Recent advances in modelling and experimental studies of SCR catalysts in ammonia-fuelled engine applications

1 Introduction

The urgent need to reduce greenhouse gas emissions and transition toward sustainable energy systems has intensified interest in alternative carbon-free fuels. Among these, ammonia (NH₃) has emerged as a promising candidate due to its high hydrogen (H₂) density, established production and distribution infrastructure, and potential for carbon-neutral synthesis (Mørch et al., 2011; Okafor et al., 2019). Unlike conventional hydrocarbon fuels, NH₃ combustion does not produce carbon dioxide (CO₂); however, it introduces significant challenges associated with pollutant formation, particularly nitrogen oxides (NO_x) and unburned NH₃.

NH₃ combustion in internal combustion engines is characterised by low laminar burning velocities (LBV), narrow flammability limits, and high ignition energy requirements, often resulting in incomplete combustion and elevated NH₃ slip (Liu et al., 2023; Mørch et al., 2011). Experimental studies on NH₃-fuelled and NH₃ dual-fuel (DF) engines have demonstrated that NO_x emissions remain a major challenge despite the absence of carbon-containing exhaust species (Lesmana et al., 2021; Reiter and Kong, 2011). These characteristics distinguish NH₃-fuelled systems from conventional diesel and gasoline engines and necessitate dedicated aftertreatment strategies.

Selective catalytic reduction (SCR) technology remains the most effective technology for NO_x abatement under lean-burn conditions. Conventionally, SCR systems rely on NH₃ generated from aqueous urea (NH₂CONH₂) decomposition. In NH₃-fuelled engines, however, residual NH₃ present in the exhaust may partially or completely replace external reductant dosing under certain operating conditions (Lesmana et al., 2019; Cardoso et al., 2021). This creates opportunities for simplified aftertreatment architectures but also introduces challenges associated with controlling NH₃ slip, catalyst selectivity, and N₂O formation.

The composition of NH₃-combustion exhaust differs substantially from that of conventional diesel engines. Variations in NO:NO₂ ratios, high water vapour concentrations, fluctuating NH₃ slip levels, and comparatively low exhaust gas temperatures (EGT) can significantly alter catalyst performance and reaction pathways (Abu-Jrai et al., 2017; Kuta et al., 2023). Consequently, catalyst formulations originally developed for diesel applications may not operate optimally under NH₃-fuelled conditions.

Several candidate aftertreatment architectures have been proposed for NH₃-fuelled engines, differing in catalyst arrangement and the utilisation of engine-out NH₃ slip. Figure 1 summarises representative configurations currently being explored. Depending on the combustion strategy and engine-out exhaust composition, systems may incorporate combinations of diesel oxidation catalysts (DOC), SCR catalysts, ammonia slip catalysts (ASC), and dedicated N₂O decomposition catalysts. Unlike conventional urea-SCR systems, NH₃-fuelled engines may utilise engine-out NH₃ slip as the reductant, reducing or eliminating the need for external urea dosing. The most suitable architecture depends on engine operating conditions, catalyst performance, and emissions-control requirements.

Significant advances have been made in SCR catalyst development over the past 2 decades. Vanadia-based catalysts remain widely used due to their high activity within the 300 °C–450 °C temperature range (Busca et al., 1998). More recently, Cu- and Fe-exchanged zeolites (CuZ and FeZ) have gained prominence because of their superior hydrothermal stability and broader operating windows (Guo Y, 2017; Paolucci et al., 2017). Detailed spectroscopic and kinetic studies have revealed the critical role of redox-active metal centres, Brønsted acidity, and NH₃ storage capacity (NSC) in governing catalyst activity and selectivity.

Advances in experimental characterisation techniques, including operando spectroscopy, transient reactor studies, and isotopic labelling experiments, have substantially improved understanding of SCR reaction pathways. These studies have identified key surface intermediates involved in NH₃ activation and NO_x reduction, supporting mechanistic descriptions based on Langmuir–Hinshelwood (L-H), Eley-Rideal (E-R), and Mars-van Krevelen (MvK) pathways (Paolucci et al., 2017; Ramis et al., 1996).

In parallel, modelling approaches have evolved from empirical kinetic expressions to detailed microkinetic and reactor-scale simulations. Modern SCR models incorporate NH₃ adsorption, surface reactions, diffusion limitations, and transient storage phenomena, enabling prediction of catalyst performance under realistic engine operating conditions (Olsson et al., 2008; Song et al., 2015). Computational fluid dynamics (CFD) models have further enhanced understanding of flow distribution, reductant mixing, and temperature non-uniformities within full-scale aftertreatment systems. Despite these advances, relatively few studies have coupled combustion and aftertreatment models specifically for NH₃-fuelled engines, where exhaust composition differs markedly from conventional applications.

An additional challenge arises from the formation of nitrous oxide (N₂O), a potent greenhouse gas (GHG) with a global warming potential (GWP) approximately 273 times greater than that of CO₂ over a 100-year timescale (IPCC, 2021). Experimental investigations have shown that N₂O formation depends strongly on catalyst composition, operating temperature, and the NO₂:NO_x ratio at the SCR inlet (Jung et al., 2022; Madia et al., 2002). Under NH₃-rich conditions typical of NH₃ combustion, side reactions involving NH₃ oxidation and NH₄NO₃ decomposition can significantly increase N₂O emissions.

While reviews such as Gopinathan and Mohamed Ibrahim, 2025 have examined NH₃ combustion strategies and engine performance, comparatively less attention has been given to the behaviour of SCR catalysts under sustained NH₃-rich exhaust conditions and the associated mechanisms of N₂O formation. This review provides a comprehensive assessment of recent advances in experimental and modelling studies of SCR catalysts for NH₃ combustion exhaust treatment. Particular emphasis is placed on catalyst materials, reaction mechanisms, kinetic modelling approaches, and the coupling between combustion processes and catalytic aftertreatment. By critically examining current developments and identifying remaining research gaps, this review aims to support the development of robust, efficient, and low-emission SCR systems for future NH₃-based engine applications.

2 Characteristics of ammonia combustion exhaust

NH₃ combustion produces an exhaust composition that differs fundamentally from that of conventional hydrocarbon-fuelled engines, creating unique challenges and opportunities for SCR systems. While the absence of carbon eliminates direct CO₂ emissions, combustion of NH₃ introduces substantial concentrations of nitrogen-containing species, including NO_x and unburned NH₃, which strongly influence catalyst performance and aftertreatment design. In addition, exhaust temperature, oxygen (O₂) concentration, water vapour content, and the presence of reactive intermediates can significantly alter SCR reaction pathways and catalyst selectivity (Kobayashi et al., 2019; Valera-Medina et al., 2018). Engine-out temperature and exhaust composition are key design parameters for NH₃-based SCR systems because they directly influence catalyst activity and sizing, and NH₃ utilisation (Kuta et al., 2022).

2.1 NO_x formation and speciation

Unlike hydrocarbon combustion, where NO_x is predominantly formed through thermal-NO mechanisms, NO_x formation during NH₃ combustion is largely governed by fuel-bound nitrogen chemistry. NH₃ oxidation proceeds through a complex network of elementary reactions involving NH₂, NH, N, HNO and N₂H_x intermediates, which ultimately lead to the formation of NO and NO₂ (Glarborg, 2022; Kobayashi et al., 2019). Consequently, NH₃-fuelled engines often exhibit higher engine-out NO_x emissions than conventional diesel engines, particularly under stoichiometric and moderately lean conditions where fuel-nitrogen conversion is favoured (Valera-Medina et al., 2018).

The relative proportions of NO and NO₂ are critical for SCR performance. In conventional diesel exhaust, NO typically dominates, whereas NH₃ combustion can lead to more variable NO:NO₂ ratios depending on combustion conditions and post-combustion oxidation processes (Kobayashi et al., 2019). The presence of NO₂ is beneficial for the “fast SCR” reaction, where an equimolar NO:NO₂ ratio enhances reaction rates, extends catalyst activity to lower temperatures, and improves overall NO_x conversion efficiency (Brandenberger et al., 2008). The NO:NO₂ ratio produced during NH₃ combustion is particularly important for downstream SCR performance. Conditions that generate a balanced NO:NO₂ ratio enhances fast-SCR activity and improves low-temperature NO_x conversion, whereas NO-dominated exhaust generally relies on the slower standard-SCR pathway (Koebel et al., 2000; Song et al., 2015). However, excessive NO₂ concentrations may promote undesirable side reactions, including the formation of ammonium nitrate (NH₄NO₃) and subsequent N₂O emissions, particularly under low-temperature operating conditions (Madia et al., 2002). Reactor-scale studies have demonstrated that SCR performance is highly sensitive to inlet NO₂:NO_x ratios and temperature-dependent transitions between standard SCR, fast SCR and NH₃ oxidation pathways, highlighting the importance of accurately representing coupled transport and reaction kinetics in predictive models (Wang et al., 2020). SCR reaction pathways are summarised in Table 1.

TABLE 1. SCR reaction pathways.

2.2 Ammonia slip and its dual role

A defining characteristic of NH₃-fuelled engines is the presence of significant unburned NH₃ in the exhaust stream. NH₃ possesses a low LBV, high autoignition temperature, and relatively slow oxidation kinetics, all of which contribute to incomplete combustion and NH₃ slip, especially under low-load and lean-burn operation (Lesmana et al., 2021; Valera-Medina et al., 2018).

From an emissions perspective, NH₃ slip is undesirable due to the toxicity of NH₃ and its contribution to secondary particulate formation in the atmosphere. However, from an aftertreatment perspective, the presence of NH₃ represents a potentially valuable intrinsic reductant for SCR. Unlike conventional diesel SCR systems that rely on urea decomposition to generate NH₃, NH₃-fuelled engines may provide sufficient NH₃ directly from combustion, reducing or potentially eliminating the need for external reductant dosing (Kuta et al., 2023).

This concept, often referred to as passive SCR, has attracted considerable attention because it simplifies aftertreatment architecture and reduces system complexity. Nevertheless, experimental investigations have demonstrated that the engine-out NH₃ concentration is highly dependent on operating conditions and is frequently insufficient to achieve complete NO_x conversion across the entire engine map (Lesmana et al., 2021). Consequently, strategies such as fuel bypass, supplementary NH₃ injection, or adaptive catalyst control may still be required. The exhaust NH₃:NO_x ratio is of particular importance for downstream SCR operation. Experimental studies have reported optimum SCR performance at NH₃:NO_x ratios of approximately 1–1.4, indicating that combustion conditions producing balanced NO_x and NH₃ slip may reduce the external urea demand of the aftertreatment system (Zhao et al., 2023).

Excessive NH₃ slip presents an additional challenge because downstream ammonia slip catalysts (ASC) may oxidise NH₃ to NO, NO₂, or N₂O instead of the desired N₂, thereby creating secondary emissions (Ramis et al., 1996; Schaber et al., 2004). ASC should not be viewed solely as NH₃ cleanup devices; under NH₃-rich exhaust conditions they can influence NOx conversion and N₂O formation simultaneously, making them an integral component of future NH₃-engine aftertreatment architectures (Wu et al., 2024). Balancing NO_x reduction efficiency against NH₃ slip therefore represents a key optimisation challenge for NH₃-fuelled SCR systems.

2.3 Exhaust gas temperature characteristics

Exhaust gas temperature (EGT) is one of the most influential parameters governing SCR performance. Compared with conventional diesel combustion, NH₃ combustion often produces lower adiabatic flame temperatures, particularly under lean-burn conditions and in NH₃-H₂ or NH₃-diesel DF operation (Valera-Medina et al., 2018).

As a result, exhaust temperatures frequently fall within the low-temperature region where SCR activity is kinetically limited. Temperatures below approximately 200 °C–250 °C reduce NH₃ activation and slow both adsorption and surface reaction rates, leading to poor NO_x conversion efficiency (Brandenberger et al., 2008; Damma et al., 2019). This challenge is particularly relevant during cold-start operation, low-load engine conditions, and auxiliary power applications.

Conversely, high-temperature operation introduces a different limitation. Above approximately 450 °C–500 °C, NH₃ oxidation reactions become increasingly competitive with SCR reactions, resulting in decreased NO_x conversion efficiency and increased formation of NO and N₂O (Nicosia et al., 2008). Maintaining catalyst operation within an optimal temperature window is therefore essential for achieving high N₂ selectivity.

The combination of low-temperature activity requirements and high-temperature stability demands has driven significant interest in Cu- and Fe-exchanged zeolites (CuZ and FeZ), which exhibit broader operational windows than conventional vanadia-based catalysts (Damma et al., 2019).

2.4 Presence of additional reactive species

NH₃ combustion exhaust contains several additional species that can influence SCR chemistry. Residual oxygen is typically present because NH₃ engines often operate under lean conditions to suppress NO_x formation and improve combustion stability (Valera-Medina et al., 2018).

O₂ plays a dual role in SCR systems. Moderate O₂ concentrations promote oxidation of NO to NO₂, thereby facilitating fast-SCR pathways and improving low-temperature activity. However, excessive O₂ availability can also accelerate NH₃ oxidation, reducing NH₃ utilisation efficiency and increasing N₂O formation (Kuta et al., 2023).

H₂ may also be present, particularly in NH₃-H₂ DF engines or in systems where NH₃ cracking is employed to enhance combustion. Residual H₂ can modify catalyst surface chemistry and, under suitable conditions, promote H₂-assisted SCR pathways that improve low-temperature NO_x conversion (Liu et al., 2011).

Water vapour represents another important exhaust constituent. Although generally considered a weak inhibitor, water can compete with NH₃ and NO_x species for adsorption sites, particularly at lower temperatures, thereby reducing catalyst activity (Madia et al., 2002). Long-term exposure to high water concentrations may also accelerate hydrothermal ageing of zeolite catalysts, resulting in framework degradation and loss of active metal sites (Paolucci et al., 2017).

In NH₃ DF engines, trace hydrocarbons originating from pilot fuels may also be present. These species can compete for active sites and alter catalyst selectivity, particularly under low-temperature conditions where hydrocarbon adsorption is favoured (Olsson et al., 2008).

2.5 Implications for SCR system design

The unique characteristics of NH₃ combustion exhaust necessitate a fundamentally different approach to SCR system design compared with conventional diesel applications. The simultaneous presence of NO_x and NH₃ creates opportunities for passive SCR operation, but also introduces challenges associated with NH₃ slip control, N₂O formation, and catalyst selectivity.

Catalysts must therefore exhibit:

• High activity at low exhaust temperatures;
• Strong resistance to hydrothermal ageing;
• High N₂ selectivity under NH₃-rich conditions;
• Low NH₃ oxidation activity;
• Minimal N₂O formation across a wide operating window.

These requirements increasingly favour advanced zeolite catalysts and hybrid catalyst architectures capable of maintaining performance under highly variable exhaust conditions (Damma et al., 2019; Paolucci et al., 2017). Furthermore, because engine-out NH₃ and NO_x concentrations are strongly influenced by combustion strategy, SCR optimisation cannot be considered independently of engine calibration. Integrated combustion-aftertreatment approaches are therefore emerging as a key research direction for NH₃-fuelled powertrains. The key characteristics of ammonia combustion exhaust and their implications for SCR performance are summarised in Table 2.

TABLE 2. Key characteristics of NH₃ combustion exhaust and their implications for SCR performance.

3 SCR catalyst technologies for ammonia combustion exhaust

SCR catalysts are central to achieving efficient NO_x abatement in NH₃-fuelled systems. However, the distinct characteristics of NH₃ combustion exhaust, particularly NH₃ fuel-rich conditions, variable NO:NO₂ ratios, and lower exhaust gas temperatures (EGT), place additional demands on catalyst performance. Consequently, catalyst selection must consider not only activity and selectivity, but also physicochemical properties such as redox behaviour, NH₃ storage capacity (NSC), and resistance to deactivation.

3.1 Catalyst families

A comparison of major SCR catalyst families for ammonia combustion exhaust is provided in Table 3.

TABLE 3. Comparison of major SCR catalyst families for ammonia combustion exhaust (Brandenberger et al., 2008; Damma et al., 2019; Gao et al., 2013).

3.1.1 Vanadia-based catalysts

Vanadia-based catalysts, typically formulated as V₂O₅-WO₃/TiO₂, remain among the most widely deployed commercial SCR catalysts owing to their high NO_x conversion efficiency and N₂ selectivity within an intermediate temperature window of approximately 300 °C–450 °C (Busca et al., 1998; Nova et al., 2009). Their activity arises from the coexistence of Brønsted and Lewis acid sites that facilitate NH₃ adsorption and activation, together with redox-active vanadium species that promote NO_x reduction via surface O₂ transfer reactions.

However, their applicability to NH₃ combustion exhaust is limited by poor low-temperature activity and increased NH₃ oxidation at elevated temperatures. Under NH₃-rich conditions, these side reactions can decrease N₂ selectivity and promote formation of NO and N₂O (Jung et al., 2022). Long-term operation at high temperatures may also lead to vanadium volatilisation and catalyst deactivation through sintering and phase transformations of both the active phase and support (Guan et al., 2014; Nova et al., 2009).

3.1.2 Zeolite-based catalysts

Metal-exchanged zeolites have become the dominant SCR technology for modern automotive and stationary applications owing to their wide operating temperature windows and superior hydrothermal stability (Fickel and Lobo, 2010; Gao et al., 2013). Among these materials, Cu-CHA and Fe-ZSM-5 catalysts have received particular attention because of their excellent activity under fuel-lean conditions. SCR activity in CuZ and FeZ is associated with isolated exchanged metal ions and well-dispersed active centres, rather than metal oxide clusters.

Cu-exchanged zeolites exhibit outstanding low-temperature SCR activity and high NH₃ storage capacity (NSC) due to the presence of isolated Cu²⁺ ions and mobile Cu species that participate in rapid Cu²⁺/Cu⁺ redox cycling (Paolucci et al., 2017). Consequently, CuZ are particularly attractive for NH₃-fuelled engines, where lower exhaust temperatures are often encountered. FeZ generally exhibit lower low-temperature activity but maintain high NO_x conversion at temperatures above 450 °C and display superior thermal stability under severe operating conditions (Brandenberger et al., 2008; Gao et al., 2013; Shin et al., 2020 reported NOx conversion above approximately 80% between about 300 °C and 450 °C for both catalyst systems in non-road engines.

Recent spectroscopic and computational studies have demonstrated that catalyst performance depends strongly on the coordination environment and mobility of exchanged metal ions within the zeolite framework, which directly influence SCR reaction pathways and catalyst durability (Borfecchia et al., 2015; Paolucci et al., 2017).

3.1.3 Transition metal oxide catalysts

Vanadium-free transition metal oxides have attracted increasing attention as environmentally benign alternatives to conventional SCR catalysts. Among these materials, manganese oxides exhibit exceptional low-temperature activity due to the coexistence of multiple oxidation states (Mn²⁺, Mn³⁺ and Mn⁴⁺), which facilitate rapid redox cycling and oxygen transfer reactions (Chen et al., 2012; Zhang et al., 2023). Several studies have reported NO_x conversions exceeding 90% below 250 °C. However, their strong oxidation capability often promotes NH₃ overoxidation, resulting in poor N₂ selectivity and increased N₂O formation.

Ceria-based catalysts have emerged as promising candidates because of their high O₂ storage capacity and reversible Ce⁴⁺/Ce³⁺ redox cycle. These properties enhance O₂ mobility, improve resistance to water inhibition, and promote catalytic stability under transient operating conditions (Zhou et al., 2021). Iron oxide catalysts generally exhibit moderate activity over a broader temperature range but remain limited by lower low-temperature performance compared with Mn- and Cu-based systems.

3.1.4 Platinum group metal (PGM) catalysts

Platinum group metals (Pt, Pd and Rh) exhibit exceptionally high intrinsic catalytic activity and are widely employed in NH₃ oxidation catalysts and ammonia slip catalysts (ASCs). Their ability to activate NH₃ at low temperatures makes them highly effective for downstream NH₃ removal.

Nevertheless, the strong oxidation activity of PGMs often leads to poor N₂ selectivity at elevated temperatures, with significant formation of NO and N₂O (Lott and Deutschmann, 2023). Combined with their high cost and susceptibility to poisoning, these limitations have restricted their use as primary SCR catalysts. Instead, PGMs are increasingly incorporated into hybrid catalyst architectures that combine efficient NO_x reduction with downstream NH₃ oxidation functionality.

3.2 Key physicochemical characteristics

The performance of SCR catalysts is governed not only by their chemical composition but also by a range of physicochemical properties that influence adsorption, reaction kinetics, mass transport, and long-term stability. These characteristics are particularly important in NH₃-fuelled applications, where exhaust composition and operating conditions differ substantially from conventional diesel systems.

SCR activity is strongly influenced by the synergy between redox-active centres responsible for NO oxidation and acidic sites responsible for NH₃ adsorption and activation, highlighting the importance of balancing redox and acidity characteristics during catalyst design (Liu et al., 2014). High SCR activity requires an appropriate balance between acidity and redox capability rather than maximizing either property independently (Shi et al., 2023). One of the most important parameters is the nature and distribution of active sites. SCR reactions require the adsorption and activation of NH₃ on Brønsted and Lewis acid sites, followed by reaction with adsorbed NO_x species. The relative abundance and strength of these sites influence NH₃ storage, reaction rates, and selectivity. For example, Cu-exchanged CHA zeolites contain isolated Cu ions capable of participating in rapid redox cycles while simultaneously providing strong NH₃ adsorption capacity, contributing to their excellent low-temperature activity (Paolucci et al., 2017). High dispersion of active metal species has been identified as a critical factor governing NH₃-SCR activity, as dispersed sites provide greater accessibility and promote efficient redox cycling compared with bulk oxide phases. The performance of NH₃-SCR catalysts is governed not only by intrinsic redox activity but also by the synergistic interaction between redox centres and Brønsted/Lewis acid sites, with highly dispersed active species providing the most effective balance of NH₃ adsorption, NO oxidation, and catalytic turnover. Brønsted and Lewis acid sites are not merely NH₃ storage sites but actively participate in different NH₃ activation pathways, leading to different SCR reaction routes and kinetics (Nova et al., 2009). The distribution and strength of Brønsted and Lewis acid sites strongly influence NH₃ adsorption, storage, activation and overall SCR activity, making acidity a key design parameter for next-generation catalysts (Shi et al., 2023).

Redox properties are equally important because NO oxidation and catalyst regeneration depend on the ability of active species to undergo reversible oxidation–reduction cycles. Catalysts with high oxygen mobility and readily accessible lattice oxygen generally exhibit enhanced SCR activity, particularly under transient conditions. Ceria-based materials are notable in this regard due to the reversible Ce⁴⁺/Ce³⁺ redox couple, which promotes oxygen storage and oxygen mobility and NO oxidation capability, promoting the fast SCR pathway and extending catalytic activity over a broader temperature range, while improving resistance to fluctuations in exhaust composition (Zhang et al., 2023).

NH₃ storage capacity (NSC) strongly influences catalyst performance during transient operation. High NH₃ storage enables the catalyst to buffer short-term variations in NH₃ availability and maintain NO_x conversion during rapid changes in engine load. However, excessive NH₃ storage may increase the risk of NH₃ slip if stored NH₃ is subsequently released under changing operating conditions (Song et al., 2015).

The pore structure and morphology of the catalyst also play a critical role. Zeolitic catalysts possess highly ordered microporous structures that provide large surface areas and abundant adsorption sites, but diffusion limitations may arise at high gas hourly space velocities (GHSV). In contrast, mesoporous materials generally offer improved mass transport but lower adsorption capacity. Consequently, an appropriate balance between adsorption strength and diffusion accessibility is required to maximise catalyst utilisation under realistic exhaust conditions (Busca et al., 1998).

Long-term catalyst durability depends largely on resistance to hydrothermal ageing, sulphur poisoning, and thermal degradation. Exposure to high temperatures and water vapour can induce framework collapse, metal agglomeration, or phase transformations that reduce active-site availability. Sulphur-containing species may additionally block active sites through sulphate formation, particularly in catalysts exposed to pilot-fuel combustion products. Among current SCR catalyst families, Cu-CHA materials generally exhibit superior hydrothermal stability, whereas vanadia-based catalysts are more susceptible to thermal degradation at elevated temperatures (Brandenberger et al., 2008; Damma et al., 2019).

Collectively, these physicochemical properties determine the activity, selectivity, and durability of SCR catalysts and therefore represent critical design parameters for NH₃-fuelled engine applications. The key physiochemical properties governing SCR performance are summarised in Table 4.

TABLE 4. Key physicochemical properties governing SCR performance.

3.3 Catalyst requirements for ammonia combustion exhaust

The exhaust composition produced by NH₃-fuelled engines imposes a set of catalyst requirements that differ significantly from those encountered in conventional diesel applications. In addition to achieving high NO_x conversion, catalysts must tolerate elevated concentrations of unburned NH₃, large variations in EGT, high water vapour levels, and changing NO:NO₂ ratios while maintaining high selectivity towards N₂.

A primary requirement is strong low-temperature activity. NH₃ combustion generally produces lower EGT than diesel combustion, particularly under lean and low-load operation (Valera-Medina et al., 2018). Consequently, SCR catalysts must remain active below 250 °C, where conventional vanadia-based systems often exhibit limited performance. This requirement has driven substantial interest in Cu-exchanged zeolites, which can achieve high NO_x conversion efficiencies under low-temperature conditions through enhanced NH₃ storage and fast-SCR activity (Paolucci et al., 2017; Damma et al., 2019).

Catalysts must also effectively accommodate substantial fluctuations in NH₃ concentration. Unlike conventional SCR systems that rely on controlled urea dosing, NH₃-fuelled engines inherently produce varying levels of NH₃ slip depending on combustion strategy, load, and fuel substitution ratio. This creates opportunities to utilise residual NH₃ as the SCR reductant, reducing or potentially eliminating external urea injection. However, successful implementation requires catalysts capable of maintaining high conversion efficiency over a broad range of NH₃:NO_x ratios while minimising secondary NH₃ emissions (Kuta et al., 2023; Lesmana et al., 2021).

Selectivity towards N₂ is another critical consideration. Under NH₃-rich conditions, competing reactions such as NH₃ oxidation and NH₄NO₃ decomposition may increase emissions of N₂O and NO, reducing the environmental benefits of NH₃ as a carbon-free fuel. Consequently, catalyst formulations must be designed to maximise N₂ formation while suppressing undesired by-products. This challenge is particularly important during low-temperature operation, where N₂O formation pathways become increasingly significant (Jung et al., 2022).

The high-water content of NH₃ combustion exhaust additionally requires catalysts with strong hydrothermal stability and resistance to water inhibition. Water vapour can compete with reactants for adsorption sites and accelerate catalyst ageing through framework degradation or metal migration. Similarly, the use of pilot fuels may introduce sulphur-containing species that promote sulphate formation and catalyst deactivation. Recent research has therefore focused on hydrothermally stable zeolite frameworks, hydrophobic surface modifications, and sulphur-tolerant transition metal oxides capable of maintaining activity under prolonged exposure to realistic exhaust environments (Brandenberger et al., 2008; Zhang et al., 2023). Recent studies show that hydrothermal aging affects Brønsted and Lewis acid sites differently, leading to progressive changes in NH₃ adsorption behaviour and SCR activity as catalyst exposure time and temperature increase (Yao et al., 2023). Catalyst deactivation was found to correlate strongly with reductions in NH₃ storage capacity, highlighting the importance of maintaining accessible adsorption sites for sustained NO_x conversion.

Given these competing requirements, increasing attention has been directed towards multifunctional catalyst architectures. Combined SCR-ASC systems, layered catalyst designs, and hybrid Cu-Fe zeolite formulations have demonstrated potential to simultaneously achieve high NO_x conversion, low NH₃ slip, and reduced N₂O emissions over a broad operating window (Guan et al., 2014). Such integrated approaches are likely to play an important role in future NH₃-fuelled engine aftertreatment systems. Recent research has also increasingly focused on catalyst modification through promoter addition, hierarchical pore structures and tailored acidity to simultaneously improve low-temperature activity, hydrothermal stability and N₂ selectivity (Shi et al., 2023).

Overall, no single catalyst family currently satisfies all performance requirements associated with NH₃ combustion exhaust. Future catalyst development will therefore require a combination of improved material design, enhanced mechanistic understanding, and integrated optimisation of both combustion and aftertreatment systems.

4 Reaction mechanisms and kinetics

A fundamental understanding of reaction mechanisms and kinetics is essential for interpreting NH₃-SCR performance and developing reliable predictive models, particularly under the atypical conditions associated with NH₃-fuelled combustion. The SCR process comprises a complex network of adsorption, surface reactions, redox cycles, and competing side reactions that collectively determine catalyst activity, selectivity, and durability.

The activation of NH₃ is generally considered the initial and rate-determining step in both SCR and competing oxidation pathways. Following adsorption, NH₃ coordinates to Lewis acid sites or protonates on Brønsted acid sites depending on catalyst composition. Adsorbed NH₃ subsequently undergoes hydrogen abstraction to form surface amide (NH₂) species, which are widely recognised as key reactive intermediates. These NH₂ species react with NO to form unstable intermediates such as NH₂NO, which rapidly decompose to yield N₂ and H₂O. This sequence constitutes the basis of the standard SCR pathway and predominates when NO is the principal NO_x species present.

When both NO and NO₂ are available, the fast-SCR pathway becomes operative and substantially increases reaction rates, particularly at low temperatures. Fast SCR proceeds through interactions between adsorbed NH₃ species and oxidised NO_x intermediates, including nitrates and nitrites, resulting in more efficient conversion to N₂. Under NO₂-rich conditions, however, alternative pathways emerge. In particular, NH₄NO₃ can form and accumulate on catalyst surfaces at low temperatures. Subsequent decomposition of these deposits may generate N₂O, linking NO₂ chemistry directly to secondary GHG emissions. Such behaviour is especially relevant for NH₃-fuelled systems, where combustion conditions and upstream oxidation catalysts can significantly influence the NO:NO₂ ratio entering the SCR catalyst.

In parallel with NO_x reduction, several competing reactions consume NH₃ and influence overall selectivity. NH₃ oxidation becomes increasingly important at elevated temperatures or over highly oxidative catalyst formulations. Depending on catalyst composition and operating conditions, oxidation pathways may produce N₂, NO, or N₂O. The formation of N₂O is of particular concern because it can arise both from partial NH₃ oxidation and from the decomposition of NH₄NO₃ intermediates. In addition, imide (NH) species have been implicated in N₂O formation through reactions with NO, further increasing the complexity of the reaction network. Consequently, the dominant SCR pathway is strongly temperature dependent, requiring separate consideration of low-temperature SCR, standard SCR, fast SCR, and high-temperature NH₃ oxidation regimes in kinetic models (Wang et al., 2020).

4.1 Surface reaction mechanisms

At the mechanistic level, NH₃-SCR reactions are commonly described using three complementary frameworks: Langmuir–Hinshelwood (L–H), Eley–Rideal (E–R), and Mars–van Krevelen (MvK) mechanisms. In the L–H mechanism, both NH₃ and NO_x species adsorb on the catalyst surface prior to reaction, making this pathway particularly relevant for catalysts capable of NO oxidation and nitrate formation. In contrast, the E–R mechanism involves the direct reaction of gaseous NO with adsorbed NH₃-derived species such as NH₂, and is often invoked when NH₃ surface coverage is high or NO adsorption is relatively weak. For redox-active catalysts including vanadia systems and transition-metal-exchanged zeolites, the MvK mechanism provides an additional description in which lattice oxygen directly participates in oxidation reactions. During this process, metal centres such as V⁵⁺ or Cu²⁺ undergo cyclic reduction and reoxidation, with gaseous O₂ restoring the oxidised catalyst state.

Recent advances in operando spectroscopy, density functional theory (DFT), and microkinetic modelling have significantly refined understanding of these mechanisms. Studies on Cu-CHA catalysts indicate that catalytic activity is not governed solely by isolated Cu ions but also by dynamic multinuclear Cu species that form under reaction conditions. The mobility of these Cu centres facilitates oxygen activation and Cu²⁺/Cu⁺ redox cycling, contributing to the exceptional low-temperature activity of Cu-zeolite catalysts (Paolucci et al., 2017). Furthermore, recent computational investigations suggest that E-R pathways involving gaseous NO reacting directly with adsorbed NH₂ species may contribute more substantially to overall SCR activity than previously recognised, particularly under NH₃-rich conditions typical of NH₃-fuelled exhaust streams (Voglar et al., 2023).

The coexistence of L-H, E-R, and MvK pathways presents a major challenge for kinetic model development. The dominant reaction route depends strongly on catalyst composition, temperature, NH₃ coverage, O₂ availability, and exhaust-gas composition. Consequently, predictive models must account for multiple parallel and competing pathways while remaining computationally tractable. From an emissions perspective, SCR performance is ultimately governed by the balance between desired NO_x reduction reactions and competing NH₃ oxidation pathways. Conditions that favour rapid NH₃ activation and selective NO_x reduction promote high N₂ selectivity, whereas excessive oxidation activity or accumulation of reactive intermediates increases the likelihood of undesirable by-products such as N₂O. Understanding these interactions remains essential for the design of catalysts and kinetic models tailored to NH₃-fuelled combustion systems. The principle ammonia-SCR reaction mechanisms are summarised in Table 5.

TABLE 5. Principal NH₃-SCR reaction mechanisms.

5 Modelling approaches for NH₃-SCR under ammonia combustion exhaust conditions

The development of predictive models for SCR systems has become increasingly important as NH₃-fuelled engines move from laboratory demonstrations toward practical deployment. Compared with conventional diesel exhaust, NH₃ combustion produces substantially different exhaust compositions, characterised by elevated NH₃ slip, variable NO₂ ratios, high water-vapour concentrations, and often lower exhaust gas temperatures (EGT). These factors introduce additional complexity into SCR modelling and require approaches capable of capturing both catalyst-scale reaction kinetics and system-level interactions between combustion and aftertreatment (Olsson et al., 2008).

A fundamental distinction between conventional urea-SCR systems and NH₃-fuelled engines lies in the origin of the reductant. In diesel applications, NH₃ generation from urea decomposition can be actively controlled to maintain the desired NH₃ ratio. In contrast, NH₃-fuelled engines generate NH₃ slip directly from the combustion process, causing reductant availability to vary with engine operating conditions. Consequently, future SCR models must explicitly account for the coupling between engine-out NH₃ production, catalyst utilisation, and downstream NO_x reduction performance.

5.1 Global kinetic models

Early SCR models were primarily based on global kinetic expressions derived from experimental reactor data. A major advance was the incorporation of dynamic NH₃ storage and adsorption kinetics, allowing transient catalyst behaviour during changing engine conditions to be represented more accurately (Nova et al., 2011). These models describe NO_x conversion using a limited number of lumped reactions representing standard SCR, fast SCR, and NH₃ oxidation pathways. Their relatively low computational cost makes them attractive for engine control applications and one-dimensional aftertreatment simulations (Koebel et al., 2000). However, because most were developed for conventional urea-SCR systems, they often struggle to accurately predict catalyst behaviour under NH₃-rich exhaust conditions, where surface coverage effects, NH₃ storage dynamics, and competing oxidation reactions become increasingly important.

5.2 Microkinetic and reactor-scale models

To improve mechanistic fidelity, researchers have developed microkinetic models that explicitly represent adsorption, desorption, surface reactions, and catalyst redox cycles. These models have proven particularly valuable for Cu- and Fe-exchanged zeolites, where SCR activity depends strongly on the oxidation state and mobility of active metal centres. Combined experimental and theoretical studies have demonstrated that Cu²⁺/Cu⁺ redox cycling plays a central role in low-temperature SCR activity over Cu-CHA catalysts (Paolucci et al., 2017). Similarly, DFT and microkinetic investigations have provided detailed insights into the formation and consumption of nitrates, nitrites, and NH₂ intermediates, improving descriptions of both standard and fast SCR pathways. DFT studies further indicate that NH₃ activation proceeds differently on Brønsted and Lewis acid sites, highlighting the importance of catalyst acidity in determining SCR reaction pathways (Nova et al., 2009). DFT-derived microkinetic models provide valuable insight into the rate-limiting elementary reactions governing fast SCR, improving mechanistic understanding beyond conventional global kinetic models (Brüggemann et al., 2011).

Recent developments in chemical kinetic modelling have demonstrated that reduced reaction mechanisms can successfully reproduce SCR behaviour when key pathways, including standard SCR, fast SCR, and NH₃ oxidation, are consistently parameterised (Burluka and Manning, 2025). However, these studies also highlight the strong sensitivity of SCR performance to NO oxidation, NH₃ storage, and the temperature-dependent competition between NO_x reduction and NH₃ oxidation. Under NH₃-fuelled engine conditions, where NH₃ slip and exhaust composition differ substantially from diesel exhaust, these sensitivities are expected to become even more pronounced, emphasising the need for revalidated kinetic frameworks.

5.3 Multi-scale and CFD modelling

At the reactor scale, one-dimensional and computational fluid dynamics (CFD) models are widely employed to describe the interaction between chemical kinetics and transport phenomena. These approaches couple reaction kinetics with heat transfer, species diffusion, and flow distribution within monolithic catalyst substrates. Reactor-scale simulations have shown that SCR reactions are often confined to a relatively thin near-surface region of the catalyst washcoat, whereas competing oxidation reactions can occur over a larger fraction of the catalyst wall. Such findings highlight the importance of accurately resolving washcoat diffusion, mass transfer limitations, and local temperature gradients. CFD studies have also demonstrated that catalyst geometry can significantly influence performance, with thin-walled monoliths and dual-layer catalyst architectures improving NH₃ utilisation and NO_x conversion while reducing undesired side reactions (Ogidiama, 2014).

Despite these advances, most reactor-scale studies have focused on conventional diesel aftertreatment systems. Similar transport-related challenges are expected under NH₃-fuelled engine conditions, where local variations in NH₃ slip may create regions of over- or under-stoichiometric catalyst operation, affecting both NO_x conversion and N₂O formation. Extending reactor-scale modelling frameworks to NH₃-rich exhaust environments therefore represents an important research priority.

More recently, multi-scale modelling frameworks have emerged that integrate combustion and aftertreatment processes within a single simulation environment. These approaches couple engine combustion models, which predict NO_x and NH₃ formation, with downstream catalyst models describing SCR performance (Song et al., 2015). Such integration is particularly important for NH₃-fuelled engines because exhaust composition depends strongly on combustion strategy, fuel substitution ratio, equivalence ratio, and engine load. Consequently, SCR performance cannot be accurately predicted without simultaneously considering combustion-generated NH₃ slip and NO_x formation.

5.4 Emerging data-driven approaches

Data-driven approaches are also gaining attention as complementary tools for catalyst modelling and control. Machine-learning techniques, including artificial neural networks, have demonstrated potential for predicting catalyst conversion efficiency, NH₃ slip, and emissions under varying operating conditions while significantly reducing computational cost (Liu et al., 2023). However, these methods remain dependent on large, high-quality experimental datasets and generally provide limited mechanistic insight. Hybrid frameworks that combine physically based kinetic models with data-driven optimisation may therefore offer the greatest potential for future NH₃-engine applications.

Although substantial progress has been achieved in global kinetics, microkinetic modelling, CFD simulations, and multi-scale frameworks, important limitations remain. Fully integrated modelling approaches capable of linking atomistic reaction chemistry, catalyst-scale processes, reactor transport phenomena, and engine operation remain relatively scarce (Lott and Deutschmann, 2023; Voglar et al., 2023). Furthermore, microkinetic models require large numbers of kinetic parameters and elementary reaction steps, many of which cannot be measured directly and must instead be estimated through parameter fitting. This introduces uncertainty when models are extrapolated beyond their calibration range.

Another significant limitation is the inadequate treatment of water adsorption and inhibition effects. Water vapour is one of the most abundant species in NH₃ and H₂ combustion exhaust, yet many SCR models continue to represent its influence using simplified competitive adsorption terms (Voglar et al., 2023). Recent hydrothermal ageing models have proposed active-site density as a physically meaningful descriptor of catalyst degradation, linking activity loss directly to changes in Brønsted and Lewis acid site populations during ageing (Yao et al., 2023). Dual-site kinetic models accounting separately for Brønsted and Lewis acid sites have demonstrated improved capability for predicting catalyst behaviour following hydrothermal exposure. Theoretical studies further suggest that water can influence proton-transfer processes and NH₃ adsorption equilibria beyond simple site competition, indicating that current representations may underestimate its impact on SCR kinetics (Nova et al., 2009).

Accurate prediction of N₂O formation remains an additional challenge. N₂O production involves multiple competing pathways, including NH₃ oxidation, NH₄NO₃ formation and decomposition, and catalyst-specific surface reactions, making reliable prediction difficult across a wide operating window (Jung et al., 2022). Transient operating conditions, catalyst ageing, and interactions between combustion and aftertreatment systems introduce further uncertainty.

Overall, SCR modelling capabilities have advanced considerably over the past 2 decades, yet most existing frameworks remain rooted in diesel-derived exhaust conditions. Future modelling efforts should focus on experimentally validated, multi-scale approaches capable of capturing NH₃-rich exhaust chemistry, catalyst ageing, water inhibition effects, and N₂O formation pathways under realistic NH₃-combustion conditions. Such models will be essential for optimising catalyst design, reducing NH₃ slip, mitigating N₂O emissions, and enabling efficient operation of next-generation NH₃-fuelled power and propulsion systems. A comparison of SCR modelling approaches is provided in Table 6.

TABLE 6. Comparison of SCR modelling approaches.

6 Experimental advances in NH₃-SCR for ammonia combustion exhaust

Experimental investigations have played a central role in advancing SCR technology for NH₃-fuelled engines by providing the validation required for catalyst development, mechanistic understanding, and kinetic model formulation. Compared with conventional diesel exhaust, NH₃ combustion generates exhaust streams characterised by elevated NH₃ slip, variable NO:NO₂ ratios, and lower exhaust gas temperatures (EGT), all of which influence catalyst activity, selectivity, and durability (Lesmana et al., 2021; Valera-Medina et al., 2018). As a result, significant research effort has focused on evaluating the suitability of existing SCR catalyst technologies under NH₃-rich conditions.

Early studies primarily examined conventional catalyst families, including vanadia-based systems and transition-metal-exchanged zeolites. Commercial V₂O₅-WO₃/TiO₂ catalysts demonstrated high NO_x conversion efficiencies and excellent N₂ selectivity within their optimal operating window of approximately 300 °C–450 °C, although activity declined substantially at lower temperatures and NH₃ oxidation became increasingly significant at elevated temperatures (Koebel et al., 2000; Nova et al., 2009). Subsequent investigations of Cu- and Fe-exchanged zeolites revealed markedly improved low-temperature performance, with Cu-CHA catalysts achieving near-complete NO_x conversion below 250 °C under fast-SCR conditions (Gao et al., 2013). However, this enhanced activity is often accompanied by increased NH₃ oxidation and N₂O formation under NH₃-rich conditions when compared with vanadia-based catalysts (Jung et al., 2022).

Consequently, experimental catalyst development has increasingly focused on improving low-temperature activity while maintaining high N₂ selectivity and minimising N₂O formation. For Cu-CHA catalysts, spectroscopic and kinetic studies have shown that catalytic performance depends strongly on the distribution, mobility, and redox behaviour of Cu ions within the zeolite framework (Paolucci et al., 2017). Optimisation of copper loading and framework composition has been shown to improve NH₃ storage capacity and low-temperature activity while suppressing undesirable oxidation pathways. Similarly, Fe-exchanged zeolites (FeZ) have attracted considerable attention for high-temperature applications owing to their superior thermal stability and resistance to hydrothermal ageing compared with Cu-based catalysts (Brandenberger et al., 2008).

6.1 Influence of operating conditions

Beyond catalyst formulation, reactor-scale studies have provided valuable insight into the influence of operating conditions on SCR performance. Temperature, gas hourly space velocity (GHSV), O₂ concentration, and feed composition have all been shown to strongly affect NO_x conversion efficiency and selectivity. Among these variables, the NO₂:NO_x ratio is particularly important because approximately equimolar concentrations of NO and NO₂ promote the fast-SCR pathway and substantially enhance low-temperature activity (Song et al., 2015). This finding is especially relevant for ammonia-fuelled engines, where combustion conditions and upstream oxidation processes can significantly alter NO:NO₂ speciation entering the catalyst.

Transient reactor studies have further highlighted the importance of NH₃ storage and release phenomena. Temperature-programmed desorption (TPD), concentration step-change experiments, and transient response analyses have demonstrated that catalyst NH₃ storage capacity (NSC) strongly influences performance during dynamic operation (Nova et al., 2011; Olsson et al., 2008). While increased NH₃ storage can sustain NO_x conversion during temporary fluctuations in reductant availability, it may also lead to delayed NH₃ breakthrough and increased NH₃ slip. These effects are particularly important in NH₃-fuelled systems, where NH₃ simultaneously acts as the fuel, pollutant, and SCR reductant.

6.2 Engine-based and integrated aftertreatment studies

More recently, experimental research has increasingly shifted from laboratory-scale reactors towards integrated engine-aftertreatment investigations. Studies on NH₃ and NH₃-diesel DF engines have demonstrated the feasibility of utilising engine-out NH₃ slip as a reductant for downstream SCR reactions (Cardoso et al., 2021; Lesmana et al., 2021). These investigations confirmed that passive utilisation of NH₃ slip can significantly reduce external urea demand; however, the naturally occurring NH₃:NO_x ratio often varies substantially with engine operating condition and is frequently insufficient to achieve complete NO_x removal across the full operating range. Consequently, supplementary NH₃ dosing, combustion optimisation, or alternative control strategies may still be required to achieve near-zero emissions.

Alongside NO_x conversion, the environmental significance of N₂O emissions has received increasing attention. Experimental studies have shown that N₂O formation depends strongly on catalyst composition, temperature, and NO₂ concentration (Jung et al., 2022). Excessive NO₂ promotes the formation of NH₄NO₃ species, which subsequently decompose to produce N₂O at intermediate temperatures. Comparative catalyst evaluations have further demonstrated that Cu-zeolites generally exhibit higher N₂O formation tendencies than vanadia-based catalysts, while Fe-zeolites display intermediate behaviour depending on operating conditions. These findings highlight the need to evaluate catalyst performance using both NO_x conversion efficiency and N₂O selectivity, particularly for NH₃-fuelled applications where NH₃-rich conditions are common.

Advances in characterisation techniques have significantly improved mechanistic understanding of SCR catalysts and reaction pathways. In situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS), X-ray absorption spectroscopy (XAS), and operando spectroscopic techniques have enabled direct observation of adsorbed NH₃ species, nitrates, nitrites, and catalyst redox transitions during operation (Paolucci et al., 2017). These studies have provided experimental evidence supporting both L-H and E-R reaction pathways and have played an important role in validating modern microkinetic models.

Despite substantial progress, several experimental challenges remain. Most published studies continue to use simplified synthetic gas mixtures that do not fully replicate the complexity of NH₃-combustion exhaust. Long-term investigations examining the combined effects of high NH₃ concentrations, water vapour, sulphur species, and thermal cycling remain relatively scarce. Furthermore, the interactions between SCR and ASC catalysts, and upstream oxidation catalysts under realistic operating conditions are still not fully understood.

Overall, experimental research has established the technical feasibility of SCR for NH₃-fuelled engines while providing critical insight into catalyst activity, selectivity, reaction mechanisms, and durability. However, further integrated engine-scale investigations, transient studies, and long-term ageing assessments are required to optimise catalyst systems for practical NH₃-combustion applications and to balance NO_x reduction with control of NH₃ slip and N₂O emissions.

7 Coupling between combustion and aftertreatment

The performance of SCR systems in NH₃-fuelled engines is intrinsically linked to upstream combustion processes. Unlike conventional diesel engines, where NH₃ availability can be independently regulated through urea dosing, NH₃-fuelled systems generate both NO_x and NH₃ slip directly from combustion. Consequently, the composition of the exhaust entering the catalyst is determined by engine operating conditions, creating a strong interdependence between combustion and aftertreatment performance (Cardoso et al., 2021; Lesmana et al., 2021).

Among the combustion parameters influencing SCR operation, engine load, combustion phasing, and equivalence ratio exert the greatest effect on exhaust composition. Experimental studies on NH₃-fuelled spark-ignition (SI) and compression-ignition (CI) engines have shown that low-load operation generally produces higher NH₃ slip because of incomplete fuel conversion, while NO_x emissions remain comparatively low due to reduced combustion temperatures (Kobayashi et al., 2019; Valera-Medina et al., 2018). As load increases, improved combustion efficiency reduces NH₃ slip, but higher in-cylinder temperatures promote fuel-NO formation, resulting in increased NO_x emissions. This inverse relationship complicates maintenance of the optimum NH₃:NO_x ratio required for efficient SCR operation.

The equivalence ratio has a similarly important influence. Lean combustion promotes oxidation of fuel-bound nitrogen species and favours NO formation, whereas richer conditions suppress NO_x production but increase emissions of unburned NH₃ and other partially oxidised species (Otomo et al., 2018). Experimental investigations by Lesmana et al., 2019 demonstrated that engine-out NH₃ concentrations vary substantially with equivalence ratio (ɸ) and are frequently insufficient to achieve complete NO_x conversion without supplementary reductant supply. These observations suggest that passive utilisation of NH₃ slip as the sole SCR reductant may only be feasible over limited operating regions.

H₂-assisted NH₃ combustion introduces additional interactions between combustion and aftertreatment systems. H₂ enrichment improves combustion stability, accelerates flame propagation, and enhances NH₃ conversion efficiency, thereby reducing NH₃ slip and extending lean operating limits (Lhuillier et al., 2020). However, the increased concentrations of reactive radicals such as O, OH, and HO₂ can alter NO oxidation chemistry and influence the NO:NO₂ ratio entering the catalyst. Liu et al., 2011 reported that residual H₂ in the exhaust can promote NO oxidation to NO₂, potentially enhancing low-temperature fast-SCR activity. H₂ addition also increases exhaust water vapour concentrations, which may further influence downstream catalyst performance through competitive adsorption and changes in reaction kinetics. Consequently, optimisation of NH₃-H₂ combustion strategies should consider aftertreatment performance in addition to combustion efficiency and stability.

7.1 NH₃-NO_x ratio and catalyst performance

The NH₃:NO_x ratio at the SCR inlet remains the most important parameter governing catalyst performance. Standard SCR chemistry requires approximately equimolar quantities of NH₃ and NO_x for complete conversion, whereas deviations from this ratio either limit NO_x reduction or increase NH₃ slip (Koebel et al., 2000). In NH₃-fuelled engines, however, this ratio is controlled primarily by combustion behaviour rather than external dosing systems. Although (Cardoso et al., 2021) proposed that engine-out NH₃ slip could potentially eliminate the need for urea injection, experimental studies indicate that substantial fluctuations in NH₃:NO_x ratio occur during load changes and transient operation (Lesmana et al., 2021). Furthermore, while increasing NH₃ availability generally improves NO_x conversion, excessive NH₃ can promote NH₃ oxidation and N₂O formation, highlighting the need for careful management of reductant availability (Wu et al., 2024).

Thermal coupling between the engine and catalyst system is equally important. NH₃ combustion frequently produces lower exhaust gas temperatures (EGT) than conventional diesel combustion, particularly under lean and low-load conditions (Valera-Medina et al., 2018). These temperatures may fall below the optimal operating window of many SCR catalysts, reducing NH₃ activation rates and limiting NO_x conversion efficiency. Conversely, high-load operation can expose catalysts to temperatures at which NH₃ oxidation and N₂O formation become increasingly significant (Jung et al., 2022). As a result, catalyst performance cannot be considered independently of the thermal characteristics of the combustion process.

7.2 Integrated combustion-aftertreatment optimisation

The strong coupling between combustion and aftertreatment extends beyond NH₃ slip and NO_x emissions to include GHG formation. Improvements in combustion efficiency can simultaneously reduce NH₃ slip and engine-out N₂O emissions, thereby lowering the burden placed on downstream catalysts (Pedersen et al., 2023). demonstrated that controlling combustion-generated NH₃ and N₂O is as important as catalyst optimisation when minimising overall emissions from NH₃-fuelled engines.

Recognising these interactions, recent modelling studies have increasingly adopted integrated engine-aftertreatment frameworks that simultaneously predict combustion emissions, NH₃ slip, catalyst NO_x conversion efficiency, and secondary pollutant formation (Zhang et al., 2023; Gopinathan and Mohamed Ibrahim, 2025). These coupled approaches enable optimisation of combustion parameters and catalyst operation at the system level rather than treating them as separate subsystems. Such integration is particularly important for NH₃-fuelled engines because NH₃ functions simultaneously as the fuel, pollutant, and SCR reductant.

Figure 2 illustrates the coupling between combustion and aftertreatment in NH₃-fuelled engines. Engine operating conditions govern the formation of NO_x, NH₃ slip, H₂O, and N₂O, which determine the inlet conditions to the SCR catalyst. In turn, catalyst performance depends on these engine-out species and influences the overall emissions released to the atmosphere.

Overall, available evidence demonstrates that SCR performance in NH₃-fuelled engines cannot be optimised independently of combustion. Exhaust composition, temperature, NH₃ slip, NO_x speciation, and N₂O formation are all strongly governed by combustion behaviour, while catalyst performance in turn influences overall emissions. Future developments are therefore likely to focus on integrated combustion–aftertreatment optimisation, where engine calibration, catalyst selection, and emissions control are treated as components of a single interconnected system.

8 N₂O formation and mitigation strategies

Nitrous oxide (N₂O) is emerging as one of the principal environmental challenges associated with NH₃-fuelled combustion systems because reductions in NO_x emissions may be accompanied by increases in greenhouse-gas emissions. With a global warming potential approximately 273 times greater than that of CO₂ over a 100-year time horizon, even relatively small quantities of N₂O can significantly affect the overall environmental performance of NH₃-fuelled engines. Although NH₃ combustion generally produces limited engine-out N₂O because of the high combustion temperatures involved, substantial quantities can be generated within SCR and NH₃ oxidation catalysts as undesired by-products of NO_x abatement reactions (Guan et al., 2014; Jung et al., 2022). Consequently, minimising N₂O emissions has become an increasingly important objective in the development of aftertreatment systems for NH₃-fuelled applications.

N₂O formation within NH₃-SCR systems is generally attributed to two principal pathways. The first involves partial oxidation of NH₃, which becomes increasingly important at elevated temperatures and on highly oxidative catalyst surfaces. Depending on catalyst composition and operating conditions, NH₃ oxidation proceeds through reactive intermediates such as NH₂ and NH species and may ultimately produce N₂, NO, or N₂O. Excess NH₃ concentrations and high oxygen availability generally favour N₂O formation through these pathways (Ramis et al., 1996).

The second major pathway involves the formation and decomposition of NH₄NO₃. Under low-temperature and NO₂-rich conditions, NH₃ reacts with NO₂ to form surface ammonium nitrate species that accumulate within catalyst pores. As temperature increases, these species become unstable and decompose, generating N₂O and water vapour (Madia et al., 2002). This mechanism is particularly important during fast-SCR operation and under conditions where local NO₂ concentrations exceed the optimum NO₂:NO_x ratio. Accordingly, NH₄NO₃ should be regarded not merely as a catalyst deposit but as a key low-temperature SCR intermediate that strongly influences NO_x conversion, NH₃ storage behaviour, and N₂O formation.

8.1 Influence of catalyst composition and operating conditions

The relative importance of these pathways depends strongly on catalyst composition and operating temperature. Comparative studies of commercial SCR catalysts have demonstrated substantial differences in N₂O selectivity. Vanadia-based catalysts generally exhibit relatively low N₂O formation across a broad operating window because of their balanced acid-redox properties and moderate NH₃ oxidation activity. In contrast, Cu-exchanged zeolites, although highly active for low-temperature NO_x reduction, can generate significant quantities of N₂O under NH₃-rich conditions and during NH₃ oxidation reactions (Paolucci et al., 2017). Fe-zeolite (FeZ) catalysts often exhibit lower low-temperature activity but may promote N₂O formation at elevated temperatures where NH₃ oxidation becomes increasingly significant. Experimental work by (Jung et al., 2022) further demonstrated that Fe-zeolite catalysts initiated N₂O formation earlier than Cu-zeolites owing to differences in NO depletion and NO₂ consumption pathways.

Temperature exerts a particularly complex influence on N₂O selectivity. At temperatures below approximately 250 °C, N₂O formation is dominated by NH₄NO₃ chemistry and incomplete SCR reactions. As catalyst temperature increases, decomposition of stored NH₄NO₃ can generate transient N₂O emission peaks. At higher temperatures, typically above 350 °C–400 °C, NH₃ oxidation becomes the dominant source of N₂O, especially in the presence of excess NH₃ or highly oxidative catalyst formulations (Jung et al., 2022). This dual temperature dependence complicates catalyst optimisation because strategies designed to suppress low-temperature N₂O formation may inadvertently increase high-temperature NH₃ oxidation activity.

8.2 Strategies for N₂O mitigation

Consequently, N₂O mitigation requires both catalyst-level and system-level approaches. Catalyst development efforts increasingly focus on formulations that maximise NO_x reduction to N₂ while suppressing competing NH₃ oxidation pathways. Adjustments to active metal loading, promoter composition, and catalyst acidity have been shown to improve N₂ selectivity while maintaining high deNO_x activity (Damma et al., 2019; Paolucci et al., 2017). Hybrid catalyst architectures incorporating SCR catalysts with downstream ammonia slip catalysts (ASC) have also been investigated as a means of controlling NH₃ emissions while limiting secondary N₂O formation.

Operational control is equally important. Maintaining the NH₃:NO_x ratio close to unity minimises excess NH₃ available for oxidation, while controlling the NO₂:NO_x ratio limits ammonium nitrate formation (Song et al., 2015). Thermal management through catalyst placement, insulation, and exhaust heat recovery can further reduce residence within temperature windows that favour N₂O generation. For NH₃-fuelled engines, these strategies must be considered alongside combustion optimisation because both NH₃ slip and NO_x formation originate directly from the combustion process.

Recent studies have also demonstrated that catalyst ageing can significantly influence N₂O behaviour. Hydrothermal ageing alters not only catalyst activity but also the temperature window and selectivity for N₂O formation (Liu et al., 2023). reported that ageing shifts N₂O formation towards higher temperatures and modifies the catalyst’s inhibitory effect on N₂O production. These findings suggest that optimisation based solely on NO_x conversion efficiency may be insufficient, as catalysts exhibiting superior deNO_x performance do not necessarily minimise N₂O emissions (Jung et al., 2022).

Beyond conventional SCR optimisation, direct catalytic decomposition of N₂O has emerged as a promising supplementary mitigation strategy. Fe-containing zeolites and transition-metal-exchanged catalysts have demonstrated significant N₂O decomposition activity between approximately 400 °C and 600 °C, converting N₂O directly to N₂ and O₂ (Zhuang et al., 2024). Such materials could potentially be integrated into future NH₃-engine aftertreatment architectures to provide simultaneous control of NO_x, NH₃ slip, and N₂O emissions. An additional advantage is that many N₂O decomposition catalysts are inhibited by excess O₂ because it competes for active sites and limits catalyst regeneration. While this represents a challenge in O₂-rich diesel exhaust, the lower O₂ concentrations often encountered in NH₃-fuelled engine exhaust may create more favourable conditions for N₂O decomposition. This raises the possibility of integrated catalyst systems capable of simultaneously achieving deNO_x performance and GHG mitigation.

Overall, N₂O mitigation is becoming a critical design objective for NH₃-fuelled engine aftertreatment systems. Future catalyst development must balance NO_x conversion efficiency, NH₃ slip control, catalyst durability, and N₂O selectivity simultaneously. Achieving this objective will require improved mechanistic understanding of N₂O formation pathways, optimisation of catalyst formulations, and closer integration of combustion and aftertreatment strategies under realistic NH₃-rich exhaust conditions. The major Nitrous oxide formation pathways and mitigation strategies are summarised in Table 7.

TABLE 7. Major N₂O formation pathways and mitigation strategies.

9 Challenges and research gaps

A number of important challenges remain before NH₃-SCR systems can be fully optimised for NH₃-fuelled combustion applications. Although substantial progress has been made in catalyst development, reaction mechanism identification, and numerical modelling, many existing studies remain focused on conventional diesel-derived exhaust compositions rather than the NH₃-rich environments characteristic of NH₃ combustion (Lesmana et al., 2021; Valera-Medina et al., 2018).

One of the most significant limitations is the lack of integrated experimental and modelling studies. Detailed reaction mechanisms have been developed using laboratory reactors, spectroscopic techniques, and density functional theory, while increasingly sophisticated kinetic and reactor-scale models have been proposed (Paolucci et al., 2017). However, relatively few investigations have combined mechanistic modelling with validation under realistic NH₃-engine operating conditions. Consequently, many existing models remain catalyst-specific and cannot be readily transferred across catalyst formulations, exhaust compositions, or operating regimes.

9.1 Experimental and modelling challenges

The availability of representative experimental data also remains limited. Most SCR performance maps, kinetic parameters, and model validation datasets have been generated using synthetic gas mixtures designed to emulate diesel exhaust. NH₃-fuelled engines, however, produce exhaust streams containing elevated NH₃ slip, altered NO:NO₂ ratios, lower O₂ concentrations, and high water-vapour contents. These differences may substantially influence catalyst activity, selectivity, and durability, highlighting the need for dedicated experimental studies under realistic NH₃-combustion conditions.

At the reactor and system scales, modelling capabilities remain underdeveloped for NH₃-fuelled applications. While detailed CFD frameworks coupling transport phenomena with SCR kinetics have been successfully applied to urea-SCR systems, their extension to NH₃-fuelled engines remains limited (Guo Y, 2017). Future models should explicitly incorporate engine-out NH₃ slip, coupled oxidation-reduction chemistry, NH₃ storage dynamics, and N₂O formation pathways to enable accurate prediction of catalyst performance under NH₃-rich exhaust conditions. Similarly, integrated combustion-aftertreatment models are required to capture the strong coupling between combustion-generated NH₃, NO_x formation, and downstream catalyst behaviour.

A major unresolved issue concerns the formation of N₂O. Although several pathways have been proposed, including NH₃ oxidation, NH₂NO intermediates, and NH₄NO₃ formation and decomposition, their relative contributions remain highly dependent on catalyst composition and operating conditions. This uncertainty limits both catalyst optimisation and predictive modelling, particularly because catalysts exhibiting excellent NO_x conversion do not necessarily minimise N₂O emissions. Improved mechanistic understanding of N₂O formation under NH₃-rich conditions therefore remains a critical research priority.

Transient operation represents another important knowledge gap. Most experimental studies and model validations are performed under steady-state conditions, whereas practical engines experience continual fluctuations in load, temperature, and exhaust composition. Under such conditions, NH₃ storage and release dynamics can strongly influence NO_x conversion, NH₃ slip, and N₂O formation (Olsson et al., 2008; Song et al., 2015). Developing transiently validated kinetic models and conducting dynamic engine-scale experiments will therefore be essential for future system optimisation.

9.2 Catalyst durability and ageing

Long-term catalyst durability also requires further investigation. NH₃-fuelled engines expose catalysts to prolonged NH₃-rich atmospheres, high water-vapour concentrations, and, in DF applications, potential contaminants originating from pilot fuels. These conditions may accelerate hydrothermal degradation, active-site migration, sulphation, and zeolite framework deterioration (Paolucci et al., 2017). In addition to reducing catalyst activity, ageing may alter reaction pathways and catalyst selectivity, potentially affecting NH₃ slip and N₂O formation. Recent studies further suggest that water adsorption plays a more significant role in SCR kinetics than is currently represented in most models, particularly given the high water-vapour concentrations present in NH₃-combustion exhaust (Voglar et al., 2023).

Finally, there remains a need for catalyst formulations specifically tailored to NH₃-fuelled exhaust environments. Current commercial SCR catalysts were largely developed for diesel applications and may not be optimised for sustained exposure to elevated NH₃ concentrations, H₂ co-firing, reduced O₂ availability, or the unique emissions characteristics of NH₃ combustion. Future catalyst development should therefore adopt a multi-objective approach that simultaneously considers NO_x conversion, NH₃ slip, N₂O mitigation, catalyst durability, and overall GHG performance.

Overall, future progress will require closer integration of catalyst design, combustion development, experimental validation, and multi-scale modelling. Addressing these challenges will be essential for enabling SCR systems capable of delivering effective NO_x control while minimising NH₃ slip and N₂O emissions in next-generation NH₃-fuelled power and propulsion systems. Current research gaps and future needs for ammonia combustion exhaust treatment are summarised in Table 8.

TABLE 8. Current research gaps and future needs for treatment of ammonia combustion exhaust.

10 Future research directions

The application of SCR technology to NH₃-fuelled combustion presents challenges that extend beyond those encountered in conventional diesel aftertreatment systems. The simultaneous presence of NH₃ slip, variable NO_x speciation, high water-vapour concentrations, and fluctuating exhaust temperatures creates a highly dynamic reaction environment that requires both catalyst innovation and improved system integration. Unlike conventional urea-SCR systems, where reductant dosing can be independently controlled, NH₃-fuelled engines generate both NO_x and NH₃ slip directly from the combustion process, creating a strong coupling between combustion behaviour and catalyst performance.

A key opportunity offered by NH₃-fuelled engines is the potential utilisation of engine-out NH₃ slip as the reductant for downstream SCR reactions. In principle, this could eliminate the need for external urea infrastructure and simplify aftertreatment system design (Kobayashi et al., 2019; Kuta et al., 2023). However, experimental studies have demonstrated that the NH₃:NO_x ratio varies significantly with engine operating conditions, often resulting in either insufficient NH₃ for complete NO_x conversion or excessive NH₃ slip that promotes secondary emissions. Effective utilisation of combustion-generated NH₃ therefore remains a major challenge requiring coordinated optimisation of both engine operation and catalyst performance.

Current SCR catalyst technologies each offer distinct advantages and limitations under NH₃-combustion conditions. Cu-exchanged zeolites provide excellent low-temperature activity and high NH₃ storage capacity but may exhibit increased N₂O formation under NH₃-rich conditions. Fe-zeolites offer superior thermal stability and favourable high-temperature performance, while vanadia-based catalysts remain highly selective for NO_x reduction but are constrained by narrower operating windows and long-term durability concerns (Paolucci et al., 2017). These limitations have stimulated growing interest in catalyst formulations specifically tailored to NH₃-rich exhaust environments.

Consequently, future catalyst development is increasingly focused on multifunctional and hybrid catalyst architectures. Layered Cu-Fe zeolite systems, combined SCR-ammonia slip catalyst (ASC) configurations, and catalysts engineered to suppress NH₃ oxidation are being explored to broaden operating windows and improve selectivity. Beyond conventional SCR optimisation, there is growing interest in integrating dedicated N₂O decomposition functionality within aftertreatment systems. Multifunctional catalyst platforms capable of simultaneously reducing NO_x, utilising NH₃ slip, and decomposing N₂O could provide a pathway toward near-zero emissions from NH₃-fuelled engines.

Advances in modelling are expected to play a critical role in accelerating catalyst and system development. Significant progress has been achieved through density functional theory (DFT), microkinetic modelling, and multi-scale simulation frameworks; however, predictive capability remains limited under realistic NH₃-engine conditions. Improved integration of atomistic reaction energetics, operando spectroscopy, microkinetic models, and reactor-scale simulations offers considerable potential for improving mechanistic understanding and reducing uncertainty in model parameters (Paolucci et al., 2017; Voglar et al., 2023). At the same time, machine-learning-assisted catalyst design and data-driven optimisation methods may help accelerate catalyst screening and reduce experimental costs.

Future developments will likely require a shift from isolated catalyst optimisation toward integrated combustion-aftertreatment design. SCR performance cannot be optimised independently because combustion conditions directly determine catalyst inlet temperature, NH₃ availability, NO_x concentration, and exhaust composition. Consequently, future NH₃-engine systems will require coordinated control of combustion phasing, fuel substitution ratio, exhaust thermal management, NH₃ slip utilisation, and catalyst operation. Integrated modelling and control frameworks capable of simultaneously optimising combustion and aftertreatment performance are therefore expected to become increasingly important. Future aftertreatment development is expected to move towards integrated catalyst-system optimisation, combining heterogeneous reaction engineering with advanced modelling and control strategies (Lott and Deutschmann, 2023).

A further priority is the generation of representative experimental datasets for model validation and catalyst assessment. Long-duration engine testing, transient operating cycles, hydrothermal ageing studies, and experiments using realistic NH₃-combustion exhaust compositions remain comparatively scarce. In particular, greater attention should be given to the effects of high water-vapour concentrations, NH₃-rich operating conditions, catalyst ageing, and transient operation, as these factors are expected to strongly influence catalyst activity, selectivity, and durability.

Ultimately, future research must move beyond maximising NO_x conversion alone and instead pursue holistic optimisation of emissions performance. Achieving this objective will require coordinated advances in catalyst science, reaction engineering, combustion development, control systems, and predictive modelling. The successful integration of these disciplines will be essential for developing NH₃-fuelled power and propulsion systems capable of simultaneously minimising NO_x, NH₃ slip, and N₂O emissions while maintaining long-term catalyst durability and operational efficiency.

11 Conclusion

NH₃ is increasingly recognised as a promising carbon-free fuel for future transportation and power-generation systems. However, the successful deployment of NH₃-fuelled engines depends on effective control of NO_x emissions and unburned NH₃, making SCR technology a critical component of future exhaust aftertreatment systems.

This review has examined the unique characteristics of NH₃ combustion exhaust and their implications for SCR performance. Compared with conventional diesel exhaust, NH₃-fuelled systems present substantially different operating conditions, including elevated NH₃ slip, variable NO:NO₂ ratios, lower exhaust temperatures, and an increased risk of N₂O formation. Together, these factors strongly influence catalyst activity, selectivity, and durability.

A review of catalyst technologies highlights that no single catalyst family currently satisfies all performance requirements. Cu-exchanged zeolites offer excellent low-temperature activity, Fe-zeolites provide superior high-temperature stability, and vanadia-based catalysts remain highly selective within intermediate temperature ranges. Consequently, increasing attention is being directed towards hybrid catalyst architectures that combine the advantages of multiple catalyst systems while mitigating their individual limitations.

The mechanistic complexity of NH₃-SCR reactions presents significant challenges for both catalyst design and predictive modelling. Multiple reaction pathways, including standard SCR, fast SCR, NH₃ oxidation, ammonium nitrate chemistry, and N₂O formation, operate simultaneously and are strongly influenced by catalyst composition and operating conditions. Although substantial progress has been made in kinetic modelling, further development is required to accurately predict catalyst behaviour under realistic NH₃-combustion conditions.

Experimental investigations have provided valuable insights into catalyst performance, reaction mechanisms, and system integration. However, important gaps remain regarding transient operation, catalyst ageing, water-vapour effects, and N₂O selectivity under NH₃-rich exhaust environments. Furthermore, the strong coupling between combustion and aftertreatment emphasises the need for integrated optimisation approaches that consider engine operation and catalyst performance simultaneously.

A central conclusion emerging from this review is that NH₃-fuelled engines fundamentally alter the operating environment of SCR systems. Elevated NH₃ slip, high water-vapour concentrations, altered NO_x speciation, and the growing importance of N₂O emissions challenge many of the assumptions established for conventional diesel aftertreatment. As a result, catalyst development, kinetic modelling, combustion optimisation, and emissions-control strategies can no longer be considered independently. Future progress will depend on the development of catalyst systems specifically tailored to NH₃-combustion exhaust, supported by realistic experimental validation and integrated multi-scale modelling frameworks. Achieving robust control of NO_x, NH₃ slip, and N₂O emissions will be essential for realising the full potential of NH₃ as a sustainable fuel for next-generation power and propulsion systems.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Muhammad Waqas, Cummins, United States

Reviewed by: Nur Irmawati Om, Universiti Tenaga Nasional, Malaysia

Joseph Mmbaga, University of Alberta, Canada

Footnotes

ASC, Ammonia slip catalyst; CFD, Computational fluid dynamics; CHA, Chabazite; DOC, Diesel oxidation catalyst; DFT, Density functional theory; DPF, Diesel particulate filter; DRIFTS, Diffuse reflectance infrared fourier transform spectroscopy; E-R, Eley-Rideal; EATS, Exhaust aftertreatment system; EGT, Exhaust gas temperature; GHSV, Gas hourly space velocity; L-H, Langmuir–Hinshelwood; LBV, Laminar burning velocity; MvK, Mars-van Krevelen; NH₃-SCR, Ammonia selective catalytic reduction; NSC, NH₃ storage capacity; SCR, Selective catalytic reduction; TPD, Temperature-programmed desorption; XAS, X-ray absorption spectroscopy.