include the saturation-ambient pressure ratio and degree of superheat.
Multi-hole injector studies show that ammonia sprays are highly sensitive to air density and temperature, producing longer, narrower sprays at low densities. Flash-boiling widens spray angles and causes collapse from vapor condensation, while higher injector temperatures enhance vaporization and alter spray characteristics. Ammonia’s temperature-sensitive density affects the injection mass and discharge coefficients. Single-hole injectors enable more precise spray analysis, revealing distinct flashing regions based on pressure ratios and showing the influence of bubble behavior on penetration during flash-boiling. Nozzle diameter, ambient pressure, and fuel temperature critically determine spray characteristics; smaller nozzles increase radial expansion, while larger ones produce larger droplets. Numerical and experimental studies developing predictive models reveal challenges in modeling ammonia sprays due to their unique thermodynamic properties, which require ammonia-specific breakup models. These findings emphasize optimizing injection parameters and nozzle design for efficient atomization and mixing in ammonia engines.
In addition, critical examination of the reviewed studies reveals several conflicting findings that warrant further investigation. For instance, while Pelé et al. [119] reported spray penetration lengths 44.7%-56.2% smaller than those of diesel, Payri et al. [126] found comparable penetration rates in the fully developed zone. This discrepancy likely stems from differences in experimental conditions and measurement techniques, as Pelé used Schlieren imaging in a constant-volume chamber, while Payri employed diffuse background illumination (DBI) in a constant-pressure flow vessel. In addition, discrepancies in spray penetration under similar conditions (e.g., 12% shorter penetration in Ref. [100] vs Ref. [102] at Rp = 5.0) arise from nozzle geometry variations (0.28 vs 0.32 mm diameter), emphasizing the need to non-dimensionalize the spray parameters with respect to the nozzle geometry. The robustness of numerical models also varies significantly; while the Lagrangian particle method shows good agreement for non-flash-boiling conditions, Zhang et al. [127] highlighted significant discrepancies under intense flash boiling, indicating that current models require substantial improvements for capturing thermal breakup effects. While large-eddy simulation (LES) models [115] robustly capture near-field flash boiling, Reynolds-averaged Navier-Stokes (RANS) approaches [122] struggle with plume-collapse effects due to unresolved turbulence-chemistry interactions, limiting their use in engine-scale simulations.
3. Liquid-ammonia injection into high pressure/temperature
This section examines liquid-ammonia injections in high-pressure, high-temperature environments that closely replicate engine operating conditions, building on the low-pressure spray studies previously discussed. This progression is essential for three reasons: Engine compression generates ambient pressures that exceed ammonia’s saturation pressure, thereby suppressing flash boiling; elevated temperatures create competing effects between increased evaporation and reduced flash-boiling tendency; and reactive conditions fundamentally alter spray development through heat release and combustion-induced flow fields. Understanding these behaviors under engine-relevant conditions is critical for translating fundamental spray physics into practical combustion system designs.
3.1. Non-reactive ammonia injection and spray study
Research on non-reactive ammonia injection and spray is essential for optimizing injection strategies and ensuring efficient atomization and distribution in combustion systems. Non-reactive conditions allow researchers to isolate and analyze fundamental physical processes, including spray formation, droplet size distribution, and penetration. The constant volume combustion chamber (CVCC) provides controlled environments for studying spray characteristics under precise pressure and temperature conditions that simulate combustion device operations, while enabling high-quality optical access for advanced diagnostics [128]. Table 6 [129], [130], [131], [132] lists recent non-reactive ammonia injection and spray studies.
Wang et al. [129] and Li et al. [130] investigated high-pressure ammonia spray combustion characteristics under diesel-like conditions. Under non-reacting conditions, ammonia sprays exhibited macroscopic characteristics similar to diesel sprays, with significant evaporation at room temperature. Empirical diesel spray correlations can be used to predict ammonia spray tip penetration by incorporating Reynolds number and Ohnesorge number. Microscopic imaging revealed that the ammonia sprays showed greater susceptibility to needle-motion-induced turbulence or cavitation during injection transients. Fig. 15 [130] shows that the ammonia spray liquid length was 44.7%-56.2% smaller than that of diesel, with differences increasing with ambient density. While liquid-phase development in ammonia sprays initially lagged behind that of diesel from the start of the injection, ammonia sprays showed pronounced liquid penetration increases during the quasi-steady state due to their large latent heat of evaporation.
Ma et al. [102] investigated the spray characteristics and jet fluctuations of high-pressure DI liquid ammonia across ambient temperatures of 291-600 K. Under high-temperature and high-pressure conditions, liquid-ammonia sprays were significantly influenced by the interplay between flash boiling and evaporation effects. Increasing the ambient temperature produced three-stage axial spray development: Initial stages dominated by flash boiling caused rapid spray area increase, while later stages showed high-temperature evaporation dominance, resulting in marked spray area reduction. At 600 K, pronounced evaporation reduced the spray, as intense heat transfer overcame liquid ammonia’s latent heat of evaporation. Spray centroid oscillation deviations increased with rising temperatures, indicating greater jet instability, with maximum deviation growth rates reaching 59.16% in near-nozzle regions.
Payri et al. [126] experimentally investigated DI liquid-ammonia spray characteristics under non-reacting diesel-like conditions. Fig. 16 [126] shows spray images captured using DBI and Schlieren techniques for both fuels. Ammonia jets exhibited cone-like shapes similar to those of diesel, with narrower spray angles. Ammonia demonstrated slower initial penetration rates compared with diesel but achieved comparable rates in fully developed zones. Ammonia liquid length was shorter than that of diesel, influenced by ambient temperature, density, and nozzle diameter, but not injection pressure. A one-dimensional (1D) spray model originally developed for diesel fuels showed good agreement with experimental results for quasi-steady liquid length and vapor penetration when applied to ammonia.
Yang et al. [131] experimentally investigated high-pressure liquid-ammonia spray characteristics under various ambient conditions. Their results revealed significant differences in spray development across flash-boiling, transitional flash-boiling, and non-flash-boiling conditions. Under flash-boiling conditions, the spray tip velocity initially increased then gradually decreased, while non-flash-boiling conditions showed significant initial velocity decrease followed by forward penetration at a similar velocity. Higher ambient temperatures increased vaporization, shortening liquid spray penetration length. This research provides comprehensive reference data for numerical simulations and valuable theoretical and practical guidance for ammonia spray combustion studies and ammonia-fueled engine development.
Wu et al. [132] investigated the spray and evaporation characteristics of high-pressure liquid-ammonia injection compared with those of diesel. Liquid penetration initially increased then decreased with rising ambient temperature, with large-nozzle sprays showing higher inflection-point temperatures. Small-nozzle sprays exhibited greater ambient condition sensitivity at high temperatures, reducing spray stability. Ammonia demonstrated higher axial diffusion rates but lower radial diffusion rates than diesel. At 800 K, distinct spray-structure differences emerged: Capturing and recirculation regions from pilot-evaporated ammonia prevented complete droplet evaporation, unlike diesel. Ammonia vapor cone angles were 4°-6° smaller than those of diesel under evaporating conditions, with liquid cone angle differences decreasing at higher ambient temperatures. Ammonia spray volume exceeded diesel spray volume due to ammonia’s high latent heat of vaporization.
Zhang et al. [127] presented a computational fluid dynamics (CFD) study on the fuel/air mixing properties of liquid ammonia under DI engine conditions. Current Lagrange-based spray models effectively replicated liquid-ammonia spray characteristics under non-flash-boiling conditions but showed significant discrepancies with experimental measurements under intense flash-boiling conditions. This finding highlights the need for more precise flash-boiling models incorporating thermal breakup effects to accurately predict ammonia fuel/air-mixing properties across the full spectrum of engine-relevant conditions.
Research on liquid-ammonia injections have employed established spray methods using a constant volume chamber (CVC) with optical techniques such as diffused back-illumination imaging to observe fundamental spray behavior under elevated pressure and temperature conditions. Studies have consistently examined how ambient temperature, pressure, and injection pressure affect ammonia spray characteristics, finding distinct differences from diesel sprays in penetration rates and evaporation characteristics. Ambient temperature has emerged as a significant influence on spray behavior and penetration length across all studies. These findings hold practical significance for ammonia-fueled engine development and sustainable transportation applications.
3.2. Ammonia spray combustion without reactive fuel
Igniting and sustaining liquid-ammonia spray combustion is challenging due to ammonia’s high auto-ignition temperature, narrow flammability range, high latent heat of vaporization, and low laminar flame speed. Achieving single ammonia spray combustion without reactive fuel additives requires appropriate atomization through high-pressure injections and optimal nozzle design to control droplet size distribution, as smaller droplets facilitate faster evaporation and mixing. Maintaining ammonia concentration within its flammability range (15%-28% by volume in air) with sufficient oxygen is essential. Fuel and ambient air preheating compensate for cooling effects from ammonia’s high vaporization latent heat. High-temperature and -pressure ambient conditions enhance combustion processes. Continued research into combustion dynamics, flame stability, and heat-release mechanisms remain vital for improving liquid-ammonia fuel combustion system performance and reliability.
An experimental investigation by Wang et al. [129] demonstrated successful high-pressure liquid-ammonia spray auto-ignition under diesel-like conditions, with 21% oxygen concentration and comparable temperature/pressure levels, marking the first observation of this phenomenon. Fig. 17 [129] shows the temporal morphology of the ammonia spray flames at various ambient and fuel temperatures. Ammonia combustion intensity was significantly lower than that of diesel and more sensitive to ambient temperature than initial fuel temperature. The lift-off length at 1000 K remarkably exceeded that at 900 K, while the apparent heat-release rate (AHRR) increased by 40% at 1000 K because of increased after-burning and earlier ignition. Ammonia’s spray flame lift-off length exceeded diesel’s, indicating slower combustion. The AHRR exhibited a sharp initial spike followed by slow heat release, suggesting a transition from premixed to mixing-controlled combustion. The cumulative heat from ammonia diffusive combustion was less than one-fourth of that from diesel spray combustion.
Wu et al. [133] experimentally investigated the ignition and flame-development characteristics of high-pressure liquid-ammonia spray combustion using simultaneous high-speed imaging of OH* and NH2* chemiluminescence with flame luminosity (FL) under controlled conditions (3 MPa ambient pressure, 950 K temperature, 40 MPa injection pressure). The combustion process comprised four stages—ignition with auto-ignition kernels, flame propagation, fully developed combustion, and post-combustion—with key radical distributions characterized throughout. Fig. 18 [133] shows the temporal evolution of OH*, NH2*, and FL images. During fully developed combustion, all signals covered the spray core region, with NH2* showing a smaller distribution and lower intensity than OH* except at the core region and flame front. OH* intensity peaked centrally, diminishing toward the flame sides. Post-combustion showed a significant FL decrease, OH* contour alignment with the flame front, and rapid NH2* reduction, disappearing before OH*. Combustion lasted approximately 5.6 ms. The study identified unique small-scale auto-ignition phenomena after main combustion, providing insights into chemical kinetics—including OH and NH2 radical formation—and contributing valuable knowledge for developing numerical models and increasing ammonia fuel efficiency.
In addition to ammonia spray combustion, Matsuura et al. [134] investigated individual ammonia-droplet combustion using hot-wire ignition and back-illumination high-speed imaging under high-pressure microgravity conditions. Their study revealed unique combustion behaviors, including droplet disruption, puffing, and re-expansion (Fig. 19 [134]). Initially, droplets vaporized at consistent rates that decreased over time, reducing the flame standoff ratio. These phenomena resulted from water vapor accumulation on droplet surfaces, which suppressed ammonia vaporization and caused superheating, leading to bubble nucleation and growth. This pioneering research on ammonia-droplet combustion provides insights into designing ammonia spray combustors and developing carbon-neutral fuel technologies for gas turbines and marine diesel engines.
3.3. Ammonia spray combustion with pilot-ignited reactive fuel
Liquid ammonia exhibits modest reactivity and limited combustion characteristics under standard conditions due to high activation-energy barriers and slow reaction kinetics, making stable flame maintenance challenging. Its substantial latent heat requires considerable vaporization energy before combustion initiation, collectively limiting standalone fuel utilization. To overcome these limitations, dual-fuel strategies employ higher-reactivity fuels such as hydrogen or hydrocarbons as combustion promoters. These active fuels react first, providing the initial energy and chemical pathways to overcome ammonia’s combustion challenges by generating heat and free radicals that break down ammonia molecules, sustaining the chain reactions necessary for continuous stable combustion. Table 7 [135], [136], [137], [138], [139], [140], [141], [142], [143], [144] presents a summary of recent studies on liquid-ammonia spray combustion using pilot-ignited reactive fuels.
Although gasoline [145] and DME [146,147] have been explored as dual-fuel combustion partners to enhance liquid-ammonia combustion characteristics, recent research predominantly focuses on diesel-piloted ignition strategies. When diesel serves as a pilot fuel, it overcomes ammonia’s poor combustion properties through multiple mechanisms: providing reliable ignition due to a lower ignition temperature than ammonia [148,149], creating local high-temperature zones facilitating ammonia decomposition, and generating hydrocarbon radicals promoting ammonia oxidation reactions. Diesel’s higher energy density and established fuel-delivery systems make it practical for dual-fuel applications. Combining diesel’s reliable ignition properties with ammonia’s carbon-free fuel potential makes this pairing particularly suitable for internal-combustion engine applications [150].
Scharl and Sattelmayer [70] investigated diesel-piloted ammonia ignition and combustion characteristics by systematically varying temporal and spatial spray interactions, diesel pilot amounts, and injection durations. Using independent single-hole injectors, they achieved variable spray interaction angles from −22.5° (converging) to 15° (diverging) parallel to the cylinder head. Their experiments demonstrated reliable ammonia ignition using pilot-ignited diesel with a lower heating value (LHV) fraction as small as 3.2%. Strong diesel-ammonia spray interaction proved crucial for effective ignition, with injection sequence adjustments mitigating misfiring. Low-load conditions showed less effective combustion, indicating ammonia’s potential limitations in certain engine operations, despite its carbon-free benefits. This research contributes valuable insights for understanding ammonia’s feasibility as a sustainable fuel alternative, addressing NOx emissions and combustion efficiency challenges while informing future engine design and fuel strategies for GHG reduction.
Based on the same rapid compression expansion machine (RCEM) system, Scharl et al. [69] explored ammonia spray combustion characteristics and mixture formation under high-pressure DI conditions. Fig. 20 [69] shows simultaneous shadowgraph and OH* chemiluminescence images of diesel-piloted ammonia combustion. Diesel-piloted ammonia spray behavior was analyzed via heat-release rate (HRR) analysis and a validated 1D spray model. The results revealed that diesel-ignited ammonia spray flames failed to stabilize, causing a continuously increasing lift-off length throughout injection, which was evident from parametrized HRR, OH*-intensity, and lift-off length data. This resulted from ammonia’s high mixing requirements for auto-ignition, influenced by its physical properties and diesel combustion product interactions. Ammonia sprays quickly became lean, with mean equivalence ratios well below stoichiometry, further complicating flame stabilization. These findings highlight ammonia fuel challenges in internal-combustion engines, suggesting that diesel post-injections and fuel pre-heating strategies may reduce pollutant formation.
Scharl and Sattelmayer [135] reported recent findings on diesel-piloted ammonia spray combustion flame emissions under internal-combustion engine conditions. The combustion process showed distinct phases, with ammonia combustion contributing to most heat release. Fig. 21 [135] displays background-corrected OH* and NH* intensities. OH*, NH*, and NH2* emissions significantly influenced FL, with NH2*-chemiluminescence (CL) primarily contributing to visible emissions during ammonia combustion. The OH*-radiation intensity from the ammonia spray was significantly lower than that from the diesel pilot combustion, despite diesel comprising only 5% of the supplied LHV. For low-soot-formation cases, the authors recommend focusing on wavelengths capturing NH2*-CL, which provide strong signals originating mainly from the flame front during ammonia combustion.
Zhang et al. [136] investigated ammonia and diesel combustion characteristics in dual DI mode, focusing on flame propagation and distribution. Fig. 22 [136] shows combustion processes at 0 and −0.5ms intervals. Simultaneous diesel and ammonia injections significantly increased combustion efficiency while minimizing unburned ammonia emissions. At an ambient temperature of 850 K, substantial diesel injection pressure created prominent flame kernels but delayed the concentrated flames from ammonia combustion on one side of the injection axis, impeding opposite-side combustion participation and causing low HRR and elevated unburned emissions. Increasing the ambient temperature to 900 K notably improved these issues. Larger spray cone angles and higher injection rates produced wider ammonia spray distribution, preventing flame backpropagation to spray junctions and improving combustion stability.
Wu et al. [137] investigated ammonia-polyoxymethylene dimethyl ether (PODE) combustion characteristics in dual-fuel DI systems, focusing on spray interaction, ignition, and flame development. Fig. 23 [137] shows radial integral OH* intensity below the theoretical interaction point versus axial distance and time, revealing gradual downstream shifts in the reaction center. The simultaneous injection (APDI-S) strategy exhibited larger axial spatial spans than PODE-pilot injection (APDI-P). During spray interaction, insufficient air mixing in collision sprays created high-equivalence-ratio regions in collision cores, delaying reactions. OH* signals weakened at 1.4-1.7 ms after the PODE injection started (ASOPI) for APDI-S and 1.9-2.1 ms for APDI-P. APDI-S showed longer OH* duration, with persistent bands below the interaction points post-combustion. Radial intensity distributions revealed three peaks with larger spatial spans in APDI-P, where the reactions occurred more on the liquid-ammonia sides. APDI-S demonstrated extended reaction durations with dispersed centers, while APDI-P showed concentrated reaction areas with earlier core region initiation. The injection timing significantly affected flame development: Simultaneous injection promoted aligned flames, while pilot injection resulted in vertical growth. Longer overlapping injection durations increased combustion efficiency.
Pathak et al. [138] investigated the temperature dependence of flame behavior in high-pressure dual-fuel diesel-piloted liquid-ammonia spray combustion, validating against optical and heat-release-rate data from RCEM experiments [69,70]. Ammonia’s lower stoichiometric air requirement created lean combustion downstream of flame lift-off, reducing combustion product temperatures. Hot combustion-product entrainment into a cool ammonia spray proved insufficient for early autoignition. Diesel combustion products addressed this by serving as heat reservoirs near the ammonia spray, facilitating liquid-ammonia evaporation and vapor autoignition but necessitating sophisticated diesel injections and combustion strategies. Across all temperatures, ammonia combustion initiated through H-abstraction via OH radicals from diesel combustion. After OH radical consumption, the chemical contribution of diesel became negligible, particularly at elevated temperatures. Rising ambient temperatures increased the significance of direct ammonia oxidation through oxygen reactions.
Zhang et al. [139] investigated ammonia-diesel high-pressure-injection dual-fuel combustion ignition characteristics under varying ambient conditions. The pilot ignition delay for diesel showed sensitivity to ambient conditions, with lower temperatures and pressures increasing delays, while the main ammonia ignition delay was primarily influenced by spray interactions and was less ambient-sensitive. The flame progression transitioned from diesel to two-layer to ammonia flame, following the ammonia injection direction (Fig. 24 [139]). This research introduced a new flame index definition for ammonia spray combustion, enhancing current understanding of the combustion regime. Increasing the ambient pressure showed no significant ignition improvement and had a potentially negative impact on reliability. The EGR strategy primarily affected ignition through ambient oxygen concentration. High ammonia energy fractions posed misfire risks due to increased mixing requirements and reduced flame temperature and HRR.
Yu et al. [140] used a validated LES model in OpenFOAM to examine ignition and pollutant formation in ammonia/n-heptane dual-fuel spray combustion under various ammonia concentrations $\phi_{\mathrm{NH}_{3}}$. Fig. 25 [140] displays the spatial and temporal pollutant distribution under different ϕNH3 conditions. For $\phi_{\mathrm{NH}_{3}}$ = 0.25, 0.5, and 0.75, NO2 was concentrated mainly at the spray base below 1150 K where first-stage ignition occurred, with NO2 regions widening as $\phi_{\mathrm{NH}_{3}}$ increased. NO2 distribution remained relatively stable temporally within the cases, being transported downstream with the n-heptane spray flow and converted to N2O and NO via the reactions NO2 + NH2 = N2O + H2O and NO2 + H = NO + OH. N2O appeared downstream of NO2 in the dual-fuel cases, followed by NO at the spray tips. NO2 production occurred in the inner spray regions between 1150–1500 K (marked by red and black lines), increasing temporally in dual-fuel cases but remaining negligible at ϕNH3 = 0.
The above diesel-piloted ammonia spray combustion research primarily examines configurations where single-hole diesel and liquid-ammonia injectors are positioned proximally on the same side. This arrangement facilitates rapid spray-plume convergence at small angles (< 15°), ensuring effective diesel-ammonia spray intersection and interaction.
Tian et al. [141] conducted visualization studies on ammonia/diesel dual DI combustion characteristics with increased spray interaction angles of 60° (Fig. 26). Higher ammonia injection pressure increased air entrainment and fuel-air mixing. Optimal combustion required an ammonia injection close to the flame. Fuel-injection rates critically determined HRR in the high-pressure dual-fuel mode. Increasing the ammonia nozzle diameter and injection pressure significantly boosted ammonia injection rates and peak HRR through increased air entrainment and mixing. Pre-diesel ammonia injections produced mixed effects, forming combustible mixtures that increased peak HRR, while causing incomplete combustion due to rapid ammonia evaporation. Higher diesel injection pressure minimally impacted ammonia combustion and was inhibited by the presence of ammonia. In the high-pressure dual-fuel mode, converged diesel and ammonia sprays followed the ammonia spray direction, although higher diesel pressure could cause directional deviation.
Chen et al. [142] experimentally investigated cross-spray and combustion behaviors in diesel-ammonia dual-fuel DI systems with diesel injectors positioned atop the CVCC and liquid-ammonia injectors on the side, creating 90° spray intersection angles. High-speed Schlieren imaging revealed that ammonia exhibited lower axial penetration, higher evaporation rates, and smaller diffusion rates than diesel, which were attributed to ammonia’s lower density, viscosity, and boiling point. Diesel-ammonia spray collision increased the overall spray diffusion and fuel-air mixing. Vertical sprays showed larger liquid-phase penetration, with adjustments to the injection intervals improving atomization. Increasing the diesel-ammonia injection intervals initially delayed ignition but subsequently reduced combustion duration and soot emissions, suggesting that appropriate interval adjustments optimize combustion efficiency and reduce emissions.
Zhang et al. [143] positioned diesel and ammonia injectors on opposing CVC sides, creating collision patterns between sprays. Various nozzle-hole tilt angles produced diverse spray intersections. Fig. 27 [143] shows a spray interaction where ignition occurs at the upper and lower plume-collision points. High-pressure ammonia injection created gas entrainment and pushed-out zones, with effective ignition occurring 16-35 mm from the nozzle in entrainment zones. Diesel flames surrounding liquid ammonia were extinguished by ammonia gasification, while pushed-out zone flames failed to ignite the ammonia. Reliable ammonia ignition and diffusion combustion, achieving 74.4% energy replacement, were obtained by entraining the diesel flame and surrounding air into ignition zones. Increased ammonia injection or prolonged diesel-flame suspension was necessary for stable combustion. Premixed diesel flames igniting the ammonia spray increased the combustion efficiency and flame consistency compared with diffused diesel flames due to the leaner, more homogeneous air-fuel mixtures.
Xu et al. [64] combined optical experiments and LES to examine diesel-ammonia flame interaction and pollutant formation mechanisms (NO and N2O) under two scenarios. Fig. 28 [64] shows the local equivalence ratio (Φ), ammonia/diesel mixture fraction (Z), heat-release rate (Qdot), and mass fractions of CO, NH3, NO, and N2O in spray cross-sections at different times. The ammonia-air mixtures displayed distinct pre-combustion characteristics between injection cases. Early interaction scenarios showed fuel-rich ammonia-air mixtures igniting directly from diesel flames, enabling self-sustained flame propagation and significant heat release that maintained high-temperature regions for continuous combustion. Late interaction scenarios resulted in rapid ammonia evaporation, which created fuel-lean ammonia/air mixtures that failed to ignite from diesel flames, causing flame extinction. NOx and N2O emissions proved highly sensitive to diesel-ammonia flame interaction. N2O formed predominantly at ammonia flame-quenching fronts, presenting significant challenges due to ammonia’s fast evaporation and slow oxidation rates in fuel-lean mixtures.
In addition, Sharma et al. [144] examined the incorporation of ammonia as a low-reactivity fuel in diesel-assisted dual-fuel combustion under reactivity-controlled compression ignition (RCCI) conditions using LES. Through flamelet analysis, they investigated the flame-structure evolution and combustion modes of dodecane/ammonia mixtures. Their results showed that ammonia combustion initiated through dodecane low-temperature combustion activation, transitioning to high-temperature combustion, with ammonia consumption occurring as flames evolved toward stoichiometric conditions. Under fuel-lean conditions, ammonia exhibited no autoignition behavior.
3.4. Ammonia spray combustion with other emerging methods
Aside from pilot-diesel ignition, researchers have explored the use of turbulent jet injection (TJI) to enhance liquid-ammonia spray combustion. Zhong et al. [151] investigated transient ammonia spray flames using pre-chamber TJI, as illustrated in Fig. 29. Hydrogen injected into the pre-chamber generated high-speed jet flames. TJI-induced ammonia spray flames progressed through three phases: pre-chamber combustion, partially premixed combustion, and mixing-controlled combustion. Pre-chamber mixture enrichment greatly enhanced TJI ignition capability, causing earlier ignition- and combustion-mode transitions. Larger nozzle diameters with lower turbulent jet velocities improved the ignition ability and extended the critical oxygen concentration for stable ignition, which was attributed to the increased jet-ammonia spray interaction and mixing behavior enabling full ignition-energy utilization. Reduced turbulence intensity decreased heat dissipation, supporting ignition processes.
Researchers have explored improved liquid-ammonia injection methods. Ichikawa et al. [152] proposed a three-layer stratified fuel-injection approach for compression-ignition engines, involving sequential injections of a high-reactivity supporting fuel (nC16H34), liquid ammonia, and a second supporting fuel through a single nozzle (Fig. 30). CVCC verification confirmed that the system effectively produced two- or three-layer sprays, evidenced by injection energy rate measurements. First-layer nC16H34 ignited ammonia sprays, although the ammonia ignition delays slightly exceeded those of 100% nC16H34 sprays. Third-layer nC16H34 facilitated second-layer ammonia combustion, creating high HRR peaks at the combustion end and significantly shortening the late combustion periods. Third-layer nC16H34 combustion drastically reduced N2O emissions by eliminating low-temperature regions in the liquid-ammonia spray diffusion flames where N2O typically forms through active combustion.
Following previous work, researchers experimentally studied ammonia stratified spray combustion using high-speed direct imaging [153]. Fig. 31 [153] shows high-speed images of ammonia stratified and 100% nC16H34 spray flames. Ammonia’s spray flame penetration lengths and cone angles matched those of 100% nC16H34, but the ammonia stratified spray flame lift-off positions moved downward, differing from those of 100% nC16H34 spray. This resulted from the second ammonia layer’s low reaction rate, large latent heat, and specific heat compared with those of nC16H34. The initially formed burnt gas lump significantly influenced the flame lift-off position in the ammonia-dominated jet-stream second layer. The researchers concluded that developing initial burnt gas lump formation methods and third-layer combustion control is crucial for ammonia stratified spray flames.
Bakir et al. [154] computationally investigated the autoignition enhancement of liquid-ammonia spray with dissolved hydrogen to simplify engine hardware and infrastructure for ammonia-hydrogen dual-fuel combustion. Trace dissolved hydrogen successfully facilitated ammonia ignition within reasonable residence times, with complete flame development being achievable through extended injection duration. Ignition timing was controlled by spray evaporation and mixing rather than H2 concentration. Chemical kinetic pathway analysis revealed that the OH radicals from H2 oxidation dominantly and necessarily oxidized the NH3, generating highly reactive intermediates and promoting the desired ignition. The emissions contained high NOx and N2O concentrations before and after ignition, particularly NO. This dissolved hydrogen-liquid ammonia blend offers cost and convenience advantages for improving liquid-ammonia spray combustion.
Cheng et al. [84] recently proposed a twin-fuel co-injection strategy using hydrogen as a combustion enhancer (Fig. 32). Their system blends high-pressure liquid NH3 (> 300 bar) with gaseous H2 (> 100 bar) before injection, enabling precise liquid-to-gas ratio control and creating homogeneous high-/low-reactivity fuel mixtures with controlled stratification. The advantages include flexible NH3/H2 ratio control based on injection pressure and duration, enhanced liquid NH3 atomization and fuel-air mixture quality through premixed injection, and improved combustion stability due to homogeneous NH3-H2 mixtures reducing the flame instability. The system can incorporate pre-chambers as volumetric ignition sources for NH3-H2 blends, further enhancing ignition and combustion stability.
Although the twin-fuel co-injection strategy offers significant advantages—such as superior combustion control with faster burn rates and lower cycle-to-cycle variations—and despite aligning with decarbonization goals by eliminating carbon-based pilot fuel, it presents several practical challenges. These include the complexity of storing liquid NH3 and gaseous H2 in dual tank systems with different safety requirements, which may increase system volume and weight by 40%-60% compared with conventional diesel systems. Precise control of the NH3/H2 ratio under transient engine conditions is difficult, with potential variations exceeding ±15%, which could lead to combustion instability or excessive NOx formation. Additionally, the initial cost of the dual-fuel system is estimated to be 2.5-3.0 times higher than that of diesel systems due to the high-pressure hydrogen components and sophisticated control systems, with the hydrogen infrastructure requirement further limiting its near-term adoption.
3.5. Summary
This section reviews liquid-ammonia injections and spray characteristics under high-pressure, high-temperature conditions, covering both non-reactive and reactive scenarios. CVC studies show that ammonia spray behavior is highly sensitive to ambient conditions, with penetration and evaporation rates differing significantly from those of diesel, necessitating optimized injection strategies and nozzle designs. Ammonia’s high auto-ignition temperature and low reactivity require pilot ignition strategies, typically using diesel. While diesel-piloted combustion achieves stable ignition, flame stabilization remains challenging due to ammonia’s slow kinetics and high vaporization heat. Optical techniques and models have been used to analyze flame development, heat release, and pollutant formation (particularly NOx and N2O). Innovative approaches include TJI, stratified injections, and blending ammonia with hydrogen or reactive fuels to increase combustion efficiency and stability while reducing emissions. These findings provide crucial insights for developing ammonia-fueled engines and sustainable transportation systems.
Comparing the maturity levels of various combustion-enhancement technologies, TJI appears most promising for immediate implementation due to its ability to extend the critical oxygen concentration for stable ignition. In addition, TJI ignition [151] achieves 40% faster flame propagation than stratified injection [152]; however, its scalability to multi-cylinder engines remains unproven. Stratified injection systems, while theoretically superior in reducing N2O emissions by 60% [152] through optimized temperature gradients, present practical challenges in precise fuel layer control and near-term adoption. The hydrogen co-injection approach offers excellent ignition enhancement but introduces significant infrastructure complexity. Based on the reviewed literature, pilot-ignited diesel injection remains the most mature technology, although it contradicts the carbon-neutral objective unless renewable diesel alternatives are employed.
4. Application of liquid-ammonia injection in engines
Research on liquid-ammonia applications in real engines began around 2022, prior to which ammonia was predominantly used in a gaseous state as engine fuel. Since 2022, researchers have explored liquid-ammonia injection feasibility and effectiveness in engine combustion. Due to ammonia’s low reactivity, single-fuel-mode combustion remains unfeasible, making dual-fuel mode the currently recognized effective approach for ammonia engine combustion. A summary of ammonia dual-fuel engine performance and emissions from other relevant references [165], [166], [167], [168], [169], [170], [171], [172] is provided in Table S1 in the Appendix.
4.1. Research progress
Zhou et al. [155] conducted a comparative analysis of the low-pressure injection dual-fuel (LPDF) mode and high-pressure injection dual-fuel (HPDF) mode, as illustrated in Fig. 33. In LPDF mode, low-pressure ammonia injection into intake ports or cylinders creates premixed combustible mixtures with air. High-reactivity diesel initiates combustion, with premixed turbulent flame-propagation speed controlling the process, particularly at small diesel fractions. Conversely, HPDF mode involves injecting both diesel and liquid ammonia near the top dead center, where fuel-air mixing rates dominate in-cylinder heat-release processes, similar to conventional diesel engines.
In addition, the researchers developed numerical models for pilot diesel-ignited ammonia dual-fuel engines in premixed and high-pressure spray combustion modes, based on commercial low-speed marine diesel engines and validated against experimental data on spray mixture formation, ignition delay, in-cylinder pressure, and pollutant emissions. Fig. 34 [135] compares the indicated thermal efficiency (ITE) and emissions across diesel-only, LPDF, and HPDF modes. Both dual-fuel modes achieved approximately 97% GHG reduction compared with diesel-only operation through substantial diesel replacement with carbon-free ammonia. Optimized LPDF increased the ITE by 5.1% through reduced cooling losses but raised NOx emissions 140.6% above diesel-only levels. Optimized HPDF achieved a diesel-comparable ITE while reducing NOx emissions by 47%. Both optimized configurations maintained 97% GHG reduction, with negligible unburned ammonia emissions.
Nadimi et al. [156] developed and validated a CFD model to study ammonia/biodiesel sprays, combustion characteristics, and emissions formation. Direct liquid-ammonia injection with biodiesel in single-cylinder diesel engines significantly reduced GHG emissions and combustion chamber temperatures. A maximum 50% ammonia energy ratio (AER) was achievable without significant performance drops. Retarding the ammonia injection timing from −25 to −10 crank angle degrees (CAD) reduced NOx, CO, and ammonia emissions by 31.4%, 39.6%, and 31.3%, respectively. Optimal conditions occurred with ammonia injection at −10 CAD and biodiesel at −16 CAD with 50% AER. In addition, this team modified a single-cylinder diesel engine with dual common rail systems for direct liquid-ammonia/pilot biodiesel injection [157]. Welding three nozzles from six-nozzle injectors reduced NH3 and CO emissions by 29.2%. Late fuel injection increased the particulate matter from 10.5 to 15.2 mg·m−3 due to fuel-rich high-temperature zones but reduced NOx and CO emissions by 1.4 and 4.4 g·(kW·h)−1, respectively, compared with early biodiesel start of injection (SOI).
Zhang et al. [158] experimentally validated ammonia/diesel dual DI in two-stroke low-speed engines using a diesel jet-controlled compression ignition (JCCI) concept, examining engine performance and emissions under various ammonia injection quantities, timings, and diesel injection timings. The study verified feasibility, demonstrating precise liquid-phase ammonia ignition control by diesel jets resulting in two-stage combustion dominated by diesel then ammonia. Adjusting the ammonia injection timing enabled different combustion modes: premixed, diffusion, and premix-diffusion co-combustion. Excessive timing advancement created super-lean mixtures, while absolute diffusion combustion extended the total combustion duration. The combustion phase showed high sensitivity to diesel injection timing. Robust ammonia combustion control shortened the total combustion duration compared with that of pure diesel, improving the ITE. Soot and CO emissions decreased through enhanced spray plume interaction and improved diesel atomization/evaporation promoting oxidation during ammonia combustion. NOx emissions increased due to the combined thermal NOx (diesel) and fuel NOx (ammonia) contributions. Adjusting the diesel injection timing or extending the combustion duration managed emissions but negatively impacted the ITE. Injection timing at −8° crank angle after top dead center (CA ATDC) provided optimal ITE and emission performance balance for both fuels.
In 2024, research on liquid-ammonia spray engines began to increase. Dong et al. [159] developed and validated a numerical model for ammonia/diesel dual DI two-stroke engines using bench test results. Their findings showed that optimizing the spray spatial arrangement, intake temperatures, and EGR rates effectively controlled nitrogenous and GHG emissions while maintaining high thermal efficiency. Optimal conditions included an 8° spray arrangement angle, 355-K intake temperature, and 20% EGR rate, effectively balancing unburned ammonia, NOx, N2O, and GHG emission reductions. Fig. 35 [159] illustrates the EGR rate effects on CO2 and GHG emissions, showing that both CO2 and N2O emissions increased with rising EGR rates, elevating overall GHG levels. At −4 °CA ATDC after the start of injection (ASOI) with a 30% EGR rate, GHG emissions exceeded diesel-only mode levels.
The same team further numerically explored liquid-ammonia diffusion combustion (LADC) based on a modified marine two-stroke engine [160]. LADC significantly increased combustion speed, reduced soot emissions by 99%, and advanced CA90 by 11.8 °CA compared with pure diesel. Optimal combustion and emission characteristics occurred at 50%-70% AER. Fig. 36 [160] shows the effects of AER on NOx and soot emissions: Increasing AER raised NOx while reducing soot. NO emissions decreased and N2O increased significantly at high AER due to the lower average cylinder temperatures reducing thermal NO, with amino groups reacting with NO to form nitrogen at 1100-1400 K. The higher ammonia reaction temperatures generated fuel-type NO, decreasing NO emissions at high AER. Diesel combustion produced negligible N2O, with elevated-temperature ammonia combustion being the primary source. Higher AER increased the liquid-ammonia injection, raising ammonia combustion and N2O emissions while reducing soot through decreased diesel injection.
Park et al. [161] investigated efficiency improvements by introducing oxygen into the intake air of a 2.5 L ammonia-fueled DI spark-ignition engine, experimentally analyzing the combustion, exhaust, and efficiency. Liquid ammonia was injected at 15 MPa, with hydrogen and oxygen introduced via intake manifold. Oxygen addition significantly increased the combustion stability and thermal efficiency, particularly under low-speed, low-load conditions. Increasing the oxygen concentration reduced the ignition delay and accelerated the combustion speed, improving stability and efficiency. Oxygen showed more pronounced effects than hydrogen on ignition delay and combustion speed. While increasing the oxygen concentration from 25% to 30% further improved the combustion stability and reduced unburned ammonia emissions, additional efficiency improvements were insignificant.
Following their previous work, the team experimentally studied diesel post-injection timing and quantity effects on combustion, emissions, and efficiency in a 12 L ammonia-diesel dual-fuel CI single-cylinder marine engine [162]. Conducting 40% diesel post-injection near a micro-pilot injection in an ammonia-diesel dual-fuel four-stroke marine engine reduced the unburned hydrocarbons by 18.9%, unburned ammonia by 11.2%, and N2O by 19.4% compared with conventional dual-fuel combustion, while maintaining equivalent thermal efficiency. Dual-fuel combustion surpassed neat diesel in GWP and CO2, NO, and CO emissions. Although post-injection slightly increased the GWP due to higher CO2, it significantly reduced the overall NOx emissions including N2O and NO, improving combustion efficiency by lowering the unburned ammonia and total hydrocarbons.
Park et al. [66] conducted engine experiments comparing combustion, thermal efficiency, and exhaust emissions at various CRs using liquid-ammonia DI with pilot-ignited diesel in a 12.5 L single-cylinder marine engine. Increasing the CR in an ammonia-diesel dual-fuel marine engine significantly improved the thermal efficiency by over 5.8% and increased combustion stability. Fig. 37 [66] shows the effects of the CR on the in-cylinder pressure and HRR, demonstrating that higher CRs increased the in-cylinder pressure and temperature, accelerating combustion efficiency. While the maximum in-cylinder pressure increased with higher CR, the peak pressure timing remained consistent. Higher CR advanced heat-release initiation, although the HRR varied due to fuel atomization differences. At CR 13, lower ambient pressure and temperature produced higher premixed flame heat release than at CR 14. Maximum HRR remained consistent across CRs despite ammonia flow-rate variations. Improved diesel flame characteristics increased the initial ammonia ignition and combustion speed, with minimal impact on late combustion and unburned ammonia emissions. A higher CR improved the thermal efficiency and combustion stability, although the reduction of unburned ammonia emissions was insignificant.
In addition, engine results from Mi et al. [163] show that the ammonia post-injection strategy effectively mitigated the suppression effect of high-concentration ammonia on diesel ignition, enabling advanced ignition timing and increased peak HRR. Combined with intake heating, this approach achieved an AER of 86% at an indicated mean effective pressure (IMEP) of 4 bar while improving the ITE by 8% and substantially reducing unburned ammonia. A feasible expansion strategy for achieving high AER in liquid ammonia-diesel dual-fuel combustion is shown in Fig. 38 [163]. At IMEP = 6 bar, the ammonia post-injection strategy with a controlled SOI for diesel and second ammonia injection timing, combined with appropriate post-injection ammonia quantity, enables an 80% substitution rate. At IMEP = 4 bar, intake heating increases the in-cylinder activity, while ammonia pre-injection is employed at low AER and post-injection at high AER. This approach achieves higher thermal efficiency and lower ammonia emissions across a wide operating range with relatively low N2O emissions.
Yang et al. [164] recently proposed an innovative ammonia thermal atmosphere compression ignition (TACI) combustion mode, as shown in Fig. 39. N-heptane is supplied to the intake manifold and enters the cylinder with the intake air during the intake stroke, where it auto-ignites under compression and undergoes homogeneous charge-compression ignition (HCCI) combustion. Following n-heptane combustion, ammonia is injected into the cylinder and auto-ignites due to the active thermal atmosphere comprising high pressure, high temperature, and active radicals. The ammonia TACI combustion mode thus consists of three stages: the n-heptane low-temperature reaction, n-heptane high-temperature combustion, and ammonia diffusion combustion. The ammonia injection timing critically affects the diffusion combustion characteristics and thermal efficiency in TACI mode, with optimal timing being determined by the combustion phase. Over-delayed injections reduce stability and efficiency while increasing N2O and NH3 emissions. The high in-cylinder temperatures from high-reactivity fuel combustion enable ultra-low N2O and NH3 emissions, although CO and hydrocarbons require aftertreatment and SCR for NOx compliance. With high ammonia substitution rates and low N2O emissions, TACI combustion demonstrates excellent GHG reduction potential, achieving over 70% reduction.
4.2. Summary
This section examines liquid-ammonia injections in engine-application research from 2022 onward. While liquid-ammonia injection in engines demonstrates technical feasibility with the potential for 97% CO2 reduction, significant challenges hinder its practical viability. Engine performance shows mixed results: Some studies report efficiency gains up to 5.1%, while others show 2.5% penalties compared with diesel-only operation, and all configurations require diesel pilot injection (3%-20% by energy), preventing complete carbon neutrality. Two-stroke engine tests show improved efficiency and reduced soot/CO, although NOx increases. Numerical models indicate that spray arrangement, intake temperatures, and EGR rates can balance emissions while maintaining efficiency. Oxygen-enriched air intake improves combustion stability, especially at low speeds/loads. Higher CRs increase efficiency and stability, with minimal impact on unburned ammonia. TACI mode displays efficient ammonia diffusion combustion in a high-temperature atmosphere.
Most concerning are the emissions trade-offs: NOx emissions increase dramatically (up to 140.6% in some modes) due to both thermal and fuel-bound nitrogen sources, while N2O formation—with a GWP 265-298 times that of CO2—can result in higher total GHG emissions than diesel under certain conditions. Combined with persistent ammonia-slip issues, narrow operational windows requiring precise control, and infrastructure challenges for high-pressure liquid-ammonia handling, these findings suggest that, despite its carbon-reduction potential, liquid ammonia’s role in transportation may be limited to niche applications unless breakthrough emission-control technologies are developed. Critical comparisons of NOx and N2O mitigation strategies reveal varying levels of effectiveness:
Optimized combustion timing. Retarding ammonia injection from −25 to −10 CAD reduces NOx by 31.4% [156] but may increase unburned ammonia. The trade-off between NOx reduction and combustion efficiency remains a key challenge.
EGR implementation. A 20% EGR rate effectively reduces NOx via dilution and temperature reduction [159]. However, increasing the EGR beyond 30% can elevate N2O emissions, potentially negating GHG benefits. The optimal EGR rate appears to be 15%-25%, based on multiple studies.
Stratified injection. The three-layer approach by Ichikawa et al. [152] dramatically reduces N2O emissions by eliminating low-temperature regions. Quantitatively, N2O emissions decreased from 150 ppm to below 20 ppm, representing an 87% reduction.
Post-injection strategies. Diesel post-injection at 40% of pilot quantity reduces N2O by 19.4% and unburned ammonia by 11.2% [162], while maintaining thermal efficiency. This provides the best balance between emission reduction and performance.
SCR systems. While highly effective for NOx reduction (up to 90% efficiency), SCR systems present challenges with ammonia slip and require careful temperature management (300-400 °C optimal range) for ammonia exhaust streams. EGR effectively suppresses thermal NOx but elevates N2O and CO2 emissions, reducing ITE by 2%-4% [155]. Stratified injection curbs N2O by 60% via temperature homogenization [152], yet injector complexity impedes adoption. SCR post-treatment achieves > 70% NOx reduction [48] but requires ammonia dosing calibration to avoid ammonia slip.
5. Summaries, future research directions, and perspectives
5.1. Summaries
This review presents a comprehensive synthesis of liquid-ammonia injection research, systematically analyzing the complete spectrum from fundamental spray physics to engine implementation. It uniquely establishes critical connections between microscopic phenomena (flash boiling, cavitation, droplet dynamics) and macroscopic engine performance, demonstrating how ammonia’s high latent heat causes spray core temperatures to drop, directly explaining extended ignition delays and 40% lower heat release at 900 K versus 1000 K. The review provides the first comparative evaluation of emerging injection strategies (LPDF vs HPDF, spray angles, stratified injection) with quantitative metrics, while identifying precise research gaps unique to liquid ammonia, including cavitation behavior, supercritical phase transitions, and thermal decomposition during atomization. By integrating breakthrough findings from 2022 to 2024, including first-ever successful auto-ignition demonstrations and novel injection and ignition strategies (TJI, stratified fuel injection, hydrogen co-injection, and TACI), this work fundamentally shows the feasibility landscape for liquid ammonia as an engine fuel and offers practical guidance beyond theoretical discussions.
5.2. Future research directions
In-depth understanding of spray dynamics. Key knowledge gaps include ammonia’s unique thermophysical properties and its influence on cavitation inception and development, flash-boiling dynamics under engine conditions, supercritical phase transitions, surface tension effects on droplet breakup, evaporation kinetics in reactive environments, aerodynamic breakup mechanisms, multi-component spray behavior with fuel blends, and thermal decomposition effects during atomization.
Advanced combustion strategies. Further research is needed to develop advanced combustion strategies that can optimize the trade-offs between thermal efficiency, combustion stability, and pollutant emissions. This includes exploring novel ignition methods, such as laser ignition or advanced spark-ignition systems, to address ammonia’s high ignition temperature and low reactivity.
Engine hardware and infrastructure. Research should focus on developing specialized engine hardware tailored for liquid-ammonia combustion. This includes designing injectors (prioritize cavitation suppression and an optimized aspect ratio for flash-boiling control), fuel systems, and combustion chambers that can handle ammonia’s unique properties, such as its high latent heat and narrow flammability range, while ensuring reliable and efficient operation.
Emission-control technologies. Research should investigate advanced emission-control technologies to mitigate the formation of NOx, N2O, and unburned NH3 during ammonia combustion. Utilizing the existing ammonia from the fuel system to replace urea in SCR can simplify the system while simultaneously reducing residual unburned ammonia and NOx emissions. Additionally, the catalyst in SCR operates optimally at around 600-700 K, allowing effective use of engine exhaust heat to maintain catalyst temperature and achieve waste heat recovery.
Thermodynamic and fluid dynamic models. Development of more accurate thermodynamic and fluid dynamic models for ammonia sprays is essential. These models should account for ammonia’s unique phase-change and resulting turbulence-chemistry interaction to improve the prediction of spray characteristics and combustion performance.
Material compatibility and safety infrastructure. Future research must develop ammonia-resistant materials for fuel injectors (nickel-based alloys, specialized polymers) and compatible seals (EPDM, fluoropolymers) capable of 10 000 + operational hours. Given ammonia’s toxicity (IDLH: 300 ppm), systems require sub-second detection at < 5 ppm with automatic engine shutdown. Safety frameworks must include ventilation (6+ air changes per hour), emergency shutdown with purging systems, personal protective equipment (PPE) specifications, spill containment for liquid ammonia, and protocols for bunkering and lube-oil contamination using amine-resistant lubricants.
Long-term durability and scalability challenges. Transitioning ammonia-fueled engines from the laboratory to commercial deployment presents critical durability and scalability issues. Injector degradation occurs due to nozzle erosion from ammonia’s low lubricity, requiring research into wear mechanisms, surface treatments, and maintenance schedules. Ammonia contamination neutralizes acidic oil additives, reducing lubricant effectiveness compared with diesel and necessitating ammonia-compatible formulations. Ammonium nitrate deposits from combustion differ from carbon deposits and may impact heat-transfer and CRs. Scaling to multi-cylinder engines introduces cylinder-to-cylinder ammonia distribution variations, complex dual-fuel control systems, and thermal-management challenges from non-uniform heat release. Demonstration programs on six-plus cylinder engines are essential to validate scalability solutions.
5.3. Perspectives
The transition to liquid ammonia as a fuel represents a significant step toward decarbonizing the transportation and energy sectors. While substantial challenges remain, ongoing research and technological advancements are gradually overcoming these obstacles. Liquid ammonia’s potential as a carbon-free fuel—combined with its high energy density and existing infrastructure—makes it an attractive option for future energy systems, particularly in maritime shipping and heavy-duty transportation.
Future research should prioritize the development of integrated solutions that combine advanced combustion technologies, optimized engine designs, and effective emission-control strategies. Collaboration between academia, industry, and policymakers will be crucial in driving the adoption of ammonia as a sustainable fuel alternative. By addressing the technical, economic, and safety challenges associated with liquid-ammonia combustion, we can pave the way for a cleaner and more sustainable energy future.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
SpatolisanoE, PellegriniLA, de AngelisAR, CattaneoS, RoccaroE. Ammonia as a carbon-free energy carrier: NH3 cracking to H2. Ind Eng Chem Res2023; 62 (28):10813-27.
[2]
WuH, Ben HouidiM, AlmatrafiF, WuB, DuJ, MagnottiG, et al. Hydrogen jet characteristics with an outwardly opening piezo injector. Phys Fluids2025; 37 (3):036103.
[3]
DavidWI, AgnewGD, Bañares-AlcántaraR, BarthJ, HansenJB, BréquignyP, et al. 2023 roadmap on ammonia as a carbon-free fuel. J Phys Energy 2024; 6 (2):021501.
[4]
MacFarlaneDR, CherepanovPV, ChoiJ, SuryantoBH, HodgettsRY, BakkerJM, et al. A roadmap to the ammonia economy. Joule2020; 4(6):1186-205.
[5]
Valera-MedinaA, Amer-HatemF, AzadAK, DedoussiI, De JoannonM, FernandesR, et al. Review on ammonia as a potential fuel: from synthesis to economics. Energy Fuels2021; 35(9):6964-7029.
[6]
IshaqH, CrawfordC. Review of ammonia production and utilization: enabling clean energy transition and net-zero climate targets. Energy Convers Manag2024; 300:117869.
[7]
Ammonia technology roadmap:towards more sustainable nitrogen fertiliser production. Report. Paris: International Energy Agency; 2021.
[8]
SalmonN, Bañares-AlcántaraR. Green ammonia as a spatial energy vector: a review. Sustain Energy Fuels2021; 5(11):2814-39.
[9]
ChehadeG, DincerI. Progress in green ammonia production as potential carbon-free fuel. Fuel2021; 299:120845.
[10]
OlabiA, AbdelkareemMA, Al-MurisiM, ShehataN, AlamiAH, RadwanA, et al. Recent progress in green ammonia: production, applications, assessment; barriers, and its role in achieving the sustainable development goals. Energy Convers Manag2023; 277:116594.
[11]
GuerraCF, Reyes-BozoL, VyhmeisterE, CaparrósMJ, SalazarJL, Clemente-JulC. Technical-economic analysis for a green ammonia production plant in Chile and its subsequent transport to Japan. Renew Energy2020; 157:404-14.
[12]
KlerkeA, ChristensenCH, NørskovJK, VeggeT. Ammonia for hydrogen storage: challenges and opportunities. J Mater Chem2008; 18(20):2304-10.
[13]
Valera-MedinaA, XiaoH, Owen-JonesM, DavidWI, BowenP. Ammonia for power. Pror Energy Combust Sci2018; 69:63-102.
[14]
GiddeyS, BadwalS, MunningsC, DolanM. Ammonia as a renewable energy transportation media. ACS Sustain Chem Eng2017; 5(11):10231-9.
[15]
DolanRH, AndersonJE, WallingtonTJ. Outlook for ammonia as a sustainable transportation fuel. Sustain Energy Fuels2021; 5(19):4830-41.
[16]
SchüthF, PalkovitsR, SchlöglR, SuDS. Ammonia as a possible element in an energy infrastructure: catalysts for ammonia decomposition. Energy Environ Sci2012; 5(4):6278-89.
[17]
TawalbehM, MurtazaSZ, Al-OthmanA, AlamiAH, SinghK, OlabiAG. Ammonia: a versatile candidate for the use in energy storage systems. Renew Energy2022; 194:955-77.
[18]
ZhouX, LiT, YangW. Ammonia-hydrogen engine with single ammonia fuel supply. Joule2025; 9(5):101922.
[19]
SunS, JiangQ, ZhaoD, CaoT, ShaH, ZhangC, et al. Ammonia as hydrogen carrier: advances in ammonia decomposition catalysts for promising hydrogen production. Renew Sustain Energy Rev2022; 169:112918.
[20]
AfifA, RadenahmadN, CheokQ, ShamsS, KimJH, AzadAK. Ammonia-fed fuel cells: a comprehensive review. Renew Sustain Energy Rev2016; 60: 822-35.
[21]
MorlanésN, KatikaneniSP, PaglieriSN, HaraleA, SolamiB, SarathySM, et al. A technological roadmap to the ammonia energy economy: current state and missing technologies. Chem Eng J2021; 408:127310.
[22]
MachajK, KupeckiJ, MalechaZ, MorawskiA, SkrzypkiewiczM, StanclikM, et al. Ammonia as a potential marine fuel: a review. Energy Strategy Rev2022; 44:100926.
[23]
AnZ, XingJ, KuroseR. Recent progresses in research on liquid ammonia spray and combustion: a review. Appl Energy Combust Sci2024:100293.
[24]
DimitriouP, JavaidR. A review of ammonia as a compression ignition engine fuel. Int J Hydrogen Energy2020; 45(11):7098-118.
[25]
StarkmanES, NewhallH, SuttonR, MaguireT, FarbarL. Ammonia as a spark ignition engine fuel: theory and application. SAE Trans1966:660155.
[26]
CorneliusW, HuellmantelLW, MitchellHR. Ammonia as an engine fuel. SAE Trans1966:650052.
[27]
ChorowskiM, LepszyM, MachajK, MalechaZ, PorwisiakD, PorwisiakP, et al. Challenges of application of green ammonia as fuel in onshore transportation. Energies2023; 16(13):4898.
[28]
PowdersMT, LuqmaniBA, PidouM, ZhuM, McAdamEJ. The use of ammonia recovered from wastewater as a zero-carbon energy vector to decarbonise heat, power and transport-a review. Water Res2024; 268:122649.
[29]
NottonG, NivetML, VoyantC, PaoliC, DarrasC, MotteF, et al. Intermittent and stochastic character of renewable energy sources: consequences, cost of intermittence and benefit of forecasting. Renew Sustain Energy Rev2018; 87:96-105.
[30]
KoonsE. Ammonia fuel: advantages and disadvantages. Report. Energy Tracker Asia; 2023.
[31]
HuangY, NieZ, ZhangL, LiuX, FengJ, YinY, et al. Ammonia fuel: a pathway for carbon-neutral of the transportation sector. Chain 2024;1(2):113-37.
[32]
KumarL, SleitiAK. Systematic review on ammonia as a sustainable fuel for combustion. Renew Sustain Energy Rev2024; 202:114699.
[33]
OjeladeOA, ZamanSF, NiBJ. Green ammonia production technologies: a review of practical progress. J Environ Manage2023; 342:118348.
[34]
MallouppasG, IoannouC, YfantisEA. A review of the latest trends in the use of green ammonia as an energy carrier in maritime industry. Energies2022; 15(4):1453.
[35]
Al-AboosiFY, El-HalwagiMM, MooreM, NielsenRB. Renewable ammonia as an alternative fuel for the shipping industry. Curr Opin Chem Eng2021; 31:100670.
[36]
FullontonA, Lea-LangtonAR, MaduguF, LarkinA. Green ammonia adoption in shipping: opportunities and challenges across the fuel supply chain. Mar Policy2025; 171:106444.
[37]
KosmajacS. MAN ES presses on with project to develop four-stroke ammonia-powered engine. Report. Schiedam: Offshore Energy; 2024.
[38]
PrevljakNH. Wärtsilä rolls out industry's first 4 -stroke ammonia engine. Report. Schiedam: Offshore Energy; 2023.
[39]
AtchisonJ. Hyundai Heavy Industries unveils marine ammonia engine. Report. Ashburn: Ammonia Energy Association; 2024.
[40]
Xinhua. China's first medium-speed high-power ammonia-fuelled engine ignites [Internet]. Beijing: Chinadaily; 2023 Dec 7 [2025 Mar 23]. Available from: https://www.chinadaily.com.cn/a/202312/07/WS65718967a31090682a5f201c.html
[41]
CorporationIHI. Carbon-neutral thermal power generation using ammonia combustion technology development and boiler design to expand use of ammonia. Report. Tokyo: IHI Corporation; 2022.
[42]
YamashitaT, AmariT, UrakataY, SumidaT, OkazakiT, TakayamaA. Development of ammonia co-firing technology for coal-fired boilers toward decarbonized society. Mitsubishi Heavy Ind Tech Rev2022; 59:4.
[43]
AtchisonJ. JERA targets 50% ammonia-coal co-firing by 2030. Report. Ashburn: Ammonia Energy Association; 2022.
[44]
TianJ, WangL, XiongY, WangY, YinW, TianG, et al. Enhancing combustion efficiency and reducing nitrogen oxide emissions from ammonia combustion: a comprehensive review. Process Saf Environ Prot2024; 183:514-43.
[45]
VerhelstS, TurnerJW, SileghemL, VancoillieJ. Methanol as a fuel for internal combustion engines. Pror Energy Combust Sci2019; 70:43-88.
[46]
YipHL, SrnaA, YuenACY, KookS, TaylorRA, YeohGH, et al. A review of hydrogen direct injection for internal combustion engines: towards carbonfree combustion. Appl Sci2019; 9(22):4842.
[47]
MazloomiK, GomesC. Hydrogen as an energy carrier: prospects and challenges. Renew Sustain Energy Rev2012; 16(5):3024-33.
[48]
KobayashiH, HayakawaA, SomarathneKKA, OkaforEC. Science and technology of ammonia combustion. Proc Combust Inst2019; 37(1):109-33.
[49]
LuQ, PengZ, ZhouS, ZhangB, ChenH, YangS. Mini-review of spray and combustion characteristics for ammonia engines. Energy Fuels2024; 38 (20):19156-73.
[50]
HeX, ShuB, NascimentoD, MoshammerK, CostaM, FernandesR. Autoignition kinetics of ammonia and ammonia/hydrogen mixtures at intermediate temperatures and high pressures. Combust Flame2019; 206:189-200.
El-AdawyM, NemitallahMA, AbdelhafezA. Towards sustainable hydrogen and ammonia internal combustion engines: challenges and opportunities. Fuel2024; 364:131090.
[53]
GoodmanJ, DhankharA, DateA, LappasP. Ammonia-air laminar flame speeds from ambient to IC engine conditions: a review. Fuel2025; 383:133769.
[54]
AlnajideenM, ShiH, NorthropW, EmbersonD, KaneS, CzyzewskiP, et al. Ammonia combustion and emissions in practical applications: a review. Carbon Neutrality2024; 3(1):1-45.
[55]
HewlettSG, Valera-MedinaA, PughDG, BowenPJ. Gas turbine co-firing of steelworks ammonia with coke oven gas or methane:a fundamental and cycle analysis. In: Proceedings of ASME Turbo Expo 2019:Turbomachinery Technical Conference and Exposition; 2019 Jun 17-21; Phoenix, AZ, USA. New York City: American Society of Mechanical Engineers; 2019.
[56]
FengR, LiZ. Current investigations on global N2O emissions and reductions: prospect and outlook. Environ Pollut2023; 338:122664.
[57]
NikiY. Reductions in unburned ammonia and nitrous oxide emissions from an ammonia-assisted diesel engine with early timing diesel pilot injection. J Eng Gas Turbine Power2021; 143(9):091014.
[58]
EyisseEF, NadimiE, WuD. Ammonia combustion: internal combustion engines and gas turbines. Energies2024; 18(1):29.
KoikeM, MiyagawaH, SuzuokiT, OgasawaraK. Ammonia as a hydrogen energy carrier and its application to internal combustion engines. J Combust Soc Jpn2016; 58(184):99-106.
[61]
YamashitaH, HayakawaA, OkuK, ColsonS, ReibelG, ChenY, et al. Visualization of liquid ammonia spray using 2p-SLIPI and comparison of liquid ammonia spray and gaseous ammonia combustion in a swirl combustor at atmospheric pressure. Fuel2024; 371:131833.
[62]
AngelilliL, Hernández PérezFE, ImHG, CiottoliPP, ValoraniM. Evaporation and clustering of ammonia droplets in a hot environment. Phys Rev Fluids2022; 7(11):114301.
[63]
WangX, BuH, ChenH, LiuJ, ChenZ, GaoJ. Numerical investigation of diesel spray combustion characteristics in the ammonia/air atmosphere. J Energy Inst2024; 116:101718.
[64]
XuL, DongP, ZhangZ, BuJ, TianJ, LongW, et al. Impact of spray interaction on ammonia/diesel dual-fuel combustion and emission under engine relevant conditions. Proc Combust Inst2024; 40(1-4):105751.
[65]
WangB, YangC, WangH, HuD, DuanB, WangY. Study on injection strategy of ammonia/hydrogen dual fuel engine under different compression ratios. Fuel2023; 334:126666.
[66]
ParkC, JangI, KimM, ParkG, KimY. Effect of high compression ratio on thermal efficiency and unburned ammonia emissions of a dual-fuel highpressure direct injection marine ammonia engine. Appl Therm Eng2025; 261:125183.
[67]
OkaforEC, YamashitaH, HayakawaA, SomarathneKKA, KudoT, TsujimuraT, et al. Flame stability and emissions characteristics of liquid ammonia spray co-fired with methane in a single stage swirl combustor. Fuel2021; 287:119433.
[68]
LiuJ, ZhangM, AnZ, WangJ, HuangZ. Effect of preheated air temperature on a liquid ammonia flash spray in a swirl combustor. Droplet2025; 4(1):e159.
[69]
ScharlV, LackovicT, SattelmayerT. Characterization of ammonia spray combustion and mixture formation under high-pressure, direct injection conditions. Fuel2023; 333:126454.
[70]
ScharlV, SattelmayerT. Ignition and combustion characteristics of diesel piloted ammonia injections. Fuel Commun2022; 11:100068.
[71]
ElbazAM, WangS, GuibertiTF, RobertsWL. Review on the recent advances on ammonia combustion from the fundamentals to the applications. Fuel Commun2022; 10:100053.
[72]
OtomoJ, KoshiM, MitsumoriT, IwasakiH, YamadaK. Chemical kinetic modeling of ammonia oxidation with improved reaction mechanism for ammonia/air and ammonia/hydrogen/air combustion. Int J Hydrogen Energy2018; 43(5):3004-14.
[73]
OkaforEC, NaitoY, ColsonS, IchikawaA, KudoT, HayakawaA, et al. Experimental and numerical study of the laminar burning velocity of CH4-NH3 -air premixed flames. Combust Flame2018; 187:185-98.
[74]
CaiT, ZhaoD, GutmarkE. Overview of fundamental kinetic mechanisms and emission mitigation in ammonia combustion. Chem Eng J2023; 458:141391.
[75]
XiaoH, Valera-MedinaA, BowenPJ. Study on premixed combustion characteristics of co-firing ammonia/methane fuels. Energy2017; 140:125-35.
[76]
LiJ, LaiS, ChenD, WuR, KobayashiN, DengL, et al. A review on combustion characteristics of ammonia as a carbon-free fuel. Front Energy Res2021; 9:760356.
[77]
WangY, WangX, ZengW, WangW, SongZ. Advancements in turbulent combustion of ammonia-based fuels: a review. Int J Hydrogen Energy2024; 88:1332-55.
[78]
ZhangM, WeiX, AnZ, OkaforEC, GuibertiTF, WangJ, et al. Flame stabilization and emission characteristics of ammonia combustion in labscale gas turbine combustors: recent progress and prospects. Prog Energy Combust Sci2025; 106:101193.
[79]
Valera-MedinaA, MarshR, RunyonJ, PughD, BeasleyP, HughesT, et al. Ammonia-methane combustion in tangential swirl burners for gas turbine power generation. Appl Energy2017; 185:1362-71.
[80]
WeiD, FangH, TangH, WangY, WeiG, ZhouH. Experimental study of combustion instability and emission characteristics of ethanol/ammonia cofiring swirl flame. Fuel2024; 362:130786.
[81]
ZhangX, TianJ, LiX, YinS, CuiZ, YangH, et al. Fundamental study on the factors influencing the stability of ammonia combustion and the effects of pre-chamber on the combustion characteristics. Appl Therm Eng2024; 254:123812.
[82]
ZhuX, DuJ, YuZ, ChengYB, WangY. $\mathrm{NO}_{x}$ emission and control in ammonia combustion: state-of-the-Art review and future perspectives. Energy Fuels2023; 38(1):43-60.
[83]
ChiongMC, ChongCT, NgJH, MashrukS, ChongWWF, SamiranNA, et al. Advancements of combustion technologies in the ammonia-fuelled engines. Energy Convers Manag2021; 244:114460.
[84]
ChengQ, MuhammadA, KaarioO, AhmadZ, MarttiL. Ammonia as a sustainable fuel: review and novel strategies. Renew Sustain Energy Rev2025; 207:114995.
[85]
KangL, PanW, ZhangJ, WangW, TangC. A review on ammonia blends combustion for industrial applications. Fuel2023; 332:126150.
[86]
PanS, MaJ, ChenX, YangW, LiangC. A simplified reaction model for combustion of ammonia. Fuel2025; 383:133818.
[87]
ZhuY, CurranHJ, GirheS, MurakamiY, PitschH, SenecalK, et al. The combustion chemistry of ammonia and ammonia/hydrogen mixtures: a comprehensive chemical kinetic modeling study. Combust Flame2024; 260:113239.
[88]
MathieuO, PetersenEL. Experimental and modeling study on the hightemperature oxidation of ammonia and related $ \mathrm{NO}_{x}$ chemistry. Combust Flame2015; 162(3):554-70.
[89]
StagniA, CavallottiC, ArunthanayothinS, SongY, HerbinetO, Battin-LeclercF, et al. An experimental, theoretical and kinetic-modeling study of the gasphase oxidation of ammonia. React Chem Eng2020; 5(4):696-711.
[90]
SherE, Bar-KohanyT, RashkovanA. Flash-boiling atomization. Pror Energy Combust Sci2008; 34(4):417-39.
[91]
Bar-KohanyT, LevyM. State of the art review of flash-boiling atomization. At Sprays2016; 26(12):1259-305.
[92]
WangL, WangF, FangT. Flash boiling hollow cone spray from a GDI injector under different conditions. Int J Multiph Flow2019; 118:50-63.
[93]
XuM, ZhangY, ZengW, ZhangG, ZhangM. Flash boiling: easy and better way to generate ideal sprays than the high injection pressure. SAE Int J Fuel Lubr2013; 6(1):137-48.
[94]
PolancoG, HoldøAE, MundayG. General review of flashing jet studies. J Hazard Mater2010; 173(1-3):2-18.
[95]
ChangM, LeeZ, ParkS, ParkS. Characteristics of flash boiling and its effects on spray behavior in gasoline direct injection injectors: a review. Fuel2020; 271:117600.
[96]
LiX, WangS, YangS, QiuS, SunZ, HungDL, et al. A review on the recent advances of flash boiling atomization and combustion applications. Pror Energy Combust Sci2024; 100:101119.
[97]
WangS, QiuS, LiX, ZhangP. Modeling non-monotonic variation of plume angle with superheat index of flash boiling spray. Energy2024; 306:132515.
[98]
DuJ, MohanB, SimJ, FangT, RobertsWL. Study of spray structure under flash boiling conditions using 2phase-SLIPI. Exp Fluids2021; 62(1):1-17.
[99]
ZengW, XuM, ZhangG, ZhangY, ClearyDJ. Atomization and vaporization for flash-boiling multi-hole sprays with alcohol fuels. Fuel2012; 95:287-97.
[100]
LiS, LiT, WangN, ZhouX, ChenR, YiP. An investigation on near-field and farfield characteristics of superheated ammonia spray. Fuel2022; 324:124683.
[101]
LiS, LiT, WangN, ZhouX, YiP, ChenR. Effect of near-field characteristics on the two-phase distribution of superheated ammonia spray. Int J Engine Res2023; 24(10):4373-81.
[102]
MaY, ZhongW, LaiS, ChenJ, PachiannanT, ZhangL, et al. Experimental study on the flash boiling spray characteristics and jet fluctuation of high-pressure direct injection liquid ammonia. Appl Therm Eng2024; 257:124032.
[103]
FangY, MaX, ZhangY, LiY, ZhangK, JiangC, et al. Experimental investigation of high-pressure liquid ammonia injection under non-flash boiling and flash boiling conditions. Energies2023; 16(6):2843.
[104]
FangY, ZhangK, MaX, ZhangY, XuL, LiY, et al. Droplet measurement of high-pressure liquid ammonia injection using PDPA. SAE Technical Paper; 2023:2023-01-1637.
[105]
LiuX, YaoX, WangZ, TangC. Single hole ammonia spray macroscopic and microscopic characteristics at flare and transition flash boiling regions. Appl Therm Eng2023; 235:121443.
[106]
LiS, LiuS, WangN, LiT, ChenR, YiP, et al. Atomization and evaporation characteristics of liquid ammonia spray under engine intake stroke conditions. Energy2025; 316:134589.
[107]
ZhongW, ChenJ, LiC, HuangY, PachiannanT, JiangZ, et al. Visualization study on flash boiling spray characteristics of high-pressure liquid ammonia with different nozzle diameters. Fuel2024; 367:131525.
[108]
ShenL, LeachF. Effect of ambient pressure on ammonia sprays using a single hole injector. SAE Tech Pap2024:2024-01-618.
[109]
ColsonS, YamashitaH, OkuK, SomarathneKDKA, KudoT, HayakawaA, et al. Study on the effect of injection temperature and nozzle geometry on the flashing transition of liquid ammonia spray. Fuel2023; 348:128612.
[110]
DesclauxA, PeléR, HespelC, Mounaïm-RousselleC. Liquid Ammonia injection on single hole injector:effect of initial conditions on flash boiling process. In:Proceedings of 32nd European Conference on Liquid Atomization and Spray System; 2023 Sep 4-7; Napoli, Italy. 2023. Hannover: TIB; 2023.
[111]
BjørgenKO, DesclauxA, Mounaïm-RousselleC, HespelC. Experimental characterization of superheated ammonia spray from a single-hole ECN spray M injector. J Eng Gas Turbines Power2025; 147(8):081018.
[112]
HiroyasuH, AraiM. Structures of fuel sprays in diesel engines. SAE Tech Pap 1990. 900475.
[113]
NaberJD, SiebersDL. Effects of gas density and vaporization on penetration and dispersion of diesel sprays. SAE Tech Pap1996:960034.
[114]
WuH, ZhangF, ZhangZ. Fundamental spray characteristics of air-assisted injection system using aviation kerosene. Fuel2021; 286:119420.
[115]
HuangZ, WangH, LuoK, FanJ. Large eddy simulation investigation of ammonia spray characteristics under flash and non-flash boiling conditions. Appl Energy Combust Sci2023; 16:100220.
[116]
AnZ, XingJ, KuroseR. Numerical study on the phase change and spray characteristics of liquid ammonia flash spray. Fuel2023; 345:128229.
[117]
WangJ, WangH, ZhengZ, YaoM. Numerical analysis of flash-boiling spray characteristics with liquid ammonia. Phys Fluids2024; 36(6):063327.
[118]
ZhouX, LiT, WangN, WuZ, CaoJ, ChenR, et al. Similarity of high-pressure direct-injection liquid ammonia spray for different-sized engines. Energy2024; 310:133267.
[119]
PeléR, Mounaïm-RousselleC, BréquignyP, HespelC, BellettreJ. First study on ammonia spray characteristics with a current GDI engine injector. Fuels2021; 2(3):253-71.
[120]
AkramMS, ChengQ, KaarioO, LarmiM. Superheated fuel sprays: a comparative study of flash boiling ammonia fuel sprays with methanol, ethanol, and gasoline for multi-hole fuel injection. Case Stud Therm Eng2024; 61:105110.
[121]
ScharlV, BjørgenKOP, EmbersonDR, LøvåsT. Investigation of fuel temperature and injection timing effects on ammonia direct injection in an optical engine. Appl Energy Combust Sci2024; 20:100299.
[122]
ZembiJ, BattistoniM, PandalA, RousselleC, PelèR, BrequignyP, et al. Numerical study of ammonia spray with a GDI engine injector. J Ammonia Energy2023; 1:59-73.
[123]
HuY, LiJ, ChenH, ZhangF, WangL. Numerical study on spray characteristics of liquid ammonia based on variable model constants. Appl Therm Eng2024; 261:125166.
[124]
RachakondaSK, GoetteD, SchmidtDP. Near-nozzle flash-boiling flow of isooctane, methanol, and ammonia in the engine combustion network spray G injector. Phys Fluids2024; 36(11):113371.
[125]
ZembiJ, BattistoniM, PandalA, PeléR, BrequignyP, HespelC, et al. Lagrangian CFD modeling of ammonia sprays: a correlation across flash boiling and evaporative conditions. Int Commun Heat Mass Transf2024; 158:107866.
[126]
PayriR, García-OliverJM, BrachoG, CaoJ. Experimental characterization of direct injection liquid ammonia sprays under non-reacting diesel-like conditions. Fuel2024; 362:130851.
[127]
ZhangY, XuL, ZhuY, XuS, BaiXS. Numerical study on liquid ammonia direct injection spray characteristics under engine-relevant conditions. Appl Energy2023; 334:120680.
[128]
WuH, DuJ, HouidiMB, AljohaniB, CenkerE, AlRamadanAS, et al. Spray and combustion characterization under an ultra-high-density condition-multifuel comparison. Proc Combust Inst2024; 40(1-4):105555.
[129]
WangN, LiT, ZhouX, LiS, ChenR. Characteristics of high-pressure ammonia spray combustion under diesel-like conditions. Appl Therm Eng2024; 257:124335.
[130]
LiS, WangN, LiT, ChenR, YiP, HuangS, et al. Experimental investigation on liquid length of direct-injection ammonia spray under engine-like conditions. Energy2024; 301:131758.
[131]
YangR, TangQ, ChengH, ZhangS, ZhangY, YaoM. Experimental study on the spray characteristics of high-pressure liquid ammonia under different ambient conditions. J Energy Inst2024; 117:101771.
[132]
WuH, MiS, QianY, ZhangT, ZhangJ, PanC, et al. Spray and evaporation characteristics of high-pressure liquid ammonia injection under flash-boiling and evaporating conditions. Fuel2025; 381:133627.
[133]
WuH, QianY, ZhangT, ZhuJ, LuX. Ignition and flame development of highpressure liquid ammonia spray combustion with simultaneous high-speed OH* and NH2* chemiluminescence imaging. Combust Flame2025; 272:113899.
[134]
MatsuuraY, BannoA, MikamiM. Single ammonia droplet combustion in a high-pressure environment in microgravity. Proc Combust Inst2024; 40(14):105503.
ZhangX, TianJ, CuiZ, YinS, YeM, YangH, et al. Visualization study on the flame propagation and distribution characteristics and exploration of optimal injection strategy in ammonia/diesel dual direct injection mode. Energy2024; 307:132499.
[137]
WuH, QianY, MiS, ZhangT, LuX. Ammonia-PODE dual-fuel direct-injection spray combustion: an optical study of spray interaction, ignition and flame development. J Clean Prod2024; 487:144647.
[138]
PathakU, ScharlV, KrnacD, SattelmayerT. Numerical investigation on temperature dependence of flame behavior in high pressure direct injection combustion of diesel piloted liquid ammonia sprays. In:Proceedings of Internal Combustion Engine Division; 2024 Oct 20-23; San Antonio, TX, USA. New York City: ASME; 2024.
[139]
ZhangJ, ChenD, LiX, LiJ, HuangH, KobayashiN. Large eddy simulation of ammonia-diesel dual fuel spray combustion: effects of ambient condition on ignition characteristics. Chem Eng J2024; 501:157698.
[140]
YuW, WeiY, DongD, ZhouM, MiX, ZhangZ, et al. Characteristics of NO/NO2/ N2O Formation and distribution in ammonia/n-heptane dual-fuel spray combustion under different ammonia concentrations. Energy Fuels2024; 38 (17):16882-95.
[141]
TianJ, ZhangX, CuiZ, YeM, WangY, XuT, et al. Visualization study on ammonia/diesel dual direct injection combustion characteristics and interaction between sprays. Energy Convers Manag2024; 299: 117857.
[142]
ChenZ, HeH, WuJ, WangL, LouH, ZhaoP, et al. An experimental study the cross spray and combustion characteristics diesel and ammonia in a constant volume combustion chamber. Energy2024; 293:130733.
[143]
ZhangZ, LongW, CuiZ, DongP, TianJ, TianH, et al. Visualization study on the ignition and diffusion combustion process of liquid phase ammonia spray ignited by diesel jet in a constant volume vessel. Energy Convers Manag2024; 299:117889.
[144]
SharmaP, BrouzetD, ChungWT, IhmeM. Examining diesel-spray assisted ignition of ammonia under reactivity-controlled conditions using large-eddy simulations. Proc Combust Inst2024; 40(1-4):105317.
[145]
HaputhanthriSO, MaxwellTT, FlemingJ, AustinC. Ammonia and gasoline fuel blends for internal combustion engines. J Energy Res Technol2015; 137 (6):062201.
[146]
RyuK, Zacharakis-JutzGE, KongSC. Performance characteristics of compression-ignition engine using high concentration of ammonia mixed with dimethyl ether. Appl Energy2014; 113:488-99.
[147]
GrossCW, KongSC. Performance characteristics of a compression-ignition engine using direct-injection ammonia-DME mixtures. Fuel2013; 103:1069-79.
[148]
XuL, XuS, BaiXS, RepoJA, HautalaS, HyvönenJ. Performance and emission characteristics of an ammonia/diesel dual-fuel marine engine. Renew Sustain Energy Rev2023; 185:113631.
[149]
LangM, SuY, WangY, ZhangY, WangB, ChenS. Experimental study on the effects of pilot injection strategy on combustion and emission characteristics of ammonia/diesel dual fuel engine under low load. Energy2024; 303:131913.
[150]
ShiT, PeiY, JinS, ZiZ, ZhangF, WuB. An investigation into particulate emission and the formation mechanism of soot precursors in ammoniadiesel dual-fuel engines. Fuel2025; 391:134734.
[151]
ZhongL, ZhaoW, WeiH, ShuG, ZhouL. A novel concept of pre-chamber turbulent jet ignition-induced liquid ammonia spray flame. Phys Fluids2024; 36(12):126107.
[152]
IchikawaY, NikiY, TakasakiK, KobayashiH, MiyanagiA. NH 3 combustion using three-layer stratified fuel injection for a large two-stroke marine engine: experimental verification of the concept. Appl Energy Combust Sci2022; 10:100071.
[153]
IchikawaY, NikiY, TakasakiK, KobayashiH, MiyanagiA. Experimental study of combustion process of NH3 stratified spray using imaging methods for NH3 fueled large two-stroke marine engine. Appl Energy Combust Sci2023; 13:100119.
[154]
BakirAH, GeH, ZhangZ, ZhaoP. Computational investigation on spray autoignition of liquid ammonia with dissolved hydrogen in spray D configuration. Fuel2024; 371:132124.
[155]
ZhouX, LiT, WangN, WangX, ChenR, LiS. Pilot diesel-ignited ammonia dual fuel low-speed marine engines: a comparative analysis of ammonia premixed and high-pressure spray combustion modes with CFD simulation. Renew Sustain Energy Rev2023; 173:113108.
[156]
NadimiE, PrzybyłaG, LøvåsT, PeczkisG, AdamczykW. Experimental and numerical study on direct injection of liquid ammonia and its injection timing in an ammonia-biodiesel dual injection engine. Energy2023; 284:129301.
[157]
NadimiE, PrzybyłaG, LøvåsT, AdamczykW. Effects of biodiesel injector configuration and its injection timing on performance, combustion and emissions characteristics of liquid ammonia dual direct injection engine. J Energy Inst2024; 114:101605.
[158]
ZhangZ, LongW, DongP, TianH, TianJ, LiB, et al. Performance characteristics of a two-stroke low speed engine applying ammonia/diesel dual direct injection strategy. Fuel2023; 332:126086.
[159]
DongP, LiuK, ZhangL, ZhangZ, LongW, TianH. Study on the synergistic control of nitrogenous emissions and greenhouse gas of ammonia/diesel dual direct injection two-stroke engine. Energy2024; 307:132657.
[160]
DongP, ChenS, ZhangL, ZhangZ, LongW, WangQ, et al. Ammonia diffusion combustion and emission formation characteristics in a single cylinder two stroke engine. Energy2024; 311:133432.
[161]
ParkC, JangI, ParkG, MinC, KimM, KimY, et al. Effect of oxygen concentrations in intake air on combustion characteristics of ammonia direct injection SI engine. Fuel2024; 376:132643.
[162]
LeeJ, ParkC, JangI, KimM, ParkG, KimY. Experimental research on the effect of diesel post-injection conditions on the efficiency and global warming potential in a single-cylinder four-stroke marine engine fueled with ammonia and diesel. Energy2025; 314:134244.
[163]
MiS, ZhangJ, ShiZ, WuH, ZhaoW, QianY, et al. Optimization of directinjection ammonia-diesel dual-fuel combustion under low load and higher ammonia energy ratios. Fuel2024; 375:132611.
[164]
YangR, YueZ, ZhangS, LvZ, YaoM. Ammonia thermal atmosphere compression ignition combustion mode to achieve efficient combustion and low greenhouse gas emissions. Energy Convers Manag2025; 325:119427.
[165]
LiZ, FanY, LiJ, WuK, ZhangZ, RenF, et al. Stratified charge assisted jet ignition mode (SCAJI) for low-speed two-stroke Otto cycle ammonia marine engine. Fuel2025; 379:133037.
[166]
MiS, ZhangJ, ShiZ, WuH, QianY, ZhuL, et al. Investigation of low carbon emission and high thermal efficiency of diesel engine combined with highpressure direct injection of hydrogen carrier: ammonia. Int J Hydrogen Energy2024; 88:86-96.
[167]
MiS, ShiZ, ZhangJ, WuH, QianY, LuX. Exploration of ultra-low carbon heavy-duty commercial vehicle technology: liquid ammoniadiesel dual fuel high-pressure direct injection. Int J Engine Res2024; 26(7): 1029-43.
[168]
XiongQ, ZhaoB, WanZ, LiangD, LiuL. Effect of ammonia jet strategy and intake air temperature on mixing characteristics in ammonia/diesel dual-fuel marine engine. Appl Therm Eng2025; 262:125192.
[169]
WangY, ZhouX, LiuL. Feasibility study of hydrogen jet flame ignition of ammonia fuel in marine low speed engine. Int J Hydrogen Energy2023; 48 (1):327-36.
[170]
LinZ, LiuS, LiuW, WangW, CaiK, QiY, et al. Numerical investigation of ammonia-rich combustion produces hydrogen to accelerate ammonia combustion in a direct injection SI engine. Int J Hydrogen Energy2024; 49:338-51.
[171]
LiT, ZhouX, WangN, WangX, ChenR, LiS, et al. A comparison between lowand high-pressure injection dual-fuel modes of diesel-pilot-ignition ammonia combustion engines. J Energy Inst2022; 102:362-73.
[172]
CuiJ, ChenW, WangB, FanY, TianH, LongW, et al. Effects of relative position of injectors on the performance of ammonia/diesel two-stroke engines. Energy2024; 309:133085.
PDF
(13104KB)
