International Journal of Hydrogen Energy 54 (2024) 1319–1328

Contents lists available at ScienceDirect

International Journal of Hydrogen Energy

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

Combustion analysis of methane-hydrogen port injection combined with

OMEn /Diesel spray injection in an RCCI engine
Ali Navid a, *, Kourosh Ghadamkheir b, Anna Hasche a, Javad Zareei c, Hartmut Krause a,
Sven Eckart a
a
Institute of Thermal Engineering, TU Bergakademie Freiberg, Freiberg, Germany
b
Mechanical Engineering Department, Ferdowsi University of Mashhad, Iran
c
Biosystems Engineering Department, Ferdowsi University of Mashhad, Iran

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

Keywords: Compressed Natural Gas (CNG) is recognized as an environmentally friendly alternative vehicle fuel, but it has a
Hydrogen low flame speed. Combining hydrogen with CNG to create ‘hydrogen compressed natural gas’ (HCNG) allows for
Methane faster and leaner combustion. Additionally, OME2− 4 with varying cetane numbers, from 63 for OME2 to 90 for
OME
OME4 , serves as a highly reactive fuel suitable for use in a dual fuel strategy known as RCCI (reactivity-controlled
Diesel
compression ignition). This project investigates the impact of adding OME2− 4 to HCNG, focusing on pressure,
RCCI engine
Emission temperature, and emissions. Results indicate that lengthening the OME chain advances the peak heat release rate
but delays the second combustion phase. In diesel + HCNG, the first-phase combustion is 2 ◦ CA shorter than
OME2 + HCNG, resulting in a 6.5% higher peak pressure than OME2 + HCNG and a 20% increase over basic
diesel. Lengthening the OME chain from OME2 to OME4 raises the peak pressure by about 6.2%. The addition of
gaseous fuels, especially hydrogen, increases peak temperature by 16%. OME4 + HCNG has a similar peak
temperature to OME3 + HCNG, while OME2 + HCNG is about 1.5% lower than OME4 + HCNG.

distribution, hydrogen and CNG blends (HCNG) offer a more viable

1. Introduction
In today’s modern world, increasing demand for transportation due
to new technologies is rapidly raising energy consumption, mainly from
fossil fuels like petroleum [1]. Diesel engines are crucial in ground
transportation but face emission challenges [2]. Recent research has
shown that it’s possible to reduce both soot and NOx emissions simul­
taneously through methods like fuel injection under pressure, super­
charging compression, and fuel additives for PM (particulate matter)
reduction, along with exhaust gas recirculation (EGR) for NOx reduction
[3–5]. Population growth and improved living standards are driving up
fossil fuel consumption, leading to a 36% increase in total energy con­
sumption over the last 15 years [6,7]. This rising energy demand ne­
cessitates increased fuel production, depleting fossil fuel reserves at a
faster rate [8].
Depleting petroleum reserves and environmental concerns have
spurred the search for alternative fuels like compressed natural gas
(CNG), hydrogen, liquefied petroleum gas (LPG), and carbon-neutral
synthetic fuels [1]. While hydrogen faces obstacles like storage and
automotive fuel option without major engine or infrastructure changes
[9]. It enables leaner burning, extending the lean operating limit,
reducing hydrocarbons and CO [5,6]. HCNG offers torque and thermal
efficiency comparable to CNG, with potential for higher thermal effi­
ciency by compromising some NOx gains. It lowers greenhouse gas
emissions like CO2 due to increased H/C ratio [10]. HCNG’s high
hydrogen flame speed optimizes combustion, enhancing the heating
value and engine performance [7,8].
Synthetic fuels like OME (oxymethylene ethers) are produced from
renewable sources and can reduce CO2 emissions [11,12]. Hydrogen for
internal combustion engines is also explored, with polyoxymethylene
dimethyl ethers (OMEn ), the molecular formula of CH3 O(CH2 O)n CH3
(Fig. 1), showing promise in reducing engine emissions, especially soot
[13,14]. These synthetic fuels, lacking C–C bonds, significantly reduce
soot formation, as soot formation involves two mechanisms: particle
formation and oxidation [2,15]. OME/air mixtures have additional
fuel-oxygen bonds, making them leaner and reducing soot formation due
to increased OH radicals [16,17]. OMEn fuels offer a significant reduc­
tion in soot emissions and avoid the trade-off between soot and nitrogen

* Corresponding author.
E-mail address: ali.navid@iwtt.tu-freiberg.de (A. Navid).

https://doi.org/10.1016/j.ijhydene.2023.11.240
Received 27 August 2023; Received in revised form 13 November 2023; Accepted 21 November 2023
Available online 7 December 2023
0360-3199/© 2023 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
A. Navid et al. International Journal of Hydrogen Energy 54 (2024) 1319–1328

Nomenclature IC internal combustion

ID ignition delay
aTDC after top dead center IP indicated power
BMEP brake mean effective pressure IMEP indicated mean effective pressure
bTDC before top dead center LHV lower heating value
CA crank angle MFB mass fuel burned
CFD computational fluid dynamic NOx Nitrogen oxide
CNG compressed natural gas NTP normal temperature and pressure
DI direct injection OME oxymethylene ethers
EVO exhaust valve opened PM particulate matter
EGR exhaust gas recirculation PtL power-to-liquid
GHG greenhouse gas RCCI reactivity controlled compression ignition
HPCR high-pressure common rail VGT variable geometry turbocharging
IVC intake valve closed

solubilization capacity, and compatibility with organic compounds and

Fig. 1. Molecular structures of OMEn .
oxides (NOx). They can be produced from renewable sources through
various methods, including power-to-liquid (PtL) and biomass processes
[18]. These fuels are fully compatible with conventional hydrocarbon
fuels and have high cetane numbers (e.g., 63 for OME2 up to 90 for
OME4), which improves diesel fuel ignition behavior [19,20].
Reactivity controlled compression ignition (RCCI) is a dual fuel
strategy that uses fuels with different reactivity in order to improve the
combustion process and reduce emissions from a diesel engine. Com­
bustion characteristics of the engine under RCCI mode are controlled by
varying the quantity and injection timings of the dual-fuel charge (high
and low reactivity fuels). In dual fuel and RCCI engines [21,22], adding
hydrogen boosts ignition and peak pressure [23,24]. Injection strategy is
crucial for combustion control. With hydrogen/methane as the low
reactivity fuel, NO increases while CO decreases [25,26]. RCCI engines
using diesel and natural gas can cut fuel consumption by up to 60% [27].
Combustion with HCNG-air and OME works in RCCI mode [28,29].
Hydrogen mainly affects early combustion stages without excessive
noise or knock [30,31].
Polyoxymethylene dimethyl ethers (OMEn ) are polyether com­
pounds with a structure similar to methylal, formed by inserting oxy­
methylene groups (–CH2 O–) into the methylal molecule [32,33]. OMEn
possess unique physicochemical properties including low toxicity,
excellent volatility, low condensation point, high permeability, good
water [34]. Their C–O and C–C bond lengths (0.143 nm and 0.154 nm,
respectively) result in higher density compared to n-alkanes, with den­
sity inversely related to temperature and directly related to chain length
[34]. For more details, refer to Table 1.
In the realm of combustion dynamics within RCCI engines, a con­
spicuous gap in comprehensive research concerning the cascading ef­
fects of OMEn fuels has come to light through an exhaustive review of
existing literature. Our current investigation not only bridges this void
but also advances our comprehension by providing a pertinent com­
parison between OME and conventional high reactivity diesel fuels
within the context of RCCI combustion. Our study primarily focuses on
elucidating the impacts of OME2− 4 and diesel fuels in conjunction with
hydrogen and methane (HCNG), known for its lower reactivity, under
demanding operating conditions of 1250 rpm at full load. The goal is to
comprehensively evaluate engine behavior, benchmarked against con­
ventional diesel combustion. To achieve this, we employ advanced
software tools. Chemkin software aids in 0-1D analysis, enabling an
exploration of key parameters like laminar flame speed, reaction path­
ways, and sensitivity across various fuel configurations. Furthermore,
we leverage AVL Fire software, along with its workflow manager mod­
ule, to facilitate intricate 2-3D simulations.
2. Materials and methods
In the methodology section of the research, a clear demarcation
between the experimental and simulation components are implemented
to ensure transparency and comprehensibility in the investigative
process.
2.1. Engine configuration
In order to simulate a diesel engine and change its configuration to
an RCCI engine, a Ford 1.8 L DI (Direct Injection) diesel engine equipped

Table 1
Fuel properties [35–37].
Property Diesel OME2 OME3 OME4 Methane Hydrogen

Chemical formula – C4 H10 O3 C5 H12 O4 C6 H14 O5 CH4 H2

Cetane/Octane number >51 63 78 90 120 130
Density @ 298 K (g/cm3 ) 0.82–0.845 0.9622 1.0220 1.0671 0.000657 8.987e-5
Viscosity @ 298 K (mPa s) 1.64–3.8 0.6809 1.1333 1.8805 0.01107 0.000892
Oxygen content (%) – 45.3 47.1 48.2 – –
H/C ratio – 2.50 2.40 2.33 4 –
O/C ratio – 0.75 0.80 0.83 – –
Surface tension (mN/m, 20 ◦ C) 25.84 28.8 30.7 32.6 – –
Lower Heating Value (MJ/kg) 42.5 20.32 19.14 18.38 50 120
Vapor pressure @ 298 (kPa) 0.011 4.2 0.4 0.03 – –

1320
A. Navid et al. International Journal of Hydrogen Energy 54 (2024) 1319–1328

with a prototype Lucas CAV HPCR (High-Pressure Common Rail) system Table 3
and an Allied Signal VGT (Variable Geometry Turbocharging) is used. Models used for the simulation process.
The VGT mechanism optimizes the airflow based on engine conditions, Description Model
and it depends on the engine speed and engine load. This will lead to
Combustion ECFM-3Z
better control of fuel injection and airflow, thus optimizing the unifor­ Turbulent model k-zeta-f
mity of the air/fuel mixture and droplet atomization. Table 2 shows Auto ignition model Two-stage
engine specifications (values for injection are correlated to 1250 rpm). NOx model Extended Zeldovich + prompt
The model is valid for a typical diesel condition at 1250 rpm and Soot Kennedy/Hiroyasu/Magnussen
Breakup model Wave
under full load condition. To discretize the Navier-Stokes set of equa­ Drag law model Schiller Naumann
tions, the finite volume method is used. Hence, SIMPLE/PISO algorithm Evaporation model Dukowicz
is used to correct the pressure-velocity field, which is suitable for un­ Wall interaction model Walljet 1
steady flows. Blob injection model is used to primary break up [38] and
since high Reynolds number motion has a great importance for pre­
dicting the model in subsequent interactions in the flow domain, k-zeta-f Table 4
is the model used to simulate the turbulence inside the combustion Initial and boundary conditions for 1250 rpm.
chamber. In order to model the combustion process, ECFM-3Z model Boundary/initial condition Value
[39] is used. By using this model, the fuel can be defined as a mixture of
Head temperature 550.15 K
several chemical species. Boundary values and derivatives are calculated
Piston temperature 575.15 K
by extrapolate and least squares fit models, respectively. The compu­ Cylinder temperature 475.15 K
tations run with MINMOD relaxed for momentum and Central differ­ diesel injection quantity 14.15 mg/cycle
encing schemes for continuity. Moreover, Upwind method is used for Inlet valve closed 50◦ CAaBDC
turbulence, energy, and scalar. The convergence criteria are set for 50 Exhaust valve open 126◦ CAaTDC
Simulation range 230-486 ◦ CA
and 5 for maximum and minimum iterations, respectively. Reduction of Initial fuel temperature (Diesel/HCNG) 300/283 K
residuals method with 0.0001 value is considered for pressure and
momentum. In order to capture heating and evaporating of fuel sprays
and breakup of droplets, Dukowicz [40,41] and wave models [32] are
incorporated. Table 3 depicts models used for the simulation in-detail.
It should be noted that since the base model is a typical diesel engine
with one fuel, and the modified system is configured with three fuels
(OME/diesel, Hydrogen, Methane), the evaporation model for RCCI
configurations is changed to multi-component. However, the initial and
boundary conditions are the same for both the typical diesel and RCCI
configurations, except for the injected fuel mass. Table 4 shows the
boundary and initial conditions.
The aforementioned engine is basically a diesel engine which is
modified in this work to be used as a RCCI engine using HCNG blend
(60% CNG and 40% hydrogen) as a low reactivity fuel and OMEn and
diesel as high reactivity fuels. HCNG enters the cylinder with the port
injection together with air and OME/diesel is injected with an injector at
the proper time.

2.2. Engine simulation

Since there is no spark, only one-fourth of a cylinder is simulated due

to the axisymmetric condition. Fig. 2(a) shows the mesh with 55,996
cells at TDC (top dead center), and Fig. 2(b) demonstrates the mesh
independence investigations for 1250 rpm and full load conditions. Grid
independence analysis is an important part of any numerical study;
hence, three different cell numbers are considered. As shown, the dif­
ference between the models with approximately 56,000 and 60,000 cells
is negligible. Finally, a model with a cell size of 1 mm and 56,000 cells is
selected for the rest of the process. Due to the high computational effi­
ciency and simplicity, the middle mesh (~56,000 cells) is used as the

Table 2
Engine specifications.
Description (unit) Value

Bore × stroke (mm) 82.5 × 82

Displacement (cm3/cylinder) 438
Compression ratio 19.5:1
Swirl ratio @ IVC 3 Fig. 2. (a) Mesh used for the simulation process at top dead center (TDC), and
Clearance (mm) 0.86 (b) mesh independence for pressure at 1250 rpm.
Connecting rod length (mm) 104
Start of injection 4 ◦ CA bTDC
Injection duration 11 ◦ CA
Injection spray angle 160◦

1321
A. Navid et al.
reference mesh. It is noteworthy that a system with an Intel Core i7 Q820
1.73 GHz CPU with 8 GB RAM is used for all simulation processes, and
each run takes approximately 150 min.
In order to validate the results, the numerical data are compared to
the experimental data. The numerical results in Fig. 3, derived from a
mesh with 56,000 cells, exhibit a high degree of agreement with the
experimental data, with a deviation of less than 1%. Fig. 3 represents the
validation diagram for pressure under 1250 rpm and full load
conditions.
The geometrical domain, consisting of 55,996 cells (approximately
56k), demonstrates a high level of consistency with experimental data
without imposing excessive computational demands on the system.
2.3. 0-D and 1-D flame simulations
For the calculation of adiabatic laminar flame speeds and their sen­
sitivities, freely propagating one-dimensional flames are investigated
numerically using the laminar premixed flame speed code of Chemkin-
Pro 2020 R2. The laminar flame speed is calculated using multicom­
ponent transport coefficients without the thermal diffusion option. The
number of grid points is set to 1400 with respect to the gradient (GRAD
= 0.04) and curvature (CURV = 0.08) adaptive mesh parameters.
Further details are explained in the work of Eckart et al. [42]. It is
noteworthy that since there is no mechanism in Chemkin that supports
both OMEn and heavy fuels such as diesel, flame speed studies are not
conducted for diesel, and the Cai mechanism (with 156 species and 1005
reactions) [7] is used for OMEn configurations. In this work, the
auto-ignition process of OME(2− 4) is studied. The novelty lies in the use
of an automatic method for the auto-ignition process of OME(2− 4) in
order to generate a reaction mechanism. Considering reaction classes
correlated to the oxidation process of OME1 in the mechanism genera­
tion validates the mechanism. Further optimization is performed to
improve the prediction accuracy for ignition delay times, which is done
through rate rules. These ignition delay times for OME(2− 4) are measured
using shock tubes, which can cover a range of initial conditions. Finally,
there is good agreement between the optimized model and the experi­
mental data.
To clarify the reaction mechanisms underlying oxidation kinetics of
OME2− 4 , the auto ignition process of these fuels at 20 bar, 825 K, λ = 1,
and 20% fuel consumption is investigated by Cai et al. [7]. The results
show that the main products of OMEn consumption is OH and HO2
radicals and they derive to primary fuel radicals. Further oxidation yield
Fig. 3. Validation of numerical results with 56k cells for the baseline diesel
engine for pressure at 1250 rpm and full load condition.
1322
International Journal of Hydrogen Energy 54 (2024) 1319–1328
to form peroxy radicals and finally hydroperoxide radicals are
isomerized.
Moreover, n-alkanes have two main kinetic reactions: a primarily
high-temperature mechanism and a lower-temperature mechanism. In
the primarily high-temperature mechanism, the major alkyl radicals
decompose into smaller alkyl radicals and olefin species. Additionally,
adding molecular oxygen to the major alkyl radicals initiates the lower-
temperature mechanism. It is noteworthy that the temperature bound­
ary between the low and high-temperature regimes varies with pressure,
which is due to the dependence of the reaction equilibrium on both
pressure and temperature [36].
3. Results and discussion
This research concerns the modification of a conventional diesel
engine into an RCCI engine that uses HCNG (composed of 60% CH4 and
40% H2 ) as a low reactivity fuel, and OME(2− 4) and diesel as high
reactivity fuels (HCNG constitutes 0.5% of the incoming air mass frac­
tion, meaning that 0.5% of the air entering the chamber is HCNG). To
ensure a fair comparison, the volume of injected fuel for the base engine
(diesel) remains constant across all cases, and the injected mass varies
based on the density of OMEn . However, the HCNG mixture remains
consistent for all RCCI cases. Table 5 provides details on the injected
mass for each fuel.
3.1. Laminar flame speed analysis
To conduct a comprehensive analysis of the results, we initially
conducted a fundamental study on the combustion of OMEn , OMEn +
HCNG, hydrogen, methane, and HCNG (comprising 60% methane and
40% hydrogen). For instance, Fig. 4 illustrates the computed laminar
flame speeds using the Cai mechanism [7]. For comparison, methane,
hydrogen, OME2 , and HCNG are plotted as neat fuels. The unburnt gas
temperature is set at 298 K, and the pressure at 1 bar to ensure a fair
comparison. It is evident that hydrogen exhibits a flame speed approx­
imately 8 times higher than methane at its maximum point (diesel is not
shown in this figure, as there is no mechanism available to cover both
diesel and OME2 ).
The maximum flame speed of hydrogen is at an equivalence ratio of
approximately φ = 1.7, while for methane, it is at φ = 1.0, which is
consistent with the literature [43]. OME2 , as a liquid fuel shows its
maximum at about φ = 1.2, which agrees very well with the measure­
ments of Eckart et al. [42]. The maximum flame speed for methane is
around 36–37 cm/s, while it is approximately 48 cm/s for OME2 , indi­
cating that the maximum flame speed of OME2 is higher than that of
methane. The mixture of OME2 + HCNG (comprising 96% mass OME2 ,
1.5% mass fraction hydrogen, and 2.5% mass fraction methane) is then
compared with these flame speeds. It can be observed that despite the
minimal amount of hydrogen, there is a noticeable increase in the flame
speed, approximately 12% higher compared to the maximum value of
neat OME2 . Fig. 4 provides a comparison of the laminar flame speeds of
these fuels.
Table 5
Injected fuel quantity.
Fuel quantity (mg/cycle)
diesel 14.15
OME2 16.403
OME3 17.390
OME4 18.07
CH4 0.368
H2 0.245
A. Navid et al.
Fig. 4. Laminar flame speed comparison for different fuels at T = 298 K and
P = 1 bar (HCNG with 60% CH4 and 40% H2 ) using Cai mechanism
in Chemkin.
Fig. 5. Rate of heat release for different fuels vs. crank angle degree. RCCI
(Reactivity controlled compression ignition) configurations have drastically
higher peak for the first combustion period.
3.2. Combustion and performance analysis
In order to analyze the combustion process in the engine, rate of heat
release is considered. Fig. 5 depicts the results for the rate of heat release
depending on the crank angle for different fuel mixtures.
RCCI configurations exhibit almost identical behavior during the
initial combustion period, where 10–50% of the fuel mass is burned
(MFB10− 50 ), primarily due to their high cetane number (see Table 1).
The formation of the spray is influenced by a balance between various
forces, including inertia, viscosity, surface tension, and aerodynamics
[44]. During the initial stage, the penetration speed is predominantly
governed by inertia and surface tension, with the inertia of a liquid jet
being primarily dependent on the injection pressure. Additionally, in
compression ignition engines, the initiation of combustion is influenced
by the cetane number. Since the injection pressure remains consistent
across all cases, it can be inferred that the initiation of combustion
1323
International Journal of Hydrogen Energy 54 (2024) 1319–1328
results from a combination of surface tension, viscosity, and cetane
number.
Fig. 5 illustrates that nearly all fuel configurations initiate combus­
tion reactions at the same time. However, the sudden rise in the curve is
attributed to this combination effect. Subsequently, this figure implies
that the first phase of combustion (MFB10− 50 ) concludes earlier for
diesel + HCNG, and as the OME chain lengthens, this duration
decreases.
As evident from the diagram, the diesel combustion curve exhibits
two peaks in close proximity to each other (the first peak at around 361
◦
CA and the second at 362 ◦ CA). However, under RCCI conditions,
although there are still two peaks, they are much further apart. In RCCI
configurations, the first peak is higher, influenced by the presence of the
HCNG mixture. This suggests that HCNG primarily affects the premixed
part of diesel combustion, and the second stage (the diffusion phase)
largely follows the typical combustion pattern. One possible reason
could be the limited time available for mixing the liquid fuel and air
within the combustion chamber.
The vapor pressure of a substance at a given temperature influences
its volatility, with higher vapor pressure indicating greater volatility
[36]. In typical compression ignition engines, the period of 50–90%
mass fraction burned (MFB50− 90 ) is governed by the diffusive process,
which, in turn, depends on fuel volatility and the air-fuel mixture in the
combustion process [45]. Table 1 shows that as the OME chain
lengthens, its vapor pressure decreases, while diesel has the lowest
vapor pressure. Consequently, the combustion duration during the
diffusion stage (MFB50− 90 ) is the longest for diesel, and as the –CH2 O–
chain shortens, this duration becomes shorter.
The higher heat release observed in the diesel + HCNG configura­
tion, coupled with a higher flame speed, suggests that the presence of
hydrogen leads to a rapid rate of heat release over a short period,
potentially resulting in higher temperatures compared to typical diesel
combustion. Fig. 6 depicts the temperature changes during the com­
bustion process for various fuel configurations.
The presence of gaseous fuels and especially hydrogen in the com­
bustion process of RCCI configurations has increased the peak temper­
ature from 2092 K in the basic diesel case to 2494 K in diesel + HCNG
case which shows 16% increase. Besides, OME4 + HCNG has almost the
same peak temperature in comparison to OME3 + HCNG. However, the
peak temperature for OME2 + HCNG is 2369 K which is around 1.5%
lower than the value for OME4 + HCNG with 2409 K at the peak.
Fig. 6. In-cylinder temperature variance during different fuel mixtures (Diesel
results are based on experimental data). Temperature rising with HCNG mix­
tures in RCCI (Reactivity controlled compression ignition) configurations
is noticeable.
A. Navid et al.
The higher flame speed for hydrogen [15,35] causes for higher
pressure at the vicinity of top dead center (TDC), which correspondingly
causes higher temperature inside the cylinder. On the other side, it
should be noticed that lower heating value for hydrogen is several times
more than diesel or OMEn . Fig. 7 depicts pressure diagram for all fuel
configurations.
This diagram indicates an inverse correlation between the maximum
pressure during combustion and the duration of the first phase of the
combustion process (MFB10− 50 ). Fig. 5 suggests that the combustion
duration in the first phase for diesel + HCNG is almost 2 ◦ CA shorter
than that for OME2 + HCNG, and correspondingly, the peak pressure for
diesel + HCNG (144.2 bar) is approximately 6.5% higher than that for
the OME2 + HCNG (134.9 bar) configuration. Notably, the difference
between the peak pressure for diesel + HCNG and basic diesel (115.0
bar) is roughly 20%. This figure demonstrates that lengthening the OME
chain from OME2 to OME4 increases the peak pressure by approximately
6.2% (from 134.9 bar to 143.3 bar).
Regarding engine performance, when modifying the combustion
process to RCCI, a slight increase in indicated mean effective pressure
(IMEP) is observed in diesel + HCNG combustion, as shown in Table 6.
This is due to the presence of some extra gaseous fuel in comparison to
pure diesel and thus, diesel + HCNG shows lower ISFC than pure diesel.
Among the various OMEs investigated, OME4 exhibits the highest IMEP.
However, in terms of thermal efficiency, OME2 + HCNG fuel stands out
as it offers the best efficiency, attributed to OME2 ’s cooler flame and a
higher ratio of work produced to heat released.
Furthermore, the study on specific fuel consumption reveals that
more OME fuel is required to produce the same power as diesel. This is
due to the lower heating value and energy density of OME fuels
compared to diesel fuel.
As mentioned earlier, the equivalence ratio represents the ratio of
fuel mass to oxidizer mass (in this case, air) compared to the same ratio
for stoichiometric conditions. The equivalence ratio is a critical
parameter in engine operation, influencing combustion efficiency,
power output, emissions control, fuel economy, and overall engine
performance. Since hydrogen has a relatively low flammability limit of
0.1, it can help maintain the combustion chamber in a lean condition.
This lean condition can keep the combustion process closer to complete
combustion [35]. Fig. 8 shows the equivalence ratios for RCCI cases at
12 ◦ CA after TDC.
Fig. 7. Pressure rise comparison of diesel and RCCI fuel configurations showing
the effect of OME chain length and diesel in RCCI (Reactivity controlled
compression ignition) configuration on the peak pressure compared to basic
diesel (Diesel results are based on experimental data).
1324
International Journal of Hydrogen Energy 54 (2024) 1319–1328
Table 6
Engine performance analysis for the various simulated fuels.
Fuel IMEP (bar) Thermal efficiency (%) ISFC (g/kWh)
Diesel 10.9 31.39 379.4
Diesel + HCNG 10.99 34.83 376.7
OME2 + HCNG 10.37 35.10 459.5
OME3 + HCNG 10.73 34.43 465.3
OME4 + HCNG 10.92 33.52 491.2
Fig. 8 reveals that the presence of methane and hydrogen has
significantly increased the homogeneity of the combustion chamber,
especially evident for OME2 . Two factors contribute to this homogene­
ity. First, methane and hydrogen exist in the gaseous phase and fill the
combustion chamber until combustion begins. Second, OME2 has a low
boiling point, high vapor pressure, and, consequently, high volatility
compared to diesel. This property aids in dispersing the fuel throughout
the entire chamber. However, as the chain lengthens, the boiling point
increases and vapor pressure decreases (as seen in Table 1), resulting in a
higher equivalence ratio. This homogeneous charge leads to uniform
combustion. Fig. 9 illustrates the temperature distribution at 12 ◦ CA
after TDC for diesel and diesel + HCNG conditions.
Fig. 9 illustrates uniform combustion during the RCCI combustion
mode, attributed to the presence of hydrogen and methane within the
combustion chamber. The homogeneous charge inside the combustion
chamber generates uniform heat, which can ultimately be converted
into positive mechanical work. It is worth noting that although Fig. 9(a)
shows higher temperatures in certain zones compared to Fig. 9(b), the
average temperature for diesel + HCNG is higher (as seen in Fig. 6). This
can be further confirmed by examining surface flame density. In tur­
bulent premixed flames, evaluating reaction rates can be conveniently
done through surface flame density. This parameter measures the
average available flame surface area per unit volume and is associated
with the mean reaction progress. Fig. 10 presents this parameter in [1/
m] for RCCI fuel configurations.
Fig. 10 directly correlates with the reactivity of the fuel mixture. This
figure demonstrates that the flame propagates much more in the diesel
RCCI condition compared to the basic diesel condition. Clearly, the
presence of HCNG, especially hydrogen with its high flame speed (as
seen in Fig. 4), has a positive effect on this propagation.
3.3. Emission analysis
Sensitivity analysis, in fact, demonstrates the impact of a reaction on
a reaction parameter, where a negative value indicates a reduction in the
final parameter, and a positive value enhances it. For instance, sensi­
tivity analysis (refer to Fig. 11) can provide additional insights into the
combustion process. It clearly illustrates the influence of HCNG
admixture on the reaction pathways during the conversion of OME2 . On
one hand, reactions involving free H radicals are significantly increased
in all cases. This is particularly evident in the case of H + O2 →O + OH,
but also with the acceleration of the reaction H + O2 →HO2 + M.
Comparatively, the number of free OH radicals in the radical pool also
increases with the addition of HCNG to OME2 .
When comparing the sensitivity analysis, it becomes evident that the
chain branching reaction H +O2 → O +OH plays a pivotal role in accel­
erating the oxidation of CO by influencing the balance between H-con­
sumption and production. Furthermore, the addition of HCNG primarily
enhances the reactions H +O2 →O +OH and HO2 + H →H2 + O2 , as
depicted in this figure. Similarly, the analysis of the Cai mechanism
reveals an intensified reaction between HCO, which occurs during the
formation of CO and H. Additionally, the simulation indicates that the
formation reactions for combustion products via CO + OH→CO2 +H are
A. Navid et al.
Fig. 8. Equivalence ratio φ at 372 ◦ CA for (a) diesel, (b) OME2 + HCNG, (c) OME3 + HCNG, (d) OME4 + HCNG, and (e) diesel+HCNG (two column picture).
Fig. 9. Temperature distribution at 372 ◦ CA for (a) diesel, and (b)
diesel+HCNG which shows high homogeneity for diesel + HCNG due to the
presence of a gaseous fuel.
significantly inhibited, whereas the oxidation of methyl groups is only
minimally affected by the presence of HCNG.
Fig. 12 presents a comparison of CO emissions for different cases. As
shown, CO emissions exhibit a direct relationship with the length of the
OME chain, with the amount of CO being nine times higher for OME4
compared to OME2 . One reason for this is that, as the chain lengthens,
the H/C ratio decreases (as seen in Table 1), leading to increased carbon
emissions production. However, the presence of hydrogen and methane
(with an H/C ratio of 4) elevates the H/C ratio within the cylinder in
RCCI configurations. Although CO production initially increases during
the combustion process, it rapidly diminishes due to complete com­
bustion. This complete combustion is facilitated by the high O/C ratio
(as indicated in Table 1), which increases as the OME chain lengthens.
1325
International Journal of Hydrogen Energy 54 (2024) 1319–1328
Fig. 10. Flame surface density for various fuel configurations at 372 ◦ CA for (a)
diesel, and (b) diesel + HCNG.
On the other hand, soot particles are impure carbon particles that
result from incomplete combustion. The presence of hydrogen and
methane increases the H/C ratio, which reduces carbon emissions.
Additionally, the presence of oxygen in OME fuels, enhancing complete
combustion, leads to a reduction in soot formation in OMEn fuel-related
combustions. However, increasing the chain length increases soot for­
mation since the C/H ratio increases with longer chains. Fig. 13 illus­
trates soot formation for different fuel configurations. It’s worth noting
that, due to the absence of C − C bonds in OMEn compounds, there
should be lower soot production compared to common hydrocarbon
fuels.
Another emission that should be considered is NO (nitrogen oxide).
A. Navid et al.
Fig. 11. Sensitivity analysis for OME2 and OME2 + HCNG at T = 298 K and
P = 1 bar with the mechanism of Cai et al.
Fig. 12. CO emission for different fuel configurations (Diesel results are based
on experimental data).
This emission is primarily produced in the initial phase of combustion
and has a direct correlation with temperature. As indicated by Figs. 5
and 6, OMEn and diesel RCCI exhibit a higher rate of heat release and
temperature than basic diesel, and this elevated peak temperature can
lead to a significantly higher NO production. Therefore, it is expected
that RCCI configurations will produce more NO emissions compared to
basic diesel. Fig. 14 illustrates the NO production during the combustion
process.
The results for OH radicals confirm the findings for NO emissions, as
OH radicals serve as precursors for NO formation. The concentration of
OH radicals has a direct correlation with the OME chain length. This
correlation may arise from the increased availability of free oxygen in
the combustion chamber with oxygenated fuels, where it becomes
evident that greater OH production results in higher NO emissions. It’s
worth noting that the high reactivity of hydrogen plays a significant role
in the increased quantity of OH radicals. Furthermore, the peak loca­
tions of OH radicals correspond to the peaks of NO fractions, which is
1326
International Journal of Hydrogen Energy 54 (2024) 1319–1328
Fig. 13. Mean Soot production for different fuel configurations at final point
(Diesel results are based on experimental data).
Fig. 14. NO mean mass fraction for different fuel configurations (Diesel results
are based on experimental data).
expected since both are indicative of a reaction layer. Fig. 15 illustrates
the production of OH radicals during the combustion process.
It is recognizable that due to the presence of hydrogen in RCCI
configurations, OH radical production starts soon compared to typical
diesel.
4. Conclusion and future perspectives
This study investigates the combustion of an HCNG blend, along with
OMEn and diesel, in an RCCI engine, where HCNG serves as the low
reactivity fuel, and OMEn and diesel serve as the high reactivity fuels.
The results reveal that as the OME chain length increases, the combus­
tion duration in the first phase (MFB10− 50 ) decreases due to lower
volatility and a higher cetane number. However, lengthening the OME
chain increases the duration of the second combustion phase
(MFB50− 90 ). The shorter combustion duration in the first phase leads to
an increase in temperature and pressure, as they have an inverse cor­
relation with combustion duration. The key findings extracted from this
study are as follows:
A. Navid et al.
Fig. 15. OH mean mass fraction production; the peak point of OH and its
correlation with the peak point of NO in the last figure is noticeable.
• The addition of HCNG to Diesel or OMEn can increase flame speed
and impact the underlying reactions.
• Diesel exhibits lower volatility compared to OMEn , resulting in a
longer diffusion combustion process due to the higher diffusivity of
the fuel in the air. Furthermore, increasing the OME chain length
enhances this characteristic.
• The inclusion of HCNG in Diesel and OMEn leads to higher pressure
and temperature near top dead center, promoting a more homoge­
neous charge in the combustion chamber and facilitating complete
combustion.
• Increasing the OME chain length is associated with higher CO
emissions due to a lower H/C ratio.
• The presence of HCNG significantly reduces soot production when
added to Diesel and OMEn . This positive effect of HCNG can be
attributed to an increase in the H/C ratio in RCCI configurations,
which is a result of the absence of C − C bonds in OME fuels.
Hydrogen is poised to play a significant role in the future of energy
worldwide. Additionally, synthetic fuels like OME may find increased
applications in combustion engines. Consequently, hydrogen can be
viewed as an independent fuel in an RCCI engine, serving as the lower
reactivity fuel alongside OMEn , which acts as the high reactivity fuel.
Furthermore, the impact of engine load on the combustion process is
another important consideration for this combination.
OME, a synthetic fuel, entails a costlier alternative compared to
traditional diesel, necessitating a prudent consideration of potential
economic limitations for consumers when implementing this fuel in an
engine. Furthermore, the utilization of RCCI engines, as a pioneering
concept in the engineering landscape, introduces distinctive mainte­
nance challenges that demand careful attention. On the contrary, the
incorporation of hydrogen, a gaseous fuel, underscores the imperative
need for rigorous safety protocols to ensure its proper handling. Future
studies should investigate the sustainability features of the findings
presented using advanced sustainability assessment tools, such as life
cycle assessment (LCA) analysis, exergy, etc.
The practical implications of employing RCCI engines with a com­
bination of HCNG (hydrogen and methane) as a low reactivity fuel
alongside high reactivity fuels like diesel or OME (Oxygenated Molecule
of Ethanol) are promising. This approach can lead to improved fuel ef­
ficiency, reduced emissions, and a reduced dependence on fossil fuels,
making it a valuable option for environmentally friendly and efficient
power generation and transportation. However, it comes with chal­
lenges related to safety, infrastructure, and potential cost
1327
International Journal of Hydrogen Energy 54 (2024) 1319–1328
considerations, which need to be carefully addressed for widespread
adoption. Overall, RCCI engines with HCNG and high reactivity fuels
offer a practical and versatile solution for a more sustainable and
adaptable energy landscape.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgement
The authors from TU Bergakademie Freiberg express their appreci­
ation to AVL for granting them a license to use the AVL suite.
References
[1] Rahman MM, Kamil M, Bakar RA. Engine performance and optimum injection
timing for 4-cylinder direct injection hydrogen fueled engine. Simulat Model Pract
Theor 2011;19(2):734–51.
[2] Dworschak P, Berger V, Härtl M, Wachtmeister G. Neat oxymethylene ethers:
combustion performance and emissions of OME 2 OME 3 OME 4 and OME 5 in a
single-cylinder diesel engine. 2020.
[3] Barro C, Meyer P, Boulouchos K. Optical investigations of soot reduction
mechanisms using post-injections in a cylindrical constant volume chamber
(CCVC). In: SAE technical paper series. SAE International400 commonwealth drive;
2014. Warrendale, PA, United States.
[4] Drost S, Kaczmarek D, Eckart S, Herzler J, Schießl R, Fritsche C, et al. Experimental
investigation of Ethanol oxidation and development of a reduced reaction
mechanism for a wide temperature range. Energy Fuels 2021;35(18):14780–92.
[5] Wei Y, Zhu Z, Liao Y, Liu S, Shi Z, Zeng Z, et al. Numerical investigations on the
effects of EGR routes on the combustion characteristics and efficiency of a heavy-
duty SI methanol engine. Fuel Process Technol 2023;250:107861.
[6] Minic D. Hydrogen energy: challenges and perspectives. InTechOpen; 2012.
[7] Cai L, Jacobs S, Langer R, vom Lehn F, Pitsch H. Auto-ignition of oxymethylene
ethers (OMEn, n = 2–4) as promising synthetic e-fuels from renewable electricity:
shock tube experiments and automatic mechanism generation. Fuel 2020;(264):
116711.
[8] Lee C-L, Jou C-JG. Saving fuel consumption and reducing pollution emissions for
industrial furnace. Fuel Process Technol 2011;92(12):2335–40.
[9] Drost S, Eckart S, Yu C, Schießl R, Krause H, Maas U. Numerical and experimental
investigations of CH4/H2 mixtures: ignition delay times, laminar burning velocity
and extinction limits. Energies 2023;16(6):2621.
[10] British petroleum statistical review of world energy. 2001.
[11] Bauer C, Forest TW. Effect of hydrogen addition on the performance of methane-
fueled vehicles. Part I: effect on S.I. engine performance. Int J Hydrogen Energy
2001;26(1):55–70.
[12] Gopidesi RK, Sr P, Dhana Raju V. Mitigation of harmful exhaust pollutants of DI
diesel engine using emulsified fuel and hythane gas in a dual-fuel mode. Energy
Sources, Part A Recovery, Util Environ Eff 2021:1–23.
[13] Das L, Gulati R, Gutpta PK. A comparative evaluation of the performance
characteristics of a spark ignition engine using hydrogen and compressed natural
gas as alternative fuels. Int J Hydrogen Energy 2000;25(8):783–93.
[14] Schmitz N, Burger J, Ströfer E, Hasse H. From methanol to the oxygenated diesel
fuel poly(oxymethylene) dimethyl ether: an assessment of the production costs.
Fuel 2016;185:67–72.
[15] Shrestha KP, Eckart S, Drost S, Fritsche C, Schießl R, Seidel L, et al.
A comprehensive kinetic modeling of oxymethylene ethers (OMEn, n=1–3)
oxidation - laminar flame speed and ignition delay time measurements. Combust
Flame 2022;246:112426.
[16] Cheng AS, Dibble RW, Buchholz BA. The effect of oxygenates on diesel engine
particulate matter. In: SAE technical paper series. SAE International400
commonwealth drive; 2002. Warrendale, PA, United States.
[17] Delfort B, Durand I, Jaecker-Voirol A, Lacôme T, Paillé F, Montagne X. Oxygenated
compounds and diesel engine pollutant emissions performances of new generation
of products. In: SAE technical paper series. SAE International400 commonwealth
drive; 2002. Warrendale, PA, United States.
[18] Richter S, Kathrotia T, Braun-Unkhoff M, Naumann C, Köhler M. Influence of
oxymethylene ethers (OMEn) in mixtures with a diesel surrogate. Energies 2021;14
(23):7848.
[19] Bhatelia T, Lee WJ, Samanta C, Patel J, Bordoloi A. Processes for the production of
oxymethylene ethers: promising synthetic diesel additives. Asia Pac J Chem Eng
2017;12(5):827–37.
[20] EN-590. Automotive Fuels-Diesel-Requirements and test methods. 2009.
[21] Reitz RD, Duraisamy G. Review of high efficiency and clean reactivity controlled
compression ignition (RCCI) combustion in internal combustion engines. Prog
Energy Combust Sci 2015;46:12–71.
[22] Hosseini SH, Tsolakis A, Alagumalai A, Mahian O, Lam SS, Pan J, et al. Use of
hydrogen in dual-fuel diesel engines. Prog Energy Combust Sci 2023;98:101100.
A. Navid et al.
[23] Liu J, Yang F, Wang H, Ouyang M. Numerical study of hydrogen addition to DME/
CH4 dual fuel RCCI engine. Int J Hydrogen Energy 2012;37(10):8688–97.
[24] Nazemi M, Shahbakhti M. Modeling and analysis of fuel injection parameters for
combustion and performance of an RCCI engine. Appl Energy 2016;165:135–50.
[25] Ahmadi R, Hosseini SM. Numerical investigation on adding/substituting hydrogen
in the CDC and RCCI combustion in a heavy duty engine. Appl Energy 2018;213:
450–68.
[26] Kakoee A, Bakhshan Y, Aval SM, Gharehghani A. An improvement of a lean
burning condition of natural gas/diesel RCCI engine with a pre-chamber by using
hydrogen. Energy Convers Manag 2018;166:489–99.
[27] Sattarzadeh M, Ebrahimi M, Jazayeri SA. A detail study of a RCCI engine
performance fueled with diesel fuel and natural gas blended with syngas with
different compositions. Int J Hydrogen Energy 2022;47(36):16283–96.
[28] Xu Z, Jia M, Xu G, Li Y, Zhao L, Xu L, et al. Potential for reducing emissions in
reactivity-controlled compression ignition engines by fueling syngas and diesel.
Energy Fuels 2018;32(3):3869–82.
[29] Guo X, Chen Y, Huang H, Chen Y, Liu M, Lei H, et al. Development of a diesel/
natural gas mechanism model for the CFD simulation of dual-fuel engine. ACS
Omega 2021;6(33):21543–55.
[30] Ebrahimi M, Jazayeri SA. Effect of hydrogen addition on RCCI combustion of a
heavy duty diesel engine fueled with landfill gas and diesel oil. Int J Hydrogen
Energy 2019;44(14):7607–15.
[31] Mabadi Rahimi H, Jazayeri SA, Ebrahimi M. Hydrogen energy share enhancement
in a heavy duty diesel engine under RCCI combustion fueled with natural gas and
diesel oil. Int J Hydrogen Energy 2020;45(35):17975–91.
[32] Reitz RD. Modeling atomization processes in high-pressure vaporizing sprays.
Atomisation Spray Technology 1987;3(4):309–37.
[33] Arvidson M, Fakley ME, Spencer MS. Lithium halide-assisted formation of
polyoxymethylene dimethyl ethers from dimethoxymethane and formaldehyde.
J Mol Catal 1987;41(3):391–3.
1328
International Journal of Hydrogen Energy 54 (2024) 1319–1328
[34] Uchida T, Kurita Y, Kubo M. The dipole moments and the structure of
polyoxymethylene dimethyl ethers. J Polym Sci 1956;19(92):365–72.
[35] Turns Stephen R. Introduction to combustion, vol. 287. New York, NY, USA:
McGraw-Hill Companies; 1996.
[36] Kulkarni A, García EJ, Damone A, Schappals M, Stephan S, Kohns M, et al. A force
field for poly(oxymethylene) dimethyl ethers (OMEn). J Chem Theor Comput
2020;16(4):2517–28.
[37] Kang M-R, Song H-Y, Jin F-X, Chen J. Synthesis and physicochemical
characterization of polyoxymethylene dimethyl ethers. J Fuel Chem Technol 2017;
45(7):837–45.
[38] Navid A, Khalilarya S, Abbasi M. Diesel engine optimization with multi-objective
performance characteristics by non-evolutionary Nelder-Mead algorithm. Sobol
sequence and Latin hypercube … Fuel 2018;(228):349–67.
[39] Mobasheri R. Analysis the ECFM-3Z combustion model for simulating the
combustion process and emission characteristics in a HSDI diesel engine. Int J
Spray Combust Dyn 2015;7(4):353–71.
[40] Dukowicz JK. Quasi-steady droplet phase change in the presence of convection.
1979.
[41] Dukowicz JK. A particle-fluid numerical model for liquid sprays. J Comput Phys
1980;35(2):229–53.
[42] Eckart S, Cai L, Fritsche C, vom Lehn F, Pitsch H, Krause H. Laminar burning
velocities, CO, and NOx emissions of premixed polyoxymethylene dimethyl ether
flames. Fuel 2021;293:120321.
[43] Konnov AA, Mohammad A, Kishore VR, Kim NI, Prathap C, Kumar S.
A comprehensive review of measurements and data analysis of laminar burning
velocities for various fuel+air mixtures. Prog Energy Combust Sci 2018;68:
197–267.
[44] Zhang G, Hung DL. Temporal investigations of transient fuel spray characteristics
from a multi-hole injector using dimensionless analysis. Exp Therm Fluid Sci 2015;
66:150–9.
[45] Internal combustion engine fundamentals. 2018.