Journal of Cleaner Production 434 (2024) 140274

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Experimental investigation of the effect of ammonia substitution ratio on an

ammonia-diesel dual-fuel engine performance
Junheng Liu a, Jinlong Liu b, c, d, e, *
a
School of Automotive and Traffic Engineering, Jiangsu University, Zhenjiang, 212013, China
b
Key Laboratory of Clean Energy and Carbon Neutrality of Zhejiang Province, Zhejiang University, Hangzhou, 310027, China
c
Jiaxing Research Institute, Zhejiang University, Jiaxing, 314031, China
d
State Key Laboratory of Clean Energy Utilization, Zhejiang University, Hangzhou, 310027, China
e
Power Machinery and Vehicular Engineering Institute, Zhejiang University, Hangzhou, 310027, China

A R T I C L E I N F O A B S T R A C T

Handling Editor: Panos Seferlis This research experimentally examines an ammonia-diesel dual fuel system, not widely covered in existing
literature but acknowledged as an effective approach to reduce engine carbon footprints in alignment with global
Keywords: decarbonization trends, focusing on enhancing understanding of performance, combustion, and emission char­
Dual-fuel engine acteristics vital for designing and optimizing these engines for commercialization. The results indicate that in an
Ammonia-diesel
ammonia-air mixture atmosphere, diesel undergoes a prolonged ignition process, enhancing the role of premixed
ammonia substitution ratio
combustion. Increasing ammonia substitution at constant speed and load reduces the weight of mixing-controlled
Combustion process
emission performance combustion, leading to larger areas in the chamber where diesel flames cannot reach. Ammonia oxidation mainly
occurs near the diesel flame, as overly lean ammonia-air mixtures fail to sustain stable flame propagation. This
results in considerable unburned ammonia emissions, particularly in regions beyond diesel flame reach and
engine crevices, adversely affecting combustion efficiency and diminishing the thermal efficiency advantages of
such operation. Furthermore, while ammonia combustion contributes to fuel-borne nitrogen oxides (NOx) for­
mation, the overall NOx emissions from ammonia-diesel engines are reduced due to lower combustion tem­
peratures and the deNOx properties of amino groups. However, the partial oxidation of ammonia during the late
expansion stroke generates significant levels of nitrous oxide (N2O), a greenhouse gas, in the emissions, coun­
teracting the intended reduction of carbon dioxide. These experimental findings highlight the necessity of
focusing on both improving in-cylinder combustion quality and developing effective aftertreatment systems for
NH3 and N2O capture, as crucial steps towards enhancing the environmental performance and market viability of
ammonia-diesel dual fuel engines.

1. Introduction carbon dioxide (CO2) and other greenhouse gas (GHG) emissions, ne­
cessitates meticulous strategies to curtail its carbon footprint, aiming for
The internal combustion engine is a hallmark invention of the a transition towards carbon neutrality (Reiter and Kong, 2011; Duan
modern industrial revolution (Di Blasio et al., 2023; Di Luca et al., et al., 2023b). Given that engines inherently emit CO2 by combusting
2022). As a power source with high thermal efficiency, superior power high-carbon fossil fuels, transitioning to sustainable low or zero-carbon
density, and versatility in fuel compatibility (Gao et al., 2022a, 2022b), alternatives presents a direct solution to lessen or supplant petroleum
the engine remains crucial in sectors like agriculture and heavy trans­ dependence, thereby curtailing the associated carbon footprint (Yousefi
portation, highlighting its fundamental role in boosting societal pro­ et al., 2022a). Among zero-carbon fuels, ammonia stands out as a su­
ductivity and advancing human progress (Alrwashdeh et al., 2022; Yang perior hydrogen energy carrier, offering greater energy density,
et al., 2024). In recent years, a global surge in collective efforts to improved safety (Shin and Park, 2023), and a more mature
combat climate change has ushered in a shared ambition and vision of production-storage-transportation system, enhancing its competitive
achieving carbon neutrality (Duan et al., 2023a; Huang and Liu, 2024). edge (Niki et al., 2019). Consequently, the potential of
The internal combustion engine, being a substantial contributor to ammonia-powered engines has spurred extensive research and policy

* Corresponding author. Zhejiang University, Hangzhou, 310027, China.

E-mail address: [email protected] (J. Liu).

https://doi.org/10.1016/j.jclepro.2023.140274
Received 14 October 2023; Received in revised form 26 November 2023; Accepted 16 December 2023
Available online 22 December 2023
0959-6526/© 2023 Elsevier Ltd. All rights reserved.
J. Liu and J. Liu Journal of Cleaner Production 434 (2024) 140274

developments globally, focusing on ammonia-centric standards, the is the generation of nitrous oxide (N2O) during ammonia combustion, a
conceptualization of an ammonia-integrated society, and the develop­ greenhouse gas uncommon in traditional engines and resistant to miti­
ment of ammonia-adapted engine technologies. gation by diesel engine after-treatment devices (Yousefi et al., 2022b).
Ammonia (NH3), known for its high resistance to autoignition, offers The distinct characteristics of ammonia-diesel dual fuel engines under­
a viable alternative as a fuel for internal combustion engines. It can be score the urgent need for comprehensive studies examining their per­
initiated for combustion in engines through both spark ignition and pilot formance, combustion, and emission profiles. However, an examination
ignition using high cetane number fuels such as diesel, biodiesel, or of the current state of research in ammonia-diesel dual fuel engines in­
dimethyl ether (Sun et al., 2023). Despite its potential, ammonia ex­ dicates a notable scarcity in the literature, with the existing studies
hibits two significant limitations when used in spark ignition (SI) en­ primarily comprising limited three-dimensional (3D) computational
gines, which are predominantly designed for gasoline. Firstly, ammonia fluid dynamics (CFD) simulations (Wang et al., 2023a, 2024). This trend
demands a high spark energy for combustion initiation (Kurien and stems from the nascent stage of ammonia engine development, evi­
Mittal, 2023). Secondly, its laminar flame speed is considerably slower denced by the scarcity of commercial fuel injectors suitable for
than conventional fuels like gasoline (Liu and Liu, 2024). To substitute ammonia. The corrosive and toxic nature of ammonia necessitates spe­
gasoline in SI engines, it is pivotal for the alternative fuel to exhibit a cial approvals and safety protocols for experimental use. Moreover,
flame speed akin to gasoline. Existing studies suggest the addition of adapting conventional engines for ammonia fuel requires intricate
hydrogen to elevate the laminar flame speed of ammonia. It is estimated modifications, particularly in the fuel supply system. Despite the paucity
that a minimum of 10% hydrogen is necessary to achieve acceptable of experimental data on ammonia-diesel operations, researchers have
combustion efficiency in the engine combustion chamber (Mørch et al., resorted to 3D CFD simulations for preliminary analysis, driven by the
2011). Nonetheless, this approach still encounters a longer ignition lag, pressing industry and governmental demands for decarbonization
leading to increased cycle-to-cycle variations (Comotti and Frigo, 2015). (Wang et al., 2023b, 2023c). Dual-fuel engine simulations are inherently
The reason is the kernel inception process, which is heavily influenced complex due to factors such as fuel vaporization, mixing, and combus­
by the laminar flame. To attain a laminar flame speed parallel to gaso­ tion, which create multifaceted thermal-fluid-chemical interactions.
line, a mixture containing ~40% hydrogen with ammonia is essential These interactions are further complicated by varying equivalence ratios
(Yan et al., 2023). This high demand for hydrogen presents challenges, and temperature gradients. Critical considerations include both low and
particularly concerning the on-board hydrogen production system high-temperature reaction pathways that govern ignition and combus­
through ammonia decomposition using catalysts (Ryu et al., 2014). tion, encompassing numerous intermediate reactions and products,
Considering these challenges, the prospect of retrofitting diesel engines thereby amplifying mechanism complexity, uncertainty, and potential
for diesel-ammonia dual fuel operation becomes appealing (Yousefi errors. The absence of experimental data to support ammonia-diesel
et al., 2022b). Such modifications are not only simpler but also CFD models poses a significant challenge (Wang et al., 2023a).
cost-effective since they eliminate the need for an on-board hydrogen Without empirical benchmarks, accurately predicting physical and
production system. However, research on diesel-ammonia dual fuel chemical phenomena becomes difficult (Wang et al., 2024). The selec­
engines remains scant, possibly because this combustion mode can only tion and calibration of sub-models (e.g., combustion and injection
partially replace petroleum fuels rather than facilitating a full transition models) are hindered by inherent uncertainties and errors, which may
as in the case of ammonia spark ignition mode. It is noteworthy to compound when integrated, leading to increased uncertainty in simu­
mention the emergence of E-diesel (i.e. synthetic diesel fuel) (Medra­ lation outcomes (Wang et al., 2023a, 2024). If these CFD models are not
no-García et al., 2022), a carbon-neutral alternative synthesized from meticulously validated and adjusted, their reliability and the accuracy of
carbon dioxide, water, and electricity using renewable energy to pro­ resulting predictions about engine performance could be significantly
duce blue crude—which contrasts with traditional crude oil and is compromised (Wang et al., 2023a, 2024). Addressing this knowledge
refined into E-diesel—underscores the significance of exploring gap, this study aims to provide high-quality, reliable experimental data
diesel-ammonia dual fuel operation modes. for ammonia-diesel dual fuel engines. Furthermore, this study offers a
The concept of dual fuel operation is not novel, having been exten­ comprehensive analysis of engine performance based on recent indoor
sively employed in natural gas-diesel (Lounici et al., 2014; Belgiorno experimental tests, vital for understanding the unique combustion be­
et al., 2019) and methanol-diesel dual fuel (Liu et al., 2015; Wang et al., haviors of these engines and providing crucial insights for their future
2022) engines. Drawing upon these experiences, certain characteristics development and optimization. Subsequent sections will detail the
of dual fuel engines have become well-documented. For instance, under experimental apparatus and operating conditions underpinning the data
low load conditions, these engines tend to exhibit markedly low com­ gathered for ammonia-diesel dual fuel operations, followed by an
bustion efficiency, leading to a minimal alternative fuel substitution. in-depth evaluation of performance and the impacts of varying ammonia
Additionally, at very high loads, there is also a necessity to limit the substitution ratios on the engine.
alternative fuel substitution rate due to potential knocking of the un­
burned mixture and the possible high-pressure rise rate that could 2. Materials and methods
jeopardize engine components. Under medium to high load conditions,
dual fuel engines can accommodate a higher alternative fuel substitution Experiments are conducted on an ammonia-diesel dual-fuel engine,
rate, facilitating a reduction in conventional petroleum consumption. equipped with an ammonia port fuel injection system alongside a high-
Introducing ammonia as a substitute for natural gas or methanol in the pressure common rail diesel injection system. This engine originates
premixed fuel brings forth distinctive features in the dual fuel engine. from a turbocharged and intercooled four-cylinder inline diesel system
Notably, the heightened resistance to auto-ignition of ammonia negates with a high-pressure common rail but without exhaust gas recirculation.
the possibility of knocking with the compression ratios found in During the conversion process, ammonia is fumigated ahead of the
contemporary diesel engines (Pei et al., 2023). The introduction of intake valves using gaseous injectors, resulting in the formation of a
ammonia also activates the element nitrogen and opens channels for homogeneous ammonia-air mixture within the bowl-in-piston combus­
carbon and nitrogen interactions, thus affecting combustion and emis­ tion chamber. The original diesel injector, featuring eight nozzle holes
sion formation. Furthermore, the concurrent combustion of ammonia each with a diameter of 0.129 mm, remains in place, injecting diesel
and diesel introduces fuel nitrogen oxides (NOx) formation pathways, near the top dead center and thus initiating the co-combustion of diesel
which intertwine with thermal NOx (Xu et al., 2023). Ammonia intrin­ and ammonia. Table 1 details the engine specifications following its
sically possesses nitrogen oxides reducing (deNOx) properties, adding to adaptation to ammonia-diesel dual fuel combustion.
the complexity of NOx formation and consumption within the com­ Fig. 1 presents a detailed schematic representation of the engine test
bustion chamber (Niki et al., 2019). Another environmental implication bench. The engine operates on commercially available diesel fuel,

2
J. Liu and J. Liu Journal of Cleaner Production 434 (2024) 140274

Table 1 substitution rate for methanol-diesel dual-fuel engines. To be precise,

Engine specifications. the ASR is determined by the difference in cyclic diesel injection
Engine type In-line 4-cylinder, turbocharged intercooler amounts between pure diesel and dual-fuel modes, normalized by the
Displacement (L) 3.26 cyclic diesel injection amount in pure diesel mode. In subsequent dis­
Bore (mm) × Stroke (mm) 95 × 115 cussions, an effective replacement ratio of x is denoted as ASRx. In the
Connecting rod length (mm) 173
conducted experiments, an ammonia substitution ratio of up to 51% is
Geometric compression ratio (− ) 17.5
Combustion chamber Flat engine head and bowl-in-piston chamber achieved. Attaining a higher ASR in the engine presents challenges,
primarily attributed to the difficulties in vaporizing gaseous ammonia
from the cylinder bottle (as evidenced by the high vaporization latent
characterized by a sulfur content less than 10 ppm. The properties of the heat of ammonia shown in Table 2), where it is stored in liquid form
fuel utilized for the test are detailed in Table 2. During testing, the en­ under high pressures. Furthermore, the operation of the four-cylinder
gine is exposed to an ambient temperature of 27 ◦ C, faces an ambient engine at 1800 rpm necessitates a significantly higher ammonia flow
pressure of 101 kPa, operates with a coolant temperature maintained rate, thereby imposing additional constraints on increasing the ASR. It is
around 80 ± 2 ◦ C, and utilizes post-intercooling intake air kept around also important to mention that the focus of this study is on assessing the
48 ± 1 ◦ C. The dual-fuel engine is paired with an electrical dynamom­ combustion performance and emission characteristics of
eter, consistently operating at an engine speed of 1800 rpm and main­ ammonia-diesel dual fuel engines using well-characterized, industrial-­
taining a load of 200 Nm for all tests. Diesel injection pressure remains grade fuels. Both ammonia and diesel used in the experiments are
stable at 130 MPa, and the diesel injection timing is consistent at − 6◦ CA standardized fuels with well-known and consistent compositions. Given
ATDC. Testing commences in pure diesel combustion mode and then the standardized nature of these fuels, Fourier Transform Infrared (FTIR)
transitions to the dual-fuel combustion mode. As detailed in Table 2, spectroscopy analysis, typically applied in the analysis of complex and
diesel fuel is characterized by its long carbon chain, rendering it a highly varied biomass materials (Parthasarathy et al., 2013, 2017), is not
reactive fuel. This high reactivity is further evidenced by its notably low considered necessary for the scope of this study. The chemical properties
auto-ignition temperature and elevated cetane number. Conversely, of these fuels are widely recognized and documented in the industry,
ammonia, with its higher auto-ignition temperature, is categorized as a rendering additional FTIR analysis redundant for evaluating combustion
low reactivity fuel. Consequently, in the experimental setup, pure and emissions in a controlled experimental setting (Balasubramanian
gaseous ammonia is introduced into the engine via a low-pressure gas et al., 2022; Mufrodi et al., 2023). Consequently, the methodology
injector, positioned upstream of the intake valve. This ensures the for­ employed in this research aligns with standard practices in engine
mation of a homogeneous NH3-air mixture within the cylinder during combustion studies, focusing on the engine performance and emissions
the intake and compression strokes. Diesel fuel, on the other hand, is
delivered directly into the cylinder through a high-pressure liquid
Table 2
injector, strategically timed near the top dead center (TDC), to initiate
Properties of diesel and ammonia used in experimental tests.
the combustion process effectively. Specifically, in the dual-fuel setting,
Property Diesel Ammonia
as the diesel injection duration decreases, ammonia injection, facilitated
by a 0.5 MPa corrosion-resistant gas injection system, increases to Chemical formula C10~C21 hydrocarbon Pure NH3
maintain consistent output torque. As a result, the concept of the Density (kg/m3) 830~840 (@15 ◦ C) 680~690 (Liquid)
Vaporization latent heat (kJ/kg) 220~290 ~1369
ammonia substitution ratio (ASR) is introduced, representing the Auto-ignition temperature (◦ C) 260~300 ~630
effective replacement percentage of ammonia to diesel per cycle when Cetane number 48~52 –
compared to pure diesel mode under identical conditions, similarly Lower heating value (MJ/kg) 42.6–43.0 ~18.6
defined in the study conducted by Liu et al. (2022) for methanol

Fig. 1. Schematic representation of the ammonia/diesel dual fuel research engine.

3
J. Liu and J. Liu Journal of Cleaner Production 434 (2024) 140274

using established fuel types (Niki et al., 2019; Nadimi et al., 2023). Fig. 2 displays the cylinder pressure variations in the ammonia-diesel
Data collection begins once the engine has stably operated at the engine across ammonia substitution ratios from ASR0 to ~ASR50.
desired condition for a minimum of 3 min. The in-cylinder pressure trace During pure diesel operation, three distinct pressure peaks are evident,
is captured using a pressure transducer and is pegged to the manifold attributed to the delayed injection timing. Specifically, the initial pres­
intake pressure. Both the top dead center and crankshaft angular signals sure peak arises from the compression stroke. However, this peak is
are recorded using an encoder. Analysis of the combustion process, marginally lower than in the motoring condition because the injection of
based on in-cylinder pressure data, employs a single-zone heat-release liquid diesel before reaching top dead center decreases the cylinder
model utilizing the ideal gas law, operating under the assumption of temperature, leading to reduced cylinder pressure. At a medium load of
uniform pressure and temperature. Emission measurements encompass ~7.7 bar brake mean effective pressure (BMEP), a distinct premixed or
a range of pollutants including total hydrocarbons (THC), carbon rapid combustion phase emerges, characterized by the combustion of
monoxide (CO), nitric oxide (NO), NOx (a mixture of NO and nitrogen fuel that has mixed with air within the flammability limits during the
dioxide (NO2)), carbon dioxide (CO2), nitrous oxide (N2O), and ignition delay, leading to a second, albeit smaller, pressure peak within
ammonia (NH3). These measurements are conducted using an advanced just a few crank angle degrees. The third and most prominent pressure
FTIR spectroscopy exhaust gas analyzer, a technique chosen for its peak under the studied operating conditions results from the mixing-
comprehensive multi-component analysis capability. FTIR spectroscopy controlled combustion phase. Additionally, Fig. 2 illustrates how
is particularly adept at identifying multiple exhaust gas constituents due increasing the ASR shifts the cylinder pressure profile from a triple peak
to its ability to detect the unique infrared absorption features of different pattern to a dual peak one. This transition occurs because, as the
gases. This capability is crucial for our research as it enables the ammonia quantity increases and the diesel quantity decreases, the
simultaneous measurement of various exhaust components, providing a weight of the mixing-controlled combustion diminishes, leading to the
comprehensive emission profile of the dual-fuel engine. While Non- disappearance of the 3rd pressure peak. Furthermore, higher ASRs
Dispersive Infrared (NDIR) sensors are typically utilized for high- within the investigated range display a steeper pressure rise following
volume, single-component gas analysis due to their specific band-pass the compression peak, indicating a more dominant rapid combustion
filters that correspond to the absorption spectrum of particular gas phase.
molecules, FTIR offers a broader spectrum analysis. This broad- Fig. 3 depicts the impact of the ammonia substitution ratio on the
spectrum approach is essential for this research as it enables the apparent rate of heat release (AHRR). In the examined engine condi­
simultaneous measurement of various exhaust gas components, which is tions, the ammonia equivalence ratio is notably low, making it incapable
vital for understanding the complete emission profile of the dual-fuel of facilitating ammonia to initiate a flame propagation process. Instead,
engine. The versatility of FTIR analysis is particularly beneficial for ammonia oxidation is primarily concentrated around the diesel flame
this study, as the dual-fuel combustion process can result in a complex plumes, which result from diesel combustion within the ammonia-air
mixture of gases. Furthermore, the accuracy of emissions measurements mixture. Accordingly, to understand the changes introduced by
is ensured by conducting careful calibration against known standards for ammonia replacing part of the diesel, it is imperative to first describe the
all exhaust components. In addition to the multi-gas analysis, soot pure diesel combustion process. In Fig. 3, the rate-of-heat-release curve
emissions from the dual-fuel engine exhaust are evaluated using a smoke for pure diesel operation reveals minimal heat release until the end of
meter. This device employs the filter paper method, a widely recognized compression, at which point an obvious heat loss is observed during the
technique, to ascertain the filter smoke number (FSN), which quantifies ignition delay period—defined as the time between the start of injection
the soot emission levels. The combination of FTIR spectroscopy for gas (SOI) and the start of combustion (SOC)—attributed to wall heat
analysis and the smoke meter for particulate matter provides a holistic transfer, fuel vaporization, and heating. During the combustion process
assessment of the engine emission characteristics. Table 3 presents the of pure diesel, burning evolves through three distinct stages. The first
measurement accuracy and experimental uncertainty associated with involves the combustion of a premixed fuel-air mixture formed during
the main instruments used in the study. the ignition delay, resulting in high heat-release rates typical of this
phase. Following the consumption of this premixed fuel and air, the
3. Results and discussion second stage emerges as the mixing-controlled combustion phase.
Although influenced by processes like liquid fuel atomization, vapor­
This section delves into the effects of the ammonia substitution ratio ization, fuel vapor mixing with air, and preflame chemical reactions, this
on the performance, combustion, and emission characteristics of the burning rate of this phase is predominantly determined by the fuel
ammonia-diesel dual-fuel engine, grounded in experimental analysis. vapor-air mixing process. Consequently, this stage sees a progressively
Each condition is evaluated based on 200 consecutively collected cycles. declining heat-release rate that initially peaks again during the primary
If present, error bars in the subsequent figures denote the 95% confi­
dence interval, which encompasses two standard deviations.

Table 3
Accuracy and uncertainty of key measured parameters.
Instrument Measured parameters Accuracy Uncertainty (%)

Electric dynamometer Speed 0.03% –

Torque 0.2% –
Fuel consumption Diesel consumption 0.12% 0.6
meter Ammonia 0.5% 0.5
consumption
Pressure transducer Cylinder pressure 0.1 bar 0.2
Exhaust gas analyzer THC 2 ppm 1.2
CO 2 ppm 1.2
NOx 2 ppm 1.8
CO2 0.05 vol% 1.0
NO, NO2 1 ppm 1.5
N2O 1 ppm 2.0
NH3 2 ppm 2.0
Fig. 2. Cylinder pressure responses to varying ammonia substitution at con­
Smoke meter Soot 0.001 FSN 0.4
stant engine speed and load.

4
J. Liu and J. Liu Journal of Cleaner Production 434 (2024) 140274

Fig. 3. Apparent heat release rate responses to varying ammonia substitution at

constant engine speed and load.

heat-release period, with the majority of the fuel energy released in

these first two stages. The final stage is marked by the tapering end of the
heat-release diagram, characterized by a modest yet discernible heat
release persisting for a significant portion of the expansion stroke. This
lingering combustion stems from the oxidation of unburned fuel frac­
tions and energy trapped in soot and fuel-rich by-products, coupled with Fig. 4. Peak cylinder pressure and maximum pressure rise rate responses to
varying ammonia substitution at constant engine speed and load.
slowing burnout kinetics as cylinder gas temperatures diminish during
expansion.
Regarding the ammonia-diesel dual-fuel operation, Fig. 3 illustrates et al.‘s research (Sun et al., 2023). The reduction in peak cylinder
that the substitution of diesel with ammonia results in an increased pressure can be attributed to lower combustion temperature of
dominance of the premixed combustion phase, concurrently reducing ammonia. Additionally, the acceptable variation in peak cylinder pres­
the significance of the mixing-controlled combustion phase. This sure suggests a stable combustion process, as this metric encompasses
enhanced dominance of premixed combustion can be attributed to a changes in both the ignition and primary combustion phases. Even with
prolonged ignition delay, as detailed in Fig. 9 (which will be discussed in a 51% replacement of diesel by ammonia under the tested conditions,
a subsequent section), and the incorporation of NH3 into the premixed the peak pressure variation remains within acceptable bounds, under­
combustion process. Specifically, a lengthier ignition delay enables a scoring the viability of ammonia-diesel dual fuel engines. As expected,
greater portion of the liquid diesel to vaporize, thereby elevating the with higher ammonia substitution ratios, the premixed combustion
energy released during the premixed combustion phase. Furthermore, in phase, which corresponds to the rapid cylinder pressure rise stage, be­
the dual-fuel operation mode, the premixed combustion process, comes more pronounced, resulting in an accelerated pressure rise rate
encompassing both diesel and ammonia combustion, experiences (PRR). An elevated pressure rise rate can result in enhanced engine noise
amplified heat release due to the introduced ammonia energy. The and vibrations. Additionally, a sudden surge in pressure may exert
decrease in the mixing-controlled combustion phase stems from a increased mechanical stress on engine components, including the cyl­
reduced diesel injection quantity, which shortens the injection duration, inder head, cylinder walls, pistons, and connecting rods. The accom­
leading to a diminished prominence of this combustion process. During panying rapid temperature increase further introduces thermal stress,
the concurrent combustion of ammonia with diesel flame plumes, a potentially culminating in component degradation over time. As a
decrease in the energy release from the mixing-controlled combustion consequence, the maximum pressure rise rate (PRRmax) is typically
phase is observed, as evidenced in Fig. 3. This decrease is primarily capped at 10 bar/◦ CA to ensure reliable and steady operation. Notably,
attributable to the enhanced energy output from the premixed com­ under the examined condition with an ammonia substitution ratio of
bustion process, coupled with an observed increase in inefficient com­ 51%, the maximum PRR remains below this critical threshold, under­
bustion. The latter aspect, highlighting a rise in inefficient combustion scoring the safety and feasibility of replacing at least 50% of diesel with
scenarios, is comprehensively depicted in Fig. 6 and will be further ammonia at medium engine loads.
discussed in a later section of the paper. Fig. 3 also reveals that with the Fig. 5 illustrates the impact of the ammonia substitution ratio on
rise in ASR, an enhanced premixed combustion phase and diminished both the brake thermal efficiency (ηth) and operational financial cost.
mixing-controlled combustion phase produce a more pronounced first The thermal efficiency for pure diesel operation appears slightly lower
AHRR peak and a reduced second one, transitioning from a dominant than that of commercial engines. This discrepancy can be partially
2nd to a primary 1st AHRR peak as the ammonia substitution ratio shifts ascribed to the delayed injection timing resulting in phase losses. This
from ASR16 to ASR26. Further elevation of the ammonia substitution delay in injection timing is intentionally implemented during the tests to
ratio from ASR34 to ASR43 results in the disappearance of the 2nd prevent exceedingly high pressure rise rates, ensuring engine protection.
AHRR peak in the mixing-controlled combustion phase. Intriguingly, for Fig. 5 reveals that substituting 7% of diesel with ammonia leads to a
the conditions at ASR43 and ASR51 that showcase a singular heat marked decrease in thermal efficiency compared to operations using
release peak, there is an alignment with the dual peak cylinder pressure pure diesel. This reduction in efficiency can potentially be attributed to a
pattern observed in Fig. 2. In this specific investigation, the absence of decreased combustion efficiency, as detailed in Fig. 6, which will be
the 2nd AHRR peak correlates with the disappearance of the 3rd cyl­ elaborated upon later in the text. Intriguingly, as the ammonia substi­
inder pressure peak. tution ratio is increased from ASR7 to ASR43, there is a discernible
In Fig. 4, the peak cylinder pressure (PCP), corresponding to the improvement in thermal efficiency. This trend aligns with the findings
mixing-controlled combustion phase, diminishes with increasing reported by Nadimi et al. (2023). This behavior can be explained by a
ammonia substitution ratios, a trend consistent with findings from Sun combination of factors. For instance, an intake mixture with a higher

5
J. Liu and J. Liu Journal of Cleaner Production 434 (2024) 140274

Fig. 5. Brake thermal efficiency and operational financial cost responses to Fig. 7. Peak pressure location and energy release centroid responses to varying
varying ammonia substitution at constant engine speed and load. ammonia substitution at constant engine speed and load.

thermal efficiency, thereby establishing a critical inflection point under

the examined conditions. This observed trend aligns with the findings
presented by Jin et al. (2023). Their research indicates a complex tra­
jectory wherein the thermal efficiency initially experiences a decrease,
followed by an increase, and ultimately undergoes a reduction beyond
the established turning point. The brake thermal efficiency and brake
specific energy consumption (BSEC) both indicate fuel economy, but due
to the inverse relationship between the two metrics, BSEC is not dis­
cussed further here. Assessing the financial cost of power output reveals
interesting comparisons between pure diesel operation and the
ammonia dual-fuel mode. Drawing from fuel price data in Nadimi et al.‘s
study (Nadimi et al., 2023), where diesel costs 1.65 ECU/kg and
ammonia is priced at 0.43 ECU/kg, it is evident that the operational cost
decreases as the ammonia substitution ratio increases. This decrease is
attributable to the higher price of diesel compared to ammonia, and
because using a higher ammonia ratio reduces the amount of diesel
needed for the same power output. It is also vital to note that the
apparent cost advantage of the dual-fuel mode over the diesel-only mode
here is based solely on fuel costs, without accounting for other costs like
aftertreatment and maintenance.
In the context of ammonia-diesel co-combustion within the cylinder,
both exhaust gas temperature (EGT) and combustion efficiency (ηcomb)
are indicative of combustion quality. As depicted in Fig. 6, when using
Fig. 6. Exhaust gas temperature and combustion efficiency responses to vary­ ammonia to substitute for 7% of diesel energy, there is a minor increase
ing ammonia substitution at constant engine speed and load.
in the exhaust gas temperature in comparison to pure diesel combustion.
This uptick can be attributed to the augmented weight of tail-end
ammonia concentration has a lower specific heat ratio, leading to a combustion, pointing to decreased combustion efficiency. Neverthe­
slight decrease in cylinder pressure during the compression stroke and less, with the ammonia substitution ratio ascending from ASR7 to
thus reducing the counteractive work against the piston. Moreover, ASR51, a notable downtrend in the exhaust gas temperature is observed,
increased levels of ammonia substitution are associated with diminished which is in alignment with the observations made by Nadimi et al.
heat losses to cooler boundary surfaces, attributable to a reduction in the (2023). This reduction in temperature is presumably attributed to the
average cylinder temperature. Furthermore, an elevated ASR pre­ lower combustion temperatures characteristic of diesel-ammonia
cipitates an advancement in the peak cylinder pressure, a phenomenon co-combustion, coupled with a concurrent diminution in the quality of
elucidated in Fig. 7 (to be discussed in further detail subsequently). This combustion. Furthermore, Fig. 6 illustrates a consistent decline in
advancement contributes to a decrease in phase losses and augments the combustion efficiency as the ammonia substitution ratio rises. This trend
efficiency of work performed during the piston downstroke. It is also may be ascribed to the diminishing weight of the mixing-controlled
worth mentioning that a decline in ammonia combustion efficiency combustion phase, affecting the spatial dispersal of diesel flame
hampers the enhancement of thermal efficiency, as it diminishes the plumes post interaction with the chamber wall. The dip in combustion
positive work on the piston. Fig. 5 further elucidates that an increase in efficiency for dual fuel operations implies the release of a substantial
the ammonia substitution ratio to ASR51 marks a discernible decline in volume of unburnt combustion byproducts via the exhaust port. This is

6
J. Liu and J. Liu Journal of Cleaner Production 434 (2024) 140274

not favorable for enhancing thermal efficiency or minimizing emissions.

Given that ammonia oxidation in the dual fuel mode predominantly
occurs within the diesel spray coverage area, fine-tuning the injection
strategy and injector design might be viable avenues to boost diesel
distribution, optimize the use of the ammonia-air blend, and thereby
elevate combustion efficiency.
Examining the impact of the ammonia substitution ratio on com­
bustion phasing provides insights into the transition of the combustion
process from pure diesel operation to ammonia-diesel dual fuel opera­
tion. As depicted in Fig. 7, the location of peak pressure (LPP) advances
with an increasing ammonia substitution ratio. This shift can be ascribed
to the heightened prominence of the premixed combustion phase.
Furthermore, the acceptable variation in the peak pressure location in­
dicates stable ignition and combustion processes in both pure diesel and
ammonia-diesel dual fuel operations. In this research, the CAx is
designated as the crank angle corresponding to x% of cumulative heat
release. The CA50, acting as the combustion centroid, stands as a crucial
phasing parameter due to its association with efficiency and emissions.
From the ASR0 to ASR26 conditions, the CA50 consistently advances, a
result of the increased weighting of the premixed combustion phase. As
delineated earlier, the augmentation in the level of premixing can be
traced back to an extended ignition delay, a subject to be expounded
upon in Fig. 9, and the incorporation of ammonia oxidation processes. Fig. 9. Ignition delay and combustion duration responses to varying ammonia
Nevertheless, with the escalation of the ammonia substitution ratio from substitution at constant engine speed and load.
ASR34 to ASR51, there is a noticeable retardation in the CA50. This
phenomenon is likely a consequence of a pronounced delay in the the underlying cause for both phenomena is identical.
commencement of combustion, as depicted in Fig. 8 and to be discussed The ignition delay in diesel engines stems from both physical and
in greater detail subsequently, along with a reduction in the mixing- chemical processes. The physical process encompasses fuel atomization,
controlled combustion phase, culminating in a decelerated NH3 oxida­ droplet vaporization, and the blending of fuel vapor with air. On the
tion rate. chemical side, pre-combustion reactions occur between the fuel, air, and
In this study, the start of combustion is determined using a rate of residual gases, culminating in autoignition. As illustrated in Fig. 9, an
heat release analysis, identifying the crank angle at which the AHRR increase in the ammonia substitution ratio leads to a prolonged ignition
shifts from negative to positive, immediately succeeding the initial delay. In the diesel ignition process, the pre-ignition phase is signifi­
negative AHRR phase post fuel injection. As illustrated in Fig. 8, an in­ cantly influenced by a peroxide-based chemical mechanism. Initially,
crease in the ammonia substitution ratio under the ammonia-diesel this mechanism involves a cascade of reactions leading to the formation
combustion mode delays the start of combustion. This retardation can of ketohydroperoxide molecules. These reactions start with the
be ascribed to the effects of ammonia addition on the cold flame kinetics abstraction of hydrogen atoms from the long-chain hydrocarbons in
of diesel. The end of combustion (EOC) is defined by the CA95, which diesel fuel, a process facilitated by oxygen, hydroxyl, and hydroperoxy
denotes the crank angle corresponding to 95% of heat release. Notably, radicals. The predominant reaction at this stage involves oxygen addi­
the trend of EOC mirrors that of CA50 as depicted in Fig. 7, implying that tion to the resultant alkyl radicals, forming peroxy radicals, as β-scission
reactions are slow at low temperatures. Subsequently, isomerization of
the alkylperoxy radical forms the hydroperoxy-alkyl radical. This is
followed by further oxygen addition, resulting in a transition-ring
complex that isomerizes and decomposes to produce ketohydroper­
oxide and the hydroxyl radical. The decomposition of ketohydroper­
oxide yields carbonyl, aldehyde, and generates additional OH radicals,
contributing to the key chain branching characteristic of low-
temperature chemistry. However, as temperature increases, oxygen
addition reactions shift towards decomposition, thereby reducing the
extent of low-temperature chain branching. Concurrently, at higher,
intermediate temperatures, the dissociation of low-temperature in­
termediates and the alkyl radical through β-scission becomes increas­
ingly significant, leading to the formation of various products, including
alkenes, HO2, and other alkyl radicals. These compete with the low-
temperature chain branching pathway. This backward shift in the low-
temperature reaction channel, combined with the increased preference
for β-scission at intermediate temperatures, results in a globally reduced
chemical reactivity at moderate temperatures. As the temperature con­
tinues to rise, hydrogen peroxide plays a more dominant role in con­
trolling the reaction kinetics. Its decomposition into a pool of OH
radicals provides the necessary agents for runaway oxidation of the
remaining hydrocarbon species. Once the temperature reaches a
threshold where H2O2 decomposition outweighs its production, a burst
of OH radicals significantly accelerates the oxidation rate, marking the
Fig. 8. Start of combustion and end of combustion responses to varying transition to high-temperature combustion. The introduction of
ammonia substitution at constant engine speed and load.

7
J. Liu and J. Liu Journal of Cleaner Production 434 (2024) 140274

ammonia into this process competes with diesel low-temperature

oxidation for OH radicals, leading to a slower auto-ignition and an
extended ignition delay (Ou et al., 2023). Besides kinetic effects,
ammonia introduces thermal effects due to its high heat capacity,
absorbing heat and cooling the ignition environment, which slows down
temperature-dependent reactions. Concurrently, the dilution effects of
ammonia, acting as an inert diluent, decrease the concentration of
reactive species during the auto-ignition process, further contributing to
the extended ignition delay. This elongated ignition in the
ammonia-diesel dual fuel operation mode yields a more premixed
mixture during this phase, resulting in both a more pronounced pre­
mixed combustion weight and a steeper pressure rise rate. Interestingly,
at ammonia substitution ratios below 16%, the ignition delay demon­
strates minimal sensitivity to ammonia addition, with only a slight
extension. However, once the ratio surpasses ASR16, this extending ef­
fect becomes markedly significant. Meanwhile, the combustion duration
(spanning from SOC to EOC) depicted in Fig. 9 does not exhibit a robust
correlation with the ammonia substitution ratio. This implies that var­
iations in the ammonia substitution ratio have a nonlinear and complex
impact on the combustion process in the ammonia-diesel operation
mode.
In the realm of emissions, the formation and conversion of NOx
within the combustion cylinder become increasingly complex due to the
nitrogen element in ammonia, which activates the nitrogen reactivity Fig. 10. NO/NOx in molar fraction and EINOx emissions responses to varying
through introducing nitrogen-containing intermediates. In the com­ ammonia substitution at constant engine speed and load.
bustion of conventional fuels, NOx formation is predominantly governed
by the Zel’dovich mechanism (Zeldvich, 1946), also known as thermal and combustion temperatures. When ammonia content reaches high
NOx, resulting from the high-temperature reaction between nitrogen levels, fuel-borne NO begins to dominate NOx production. However, the
and oxygen. In the context of ammonia combustion chemistry, NO and deNOx chemistry, driven by reactions such as NH2 + NO ↔ N2 + H2O
NO2, act as intermediate reactants in the ammonia oxidation pathway and NH2 + NO ↔ NNH + OH, becomes prominent within the
leading to nitrogen and water. Another significant intermediate product 1100–1400K range. An increase in NH3 combustion in the air-fuel
is nitrous oxide, which adds further complexity. N2O often forms under mixture accelerates this deNOx process.
conditions of ammonia partial oxidation. Moreover, ammonia is known In the pure diesel operation mode, both CO and THC emissions are
for its deNOx effects, referring to the reduction of nitrogen oxides. kept at minimal levels because diesel engines consistently operate under
Specifically, ammonia reacts with NO and NO2 to form nitrogen gas and lean conditions. Fig. 11 indicates that transitioning from a pure diesel
water, thereby reducing the concentration of NOx. Additionally, amino mode to an ammonia-diesel dual fuel mode elevates the CO and THC
groups within the molecular structure of ammonia contribute to the emissions. Increasing the ammonia substitution ratio in the fuel mixture
complexity of NOx kinetics. These groups interact with NOx species in leads to higher emissions of CO and THC, which are indicators of
various ways, influencing both the formation and decomposition rates. incomplete combustion. This trend implies a reduction in diesel com­
The presence of amino groups can lead to alternative reaction pathways, bustion efficiency, considering that carbon is solely derived from diesel
potentially more efficient in reducing NOx emissions under certain
conditions. Overall, these diverse effects introduce a significant level of
complexity to NOx kinetics. Fig. 10 illustrates that NO remains the
predominant contributor to NOx emissions, accounting for over 90% of
the concentration. This observation aligns with the understanding that
NO2, particularly at higher temperatures, can function as an oxidizing
agent. Furthermore, the data does not indicate a strong correlation be­
tween the ammonia substitution ratio and the NO/NOx distribution.
Fig. 10 also delineates the relationship between the ammonia substitu­
tion ratio and the emission index of NOx species (EINOx). In this context,
NO undergoes conversion to NO2 first, and then the mass of this trans­
formed NO2 is aggregated with the extant NO2. It is evident from the
data that employing a dual-fuel approach with ammonia and diesel re­
duces NOx emissions compared to using pure diesel. Elevating the
ammonia substitution ratio further accentuates this decrement, a
deduction that finds resonance in the research conducted by Nadimi
et al. (2023). It is noteworthy that the rates and focal regions of NOx
production and consumption in dual-fuel operations deviate from those
of pure diesel operations. Such dual-fuel modes yield reduced engine-out
NOx emissions, a result stemming from two main factors. First, the
combustion temperature for ammonia is inherently lower, leading to
diminished thermal NOx production. Second, the amino groups formed
during the ammonia reaction possess denitrification properties,
enabling them to react with NOx and ultimately produce nitrogen. More
specifically, the decrease of NOx at lower ASRs can be credited to the Fig. 11. CO and THC emissions responses to varying ammonia substitution at
suppression of thermal NOx production due to reduced diesel content constant engine speed and load.

8
J. Liu and J. Liu Journal of Cleaner Production 434 (2024) 140274

in the combustion process. Substituting a larger proportion of diesel

with ammonia could theoretically lead to cleaner combustion, as it re­
duces the quantity of diesel combusted in each cycle. However, this
approach presents a complication. Ammonia competes with small hy­
drocarbon radicals and molecules generated during the high-
temperature cracking of diesel for the essential OH radicals needed for
combustion. This competition diverts some OH radicals for ammonia
oxidation, increasing the level of incomplete combustion of diesel. In
detail, the co-combustion of ammonia and diesel occurs at the diffusion
flame boundary, a part of the high-temperature oxidation process. Diesel
initially cracks into small C1–C4 hydrocarbons, which then undergo
mixed combustion with ammonia. During this high-temperature reac­
tion, ammonia and small hydrocarbon molecules combust indepen­
dently. Their interaction and mutual influence primarily involve
competition for free radicals. In the presence of ammonia, the complete
combustion level of small hydrocarbon radicals is reduced, as ammonia
oxidation competes for the OH radicals needed for hydrocarbon oxida­
tion. Consequently, CO and THC concentrations in the exhaust gases
increase for ammonia-diesel dual fuel combustion compared to pure
diesel operation.
Soot emissions from engines, detrimental to both human health and
the environment, necessitate stringent controls, which is why experi­
mental tests are conducted to measure engine-out soot emissions. In
diesel engines, fuel undergoes various phase transitions and chemical Fig. 12. Soot and unburned ammonia emissions responses to varying ammonia
reactions, starting from atomization in the cylinder to the eventual substitution at constant engine speed and load.
formation of particulates. Initially, large hydrocarbon fuel molecules
crack into smaller hydrocarbon radicals. Under conditions favorable for ammonia flame quenching, making it difficult to maintain a stable
soot formation, such as in rich combustion scenarios, these small mol­ ammonia flame propagation. Consequently, unburned NH3 likely orig­
ecules can polymerize to form polycyclic aromatic hydrocarbons inates from piston crevices and incomplete combustion processes. Spe­
(PAHs). The PAHs then collide and accumulate, forming agglomerates cifically, the unburned NH3 here is considered to produce from the
and evolving into solid particles. Nucleated soot particles grow through piston crevice and incomplete combustion. Fig. 12 reveals a notable
surface growth and collision coalescence, leading to larger particle increase in unburned ammonia emissions as the ammonia substitution
formation. Introducing ammonia as a substitute for a portion of diesel in ratio rises. Several factors contribute to this trend. A higher ammonia
engine operation complicates the soot formation mechanism. Nitrogen- substitution ratio enhances the ammonia mole fraction in the ammonia-
containing intermediates from NH3 oxidation, such as NOx and NHx, air mix, leading to more ammonia being trapped in crevices and
participate in the hydrocarbon high-temperature reactions, leading to bypassing the primary combustion. In parallel, with a reduced amount
new component formation. This C–N interaction results in the devel­ of diesel, the mixing-controlled combustion process is lessened, leading
opment of CN-based components and nitrogen-containing aromatic to expanded inaccessible regions of the diesel flame within the chamber.
compounds. Ammonia also shares the pool of reactive free radicals with Concurrently, as the ammonia substitution ratio increases, the higher
small hydrocarbon molecules. By altering the concentration of active ammonia mole fraction in these regions results in a greater quantity of
radicals, including OH and H, ammonia changes the dynamics between unburned ammonia. Fig. 12 also indicates an alarmingly high concen­
various aromatic formation and growth pathways. Theoretically, using tration of unburned ammonia in the exhaust gas, necessitating control
an ammonia-diesel dual fuel mixture could reduce soot formation during measures. Consequently, an augmented diesel diffusion combustion
combustion. This reduction is partly attributed to substituting diesel might be required to bolster ammonia combustion, aiming for superior
with ammonia, decreasing the availability of carbon elements for soot combustion efficiency. Furthermore, an ammonia slip catalyst (ASC)
formation. Additionally, the elevated level of premixed combustion in might be needed to mitigate the emission of slipped NH3 from the
this dual fuel operation helps mitigate local diesel-rich conditions, engine.
which typically encourage soot formation. The interaction between One distinct characteristic of ammonia-diesel co-combustion is the
carbon and nitrogen components from this fuel mixture diversifies the production of N2O, which is rarely observed in pure diesel operations. It
carbon-containing compounds present, impeding their aggregation into is crucial to note that, despite its lower emission levels compared to NOx
soot precursors like unsaturated hydrocarbons and PAHs. However, and unburned NH3, N2O cannot be overlooked due to its significant
ammonia-diesel dual fuel operations can impair soot oxidation capacity greenhouse effect. This effect can considerably undermine the overall
due to lower combustion temperatures. The introduction of ammonia decarbonization benefits. In chemical kinetics, N2O forms from the
decreases oxygen concentration in the combustion chamber, diminish­ reduction reactions of NO and NO2, such as NH + NO ↔ N2O + H,
ing its oxidative impact on soot particles. Crucially, the oxidation of especially at temperatures above 1400K. Its consumption paths could
ammonia itself competes with soot particle oxidation for reactive free include N2O + H ↔ N2 + OH and N2O + OH ↔ N2 + H2O, which become
radicals and energy, affecting the efficiency of soot oxidation. In sum­ dominant when the ambient temperature exceeds 1650K (Mi et al.,
mary, the addition of ammonia in diesel combustion acts as a trade-off: it 2023). In ammonia-diesel dual fuel operations, N2O is generated both
inhibits the formation of soot while simultaneously reducing the during ammonia combustion and the subsequent oxidation of unburned
oxidation of soot, presenting a complex balancing act in soot control. NH3 during the expansion stroke. Fig. 13 indicates that a higher ASR
Fig. 12 illustrates a decline in soot emissions with an increased ammonia leads to increased N2O emissions, implying that more ammonia is being
substitution ratio, suggesting that the benefits of reduced soot formation partially oxidized to N2O, a precursor to nitrogen gas. Intriguingly, a
surpass the drawbacks of diminished soot oxidation in the aforemen­ significant decrease in N2O emission levels is observed when the
tioned trade-off. In ammonia-diesel dual fuel operations, a rise in the ammonia substitution rate is increased from 34% to 43%. This can be
level of incomplete combustion is primarily attributed to decreased attributed to the decreased spatial distribution of diesel, potentially
ammonia combustion efficiency. Overly lean mixtures exacerbate

9
J. Liu and J. Liu Journal of Cleaner Production 434 (2024) 140274

Fig. 13. N2O and CO2 equivalent emissions responses to varying ammonia
substitution at constant engine speed and load. Fig. 14. COVIEMP and COVPCP responses to varying ammonia substitution at
constant engine speed and load.
reducing the contact surface area between small species derived from
diesel and the ammonia-air mixture. Such reduced contact is not condition, the COVIMEP remains below 3%, a threshold typically deemed
conducive to N2O formation, even if the formation rate is efficient per stable for operation. In Fig. 14, the COV of peak cylinder pressure (PCP)
unit of contact area. Since N2O is an incomplete combustion product of complements the information provided by COVIMEP. This cooperative
ammonia, enhancing ammonia combustion efficiency also serves as a data presentation is crucial, as fluctuations in peak cylinder pressure can
strategy to curtail N2O emissions. Nitrous oxide has a global warming indicate variations in auto-ignition timing and the subsequent main
potential (GWP) that is significantly higher than that of CO2. Over a combustion phase. Additionally, CCV in peak cylinder pressure in­
century, the GWP of N2O is about 298 times that of CO2. With increased fluences the engine noise, vibration, and harshness (NVH) attributes
ammonia substitution ratios in diesel engines, less diesel is consumed, (Dimitrakopoulos et al., 2017). Furthermore, erratic combustion events
leading to reduced CO2 emissions. However, the non-negligible levels of and atypical peak pressures can induce mechanical strains on engine
N2O emissions complicate the evaluation. The CO2-equivalent metric, components, potentially accelerating wear and curtailing engine
which equates emissions from various greenhouse gases based on their longevity. Interestingly, the correlation between the ammonia substi­
global warming potential, is used to place these gases on a common scale tution ratio and COVPCP in Fig. 14 adopts a hump-like profile, but its
for easier comparison. Fig. 13 illustrates that the CO2-equivalent emis­ peak value remains within permissible bounds. Transitioning from a
sions from ammonia-diesel dual-fuel engines are higher than those from purely diesel mode to an ammonia-diesel dual-fuel operation augments
pure diesel, undermining the goal of decarbonization and exacerbating the variation in peak cylinder pressure. This change is attributed mainly
the greenhouse effect. Although operating conditions with ammonia to greater inconsistencies in the auto-ignition phase. The diminishing
substitution ratios of ASR43 and ASR51 produce lower CO2-equivalent COVPCP, despite a rising COVIMEP when the ammonia substitution ratio
emissions than pure diesel, their combustion efficiencies are suboptimal, exceeds 26%, underscores the idea that variations in the subsequent
and their unburned NH3 emissions exceed acceptable limits. Before main combustion phase primarily dictate engine stability. Most
conducting detailed studies, there are expectations that NOx emissions crucially, the viability of transitioning from traditional diesel engines to
might be a significant challenge for ammonia-diesel dual-fuel engines ammonia-diesel dual-fuel operations is underscored by the tolerable
due to fuel-based NOx formation, but the findings here highlight range of cycle-to-cycle variation, which also suggests potential CO2
elevated unburned NH3 and N2O emissions as the primary concerns. emission reductions by substituting diesel with ammonia.
Given that these emissions are intrinsic to ammonia-diesel dual-fuel
engines, and acknowledging that selective catalytic reduction (SCR) for 4. Summary and conclusions
NOx reduction can also result in some N2O emissions, it is imperative to
consider integrating an additional aftertreatment device specifically This study rigorously investigates the performance characteristics of
designed to capture N2O. The environmental advantages of ammonia-diesel dual fuel engine operations, a subject that has not been
ammonia-diesel dual-fuel engines would be considerably enhanced if extensively explored in the current body of research. The primary
nitrous oxide emissions could be effectively mitigated. objective is to enhance the understanding of this operation mode, which
Cycle-to-cycle variation (CCV) is pivotal in diesel engines. It in­ is crucial for leveraging ammonia, a zero-carbon fuel, in reducing the
fluences engine performance, fuel efficiency, emissions, durability, and carbon footprint associated with engine emissions. Methodically,
user comfort, either directly or indirectly. Efficient control of CCV is experimental tests are conducted under consistent engine speed and
instrumental in optimizing engine operations and prolonging its life­ load conditions, maintaining uniform injection timing. Throughout
span. The coefficient of variation (COV) of the indicated mean effective these experiments, the proportion of ammonia substitution ratio is
pressure (IMEP) offers a snapshot of this variation across a given set of progressively increased from 0% to ~50%, concomitant with a decrease
cycles, serving as an index recording the entire operational history of in diesel quantity per cycle achieved by reducing the injection duration.
each fired cycle. In Fig. 14, a clear trend emerges with the rise in the The key insights derived from the analysis of the experimental data are
ammonia substitution ratio leading to an increase in the COV of IMEP, as follows:
suggesting heightened combustion instability. Yet, even at the ASR51

10
J. Liu and J. Liu Journal of Cleaner Production 434 (2024) 140274

(1) In the transition from mono-fuel diesel to ammonia-diesel dual diesel dual fuel operations. Second, the integration of after­
fuel operation, it is observed that the combustion mechanism treatment systems designed to capture unburned ammonia and
predominantly involves premixed combustion, subsequently fol­ nitrous oxide emissions emerges as a pivotal strategy, potentially
lowed by the mixing-controlled combustion phase. This pattern is playing a significant role in substantially reducing the carbon
attributed to the lean nature of the ammonia-air mixtures, which footprint associated with these dual fuel engines.
can lead to flame quenching in ammonia and challenges in
maintaining stable ammonia flame propagation. As the ammonia CRediT authorship contribution statement
substitution ratio increases, there is a discernible shift in com­
bustion dynamics; the prominence of premixed combustion is Junheng Liu: Conceptualization, Formal analysis, Funding acquisi­
enhanced, while the relevance of the mixing-controlled com­ tion, Methodology, Project administration, Validation, Writing – orig­
bustion phase decreases. inal draft. Jinlong Liu: Conceptualization, Formal analysis, Funding
(2) Within the realm of ammonia-diesel dual fuel operation, it is acquisition, Methodology, Project administration, Validation, Writing –
noted that the ignition delay period is extended. A higher review & editing.
ammonia substitution ratio further exacerbates this extension of
ignition delay. Such prolongation in ignition delay results in a
more substantial portion of the fuel energy being released during Declaration of competing interest
the premixed combustion phase. This phenomenon subsequently
culminates in an earlier attainment of peak pressure and an The authors declare that they have no known competing financial
increased rate of pressure rise. The ammonia-diesel dual fuel interests or personal relationships that could have appeared to influence
configuration shows potential for enhanced thermal efficiency. the work reported in this paper.
Nonetheless, it is observed that an increase in ammonia substi­
tution correlates with a marked reduction in combustion effi­ Data availability
ciency, which, in turn, impedes gains in thermal efficiency and
contributes to a decrease in exhaust gas temperature. The Data will be made available on request.
reduced combustion efficiency observed in ammonia-diesel dual
fuel operation is primarily attributed to the decreased combus­ Acknowledgments
tion efficiency of ammonia, even though this mode also leads to
diminished diesel combustion efficiency. This research has received support from the National Natural Science
(3) Replacement of diesel with ammonia in dual fuel operations Foundation of China under Grant No. 51806086, with Prof. Dr. Junheng
demonstrates a reduction in carbon dioxide and soot emissions. Liu as the beneficiary. Additionally, it is backed by the National Natural
However, an escalation in the ammonia substitution ratio is Science Foundation of China (Grant No. 52306169), the Zhejiang Pro­
observed to correlate with an increase in CO and THC emissions. vincial Natural Science Foundation of China (Grant No. LQ23E060005),
This trend is attributed to the competitive interaction between the State Key Laboratory of Engines at Tianjin University (Grant No
ammonia and the small species produced from diesel cracking for K2022-05), and the State Key Laboratory of Intelligent Agricultural
oxidation by OH radicals. Additionally, the dual fuel strategy Power Equipment at Luoyang Tractor Research Institute Co., Ltd. (Grant
involving ammonia and diesel results in lower NOx emissions, a No. SKLIAPE2023002), with Prof. Dr. Jinlong Liu acknowledged as the
consequence of the combined effects of reduced combustion grant recipient. The authors also gratefully acknowledge the Zhejiang
temperatures and the NOx reducing properties of amino groups. Key Laboratory of Clean Energy and Carbon Neutrality (The Baima Lake
With the increase in ammonia substitution ratio, there is a Laboratory) and the Jiaxing Research Institute of Zhejiang University for
notable expansion of regions within the combustion chamber that their support and assistance.
remain beyond the reach of the diesel flame. Simultaneously, a
greater proportion of ammonia is present in the engine crevices References
and these flame-inaccessible areas, owing to its increased con­
centration in the premixed fuel mixture. This scenario results in Alrwashdeh, S.S., Ala’a, M., Murtadha, T.K., 2022. Effect of turbocharger compression
an elevated level of unburned ammonia emissions being released ratio on performance of the spark-ignition internal combustion engine. Emerging
Science Journal 6 (3), 482–492.
into the exhaust. Furthermore, the substantial portion of Balasubramanian, R., Abishek, A., Gobinath, S., Jaivignesh, K., 2022. Alternative fuel:
ammonia undergoing partial oxidation during the later stages of hydrogen and its thermodynamic behaviour. Journal of Human, Earth, and Future 3
the expansion stroke contributes to heightened N2O emissions. (2), 195–203.
Belgiorno, G., Di Blasio, G., Beatrice, C., 2019. Advances of the natural gas/diesel RCCI
These elevated levels of N2O are of concern due to their potential concept application for light-duty engines: comprehensive analysis of the influence
to exacerbate the greenhouse effect, particularly when evaluated of the design and calibration parameters on performance and emissions. Natural Gas
using global warming potential metrics. Engines: For Transportation and Power Generation 251–266.
Comotti, M., Frigo, S., 2015. Hydrogen generation system for ammonia–hydrogen fuelled
(4) Under optimal design conditions, the ammonia-diesel dual fuel internal combustion engines. Int. J. Hydrogen Energy 40 (33), 10673–10686.
methodology has the potential to significantly elevate thermal Di Blasio, G., Ianniello, R., Beatrice, C., Pesce, F.C., Vassallo, A., Belgiorno, G., 2023.
efficiency while simultaneously diminishing carbon-based emis­ Additive manufacturing new piston design and injection strategies for highly
efficient and ultra-low emissions combustion in view of 2030 targets. Fuel 346,
sions, thereby contributing positively to the environmental sus­
128270.
tainability of engine operations. However, this approach has Di Luca, G., Pipicelli, M., Ianniello, R., Belgiorno, G., Di Blasio, G., 2022. Alcohol fuels in
notable drawbacks, including increased emissions of N2O and spark ignition engines. Application of Clean Fuels in Combustion Engines 33–54.
Dimitrakopoulos, N., Belgiorno, G., Tuner, M., Tunestal, P., Di Blasio, G., Beatrice, C.,
unburned ammonia, which pose significant environmental chal­
2017. PPC operation with low ron gasoline fuel. A study on load range on a euro 6
lenges due to the greenhouse gas potential of N2O and the adverse light duty diesel engine. In: The Proceedings of the International Symposium on
environmental and health impacts of unburned ammonia, and Diagnostics and Modeling of Combustion in Internal Combustion Engines. The Japan
also lead to reduced combustion efficiency, crucial in assessing Society of Mechanical Engineers, p. C308.
Duan, X., Feng, L., Liu, H., Jiang, P., Chen, C., Sun, Z., 2023a. Experimental investigation
the overall efficacy of this operation mode. The potential viability on exhaust emissions of a heavy-duty vehicle powered by a methanol-fuelled spark
of ammonia-diesel dual fuel engines is evident from two critical ignition engine under world Harmonized Transient Cycle and actual on-road driving
observations. First, the observed variations in indicated mean conditions. Energy 282, 128869.
Duan, X., Xu, L., Xu, L., Jiang, P., Gan, T., Liu, H., Ye, S., Sun, Z., 2023b. Performance
effective pressure and peak cylinder pressure suggest a stable analysis and comparison of the spark ignition engine fuelled with industrial by-
ignition and combustion process within the context of ammonia- product hydrogen and gasoline. J. Clean. Prod. 424, 138899.

11
J. Liu and J. Liu Journal of Cleaner Production 434 (2024) 140274

Gao, J., Wang, Y., Wang, S., Li, X., Chang, X., Wang, X., Yang, C., Xuan, R., 2022a. Effect Parthasarathy, P., Narayanan, K.S., Arockiam, L., 2013. Study on kinetic parameters of
of catalytic reactions on soot feature evolutions in oxidation process. Chem. Eng. J. different biomass samples using thermo-gravimetric analysis. Biomass Bioenergy 58,
443, 136392. 58–66.
Gao, J., Wang, Y., Li, X., Wang, S., Ma, C., Wang, X., 2022b. Catalytic effect of diesel PM Parthasarathy, P., Narayanan, K.S., Ceylan, S., Pambudi, N.A., 2017. Optimization of
derived ash on PM oxidation activity. Chemosphere 299, 134445. parameters for the generation of hydrogen in combined slow pyrolysis and steam
Huang, Q., Liu, J., 2024. Preliminary assessment of the potential for rapid combustion of gasification of biomass. Energy Fuel. 31 (12), 13692–13704.
pure ammonia in engine cylinders using the multiple spark ignition strategy. Int. J. Pei, Y., Wang, D., Jin, S., Gu, Y., Wu, C., Wu, B., 2023. A quantitative study on the
Hydrogen Energy 55, 375–385. combustion and emission characteristics of an ammonia-diesel dual-fuel (ADDF)
Jin, S., Wu, B., Zi, Z., Yang, P., Shi, T., Zhang, J., 2023. Effects of fuel injection strategy engine. Fuel Process. Technol. 250, 107906.
and ammonia energy ratio on combustion and emissions of ammonia-diesel dual-fuel Reiter, A.J., Kong, S.C., 2011. Combustion and emissions characteristics of compression-
engine. Fuel 341, 127668. ignition engine using dual ammonia-diesel fuel. Fuel 90 (1), 87–97.
Kurien, C., Mittal, M., 2023. Utilization of green ammonia as a hydrogen energy carrier Ryu, K., Zacharakis-Jutz, G.E., Kong, S.C., 2014. Performance enhancement of ammonia-
for decarbonization in spark ignition engines. Int. J. Hydrogen Energy 48 (74), fueled engine by using dissociation catalyst for hydrogen generation. Int. J.
28803–28823. Hydrogen Energy 39 (5), 2390–2398.
Liu, J., Yao, A., Yao, C., 2015. Effects of diesel injection pressure on the performance and Shin, J., Park, S., 2023. Numerical analysis for optimizing combustion strategy in an
emissions of a HD common-rail diesel engine fueled with diesel/methanol dual fuel. ammonia-diesel dual-fuel engine. Energy Convers. Manag. 284, 116980.
Fuel 140, 192–200. Sun, W., Zeng, W., Guo, L., Zhang, H., Yan, Y., Lin, S., Zhu, G., Jiang, M., Yu, C., Wu, F.,
Liu, J., Liu, Z., 2024. In-cylinder thermochemical fuel reforming for high efficiency in 2023. An optical study of the combustion and flame development of ammonia-diesel
ammonia spark-ignited engines through hydrogen generation from fuel-rich dual-fuel engine based on flame chemiluminescence. Fuel 349, 128507.
operations. Int. J. Hydrogen 54, 837–848. Wang, X., Gao, J., Chen, H., Chen, Z., Zhang, P., Chen, Z., 2022. Diesel/methanol dual-
Liu, J., Wu, P., Ji, Q., Sun, P., Wang, P., Meng, Z., Ma, H., 2022. Experimental study on fuel combustion: an assessment of soot nanostructure and oxidation reactivity. Fuel
effects of pilot injection strategy on combustion and emission characteristics of Process. Technol. 237, 107464.
diesel/methanol dual-fuel engine under low load. Energy 247, 123464. Wang, B., Wang, H., Hu, D., Yang, C., Duan, B., Wang, Y., 2023a. Effect of different
Lounici, M.S., Loubar, K., Tarabet, L., Balistrou, M., Niculescu, D.C., Tazerout, M., 2014. ammonia mixing methods for diesel ignition on combustion and emission
Towards improvement of natural gas-diesel dual fuel mode: an experimental performance of high pressure common rail engine. J. Energy Inst. 111, 101402.
investigation on performance and exhaust emissions. Energy 64, 200–211. Wang, B., Wang, H., Hu, D., Yang, C., Duan, B., Wang, Y., 2023b. Study on the
Medrano-García, J.D., Charalambous, M.A., Guillén-Gosálbez, G., 2022. Economic and performance of premixed natural gas/ammonia engine with diesel ignition. Energy
environmental barriers of CO2-based fischer–tropsch electro-diesel. ACS Sustain. 271, 127056.
Chem. Eng. 10 (36), 11751–11759. Wang, B., Yang, C., Wang, H., Hu, D., Wang, Y., 2023c. Effect of diesel-ignited ammonia/
Mi, S., Wu, H., Pei, X., Liu, C., Zheng, L., Zhao, W., Qian, Y., Lu, X., 2023. Potential of hydrogen mixture fuel combustion on engine combustion and emission performance.
ammonia energy fraction and diesel pilot-injection strategy on improving Fuel 331, 125865.
combustion and emission performance in an ammonia-diesel dual fuel engine. Fuel Wang, B., Yang, C., Wang, H., Hu, D., Duan, B., Wang, Y., 2024. Effects of combustion
343, 127889. and emission performance of ammonia/natural gas engines ignited by diesel. Fuel
Mørch, C.S., Bjerre, A., Gøttrup, M.P., Sorenson, S.C., Schramm, J., 2011. Ammonia/ 358, 130323.
hydrogen mixtures in an SI-engine: engine performance and analysis of a proposed Xu, L., Xu, S., Bai, X.S., Repo, J.A., Hautala, S., Hyvönen, J., 2023. Performance and
fuel system. Fuel 90 (2), 854–864. emission characteristics of an ammonia/diesel dual-fuel marine engine. Renew.
Mufrodi, Z., Shitophyta, L.M., Sulistyo, H., Aziz, M., 2023. Reaction of carbon dioxide gas Sustain. Energy Rev. 185, 113631.
absorption with suspension of calcium hydroxide in slurry reactor. Emerging Science Yan, Y., Liu, Z., Liu, J., 2023. An evaluation of the conversion of gasoline and natural gas
Journal 7 (2), 328–338. spark ignition engines to ammonia/hydrogen operation from the perspective of
Nadimi, E., Przybyła, G., Lewandowski, M.T., Adamczyk, W., 2023. Effects of ammonia laminar flame speed. J. Energy Resour. Technol. 145 (1), 012302.
on combustion, emissions, and performance of the ammonia/diesel dual-fuel Yousefi, A., Guo, H., Dev, S., Liko, B., Lafrance, S., 2022a. Effects of ammonia energy
compression ignition engine. J. Energy Inst. 107, 101158. fraction and diesel injection timing on combustion and emissions of an ammonia/
Niki, Y., Nitta, Y., Sekiguchi, H., Hirata, K., 2019. Diesel fuel multiple injection effects on diesel dual-fuel engine. Fuel 314, 122723.
emission characteristics of diesel engine mixed ammonia gas into intake air. J. Eng. Yang, R., Liu, Z., Liu, J., 2024. The methodology of decoupling fuel and thermal nitrogen
Gas Turbines Power 141 (6), 061020. oxides in multi-dimensional computational fluid dynamics combustion simulation of
Ou, J., Yang, R., Yan, Y., Liu, Z., Liu, J., 2023. Investigation of the Effect of Ammonia ammonia-hydrogen spark ignition engines. Int. J. Hydrogen 55, 300–318.
Addition on the Two-Stage Ignition Process of Dimethyl Ether Based on Chemical Yousefi, A., Guo, H., Dev, S., Lafrance, S., Liko, B., 2022b. A study on split diesel injection
Kinetic Analysis. American Society of Mechanical Engineers Internal Combustion on thermal efficiency and emissions of an ammonia/diesel dual-fuel engine. Fuel
Engine Forward Conference. ICEF2023-110108. 316, 123412.
Zeldvich, Y.B., 1946. The oxidation of nitrogen in combustion and explosions. Acta
Physicochimica 21, 577–628.

12