Compression

FULL PAPER
www.entechnol.de
Pyrolysis of Methane and Ethane in a Compression–

Expansion Process as a New Concept for Chemical Energy
Storage: A Kinetic and Exergetic Investigation
Charlotte Rudolph* and Burak Atakan
storage (EES) in applications such as

The production of chemical energy carriers utilizing electrical energy from (super-) capacitors or superconducting
renewable sources is essential for the future energy system. A motored piston magnetic energy storage (SMES). These
engine may be used as a reactor to convert mechanical to chemical energy by the types of energy storage provide high
efficiencies up to 95%, a fast response
pyrolysis of methane and ethane; this is analyzed here. The piston engine is
within milliseconds, and a long cycle life
modeled as a compression–expansion cycle with detailed chemical kinetics. The (106 cycles). Their main problem is a short
main products are hydrogen and high-energy hydrocarbons such as acetylene, storage period (seconds or hours) and a low
ethylene, and benzene. To reach the required high temperatures for conversion energy density (<5 Wh kg1) and capac-
after compression, the educt is diluted with argon. The influence of the operating ity.[4,5] Also battery energy storage is
well-known for medium power capacities
conditions (temperature, pressure, dilution) on the product gas composition, the
and short response times and a short
stored exergy, and the ratio of exergy gain to work input (efficiency) is investi- lifetime but high efficiencies of up to
gated. A conversion of >80% is predicted for an argon dilution of 93 mol% at 85%, depending on the type of battery.[6]
inlet temperatures of 573 K (methane) and 473 K (ethane), respectively. A storage Therefore, these short-term energy storage
power of 7.5 kW (methane) and 6 kW (ethane) for a 400 ccm four-stroke single- methods are a good option for short-term
cylinder is predicted with an efficiency of 75% (methane) and 70% (ethane), energy compensation.[7] A choice for
medium-term energy storage is the com-
respectively. Conditions are identified, where high yields of the target species are
pressed air energy storage (CAES), which
achieved, and soot formation can be avoided. uses excess energy to compress air and
stores it in caverns afterward. CAES has
a storage duration of hours to days at
medium energy densities (<60 Wh kg1)
[3]
1. Introduction and capacity; they have a fast response time and, depending
on their design, efficiencies of 60–90%.[3] Their major disadvan-
The future energy demand will be covered by renewable ener- tage is that certain geological requirements limit their general
gies. Their weather dependence results in a fluctuating electrical usage.[8,9] Another medium-term energy storage technology is
energy production: Energy surpluses or deficits arise. To provide the pumped hydro energy storage (PHES). PHES uses the poten-
a stable and flexible energy supply, the topic of energy storage is tial energy of water, which increases, when water is pumped
becoming increasingly important. To store energy for later usage, from a lower level to a higher level. In contrast to CAES,
several methods of energy storage exist, which differ in storage PHES is developed to commercial maturity with efficiencies
capacity and storage duration and in the type of stored energy.[1–3] of up to 85% with high capacities and a storage duration of
The most direct way to store electrical energy is electrical energy up to months,[10,11] but is also limited by geological restrictions.
Power-to-Gas (PtG) or Power-to-Liquid (PtL) do not only offer a
long-term energy storage but also a cross-sector energy compen-
C. Rudolph, Prof. B. Atakan sation. These concepts use excess energy to produce hydrogen
Institute for Combustion and Gas Dynamics – Thermodynamics
University of Duisburg-Essen
first and synthetic methane or ammonia afterward, which can
Lotharstraße 1, 47057 Duisburg, Germany be stored with a high energy density in caverns, tanks, pipelines,
E-mail: [email protected] or in chemical compounds.[2,3] If needed, the stored gas can be
The ORCID identification number(s) for the author(s) of this article
converted back to electricity using fuel cells, gas turbines or
can be found under https://doi.org/10.1002/ente.202000948. engines,[12] or converted to liquid fuels such as higher alkanes
via the Fischer–Tropsch synthesis.[13]
© 2021 The Authors. Energy Technology published by Wiley-VCH GmbH.
This is an open access article under the terms of the Creative Commons
Hydrogen can be produced in different ways. A distinction is
Attribution-NonCommercial-NoDerivs License, which permits use and made between methods, which need fossil fuels and those that
distribution in any medium, provided the original work is properly are based on renewable energies,[14] as e.g., electrolysis of water.
cited, the use is non-commercial and no modifications or adaptations Hydrogen is mainly produced by steam methane reforming
are made. (SMR) because the cost for natural gas is low and the technolo-
DOI: 10.1002/ente.202000948 gies are already developed for large scale. SMR is the catalytic
Energy Technol. 2021, 9, 2000948 2000948 (1 of 16) © 2021 The Authors. Energy Technology published by Wiley-VCH GmbH
21944296, 2021, 3, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/ente.202000948 by Universiteitsbibliotheek Gent, Wiley Online Library on [13/04/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
www.advancedsciencenews.com www.entechnol.de
reaction of methane and water to hydrogen and carbon dioxide at trends. In the shock tube study, in which the decomposition of
relatively high temperatures (>900 K) and pressures (3–25 bar) methane was examined at different temperatures and very short
according to Equation (1)).[15–17] residence times, a methane conversion of up to 80% at a temper-
ature of 2400 K was observed. In the flow reactor study with com-
CH4 þ 2H2 O ⇆ CO2 þ 4H2 Δg 0 ¼ 114 103 J mol1 (1) parably long residence times, a methane conversion of up to 85%
could already be reached at temperatures of 1500 K. In both stud-
After the carbon dioxide removal, hydrogen is obtained with a
ies, the major products are acetylene, ethylene, and benzene but
purity of 99.99%. Overall, the process can achieve a thermal
also soot is formed. Especially, long residence times lead to a
efficiency (based on the higher heating values) of 85% and the
high amount of soot as it is expected in equilibrium. As not only
hydrogen production costs for the complete process are compa-
methane but also natural gas is used for such processes, the
rably low.[14,15,18]
pyrolysis of ethane is also considered. The ethane conversion
The electrolysis of water is a method that uses water and
starts at temperatures of 1100 K. The main products are ethylene
renewable energy sources. One benefit of this method is, that
at lower temperatures of 1300 K and acetylene and methane at
no carbon dioxide is produced and therefore no post-treatment
higher temperatures >1500 K, following the net reactions
is necessary. Although the efficiencies are similar to SMR, they
(Equation (6)–(8)) below.[33,34]
are not competitive as the large-scale plants are technically not
matured.[19,20] A carbon-dioxide-free method to produce hydro- C2 H6 ⇆ C2 H4 þ H2 Δg 0 ¼ 100 103 J mol1 (6)
gen with very low costs is the decomposition of methane, accord-
ing to (Equation (2)). C2 H6 ⇆ C2 H2 þ 2H2 Δg 0 ¼ 241 103 J mol1 (7)
CH4 ⇆ CS þ 2H2 Δg 0 ¼ 50.8 103 J mol1 (2) C2 H6 þ H2 ⇆ 2CH4 Δg 0 ¼ –69 103 J mol1 (8)
In this process, methane is converted to hydrogen and solid Internal combustion (IC) engines, running in homogeneous
carbon by pyrolysis. The needed catalysts for this reaction as well charge compression ignition (HCCI) mode, can also be used as
as the temperature range has been studied many times.[21–24] chemical reactors for partial oxidation with respect to flexible
In early studies, Muradov[21,25] found out, that metal or metal and versatile energy conversion in polygeneration processes,
oxide catalysts are very advantageous, yielding hydrogen purities which was investigated theoretically[35] and experimentally.[36,37]
in hydrogen–methane mixtures of >80% at temperatures above Engines can also be considered for pyrolysis processes, where the
1100 K, which is very close to the chemical equilibrium. The endothermal reactions are initiated by the high temperatures at
additionally produced solid carbon is used industrially or can the top dead center (TDC) after compression. In addition, the
be buried in the ground. The disadvantage of this method is mixture is diluted with an atomic inert gas to further increase
the comparably low thermal efficiency (based on the higher heat- the temperatures and to improve the conversion. The enthalpy
ing values) of 35–50%,[14] if only hydrogen is sought as product. of the cylinder charge increases due to work input during the
The overall efficiency of the process can be increased, if not only compression stroke. Therefore, the product gas contains species
hydrogen and carbon but also larger amounts of higher hydro- with a higher enthalpy and exergy than the educt gas. It results
carbons are produced in the methane pyrolysis.[26–30] that the input energy, possibly from excess renewable energy, is
Bartholomé[26] developed a method in 1954 for acetylene pro- stored in the higher-energy product gas. Possible applications for
duction from methane in which the thermal decomposition of the product gas are: 1) utilization as a raw material for chemical
methane is initiated by oxidation and, after a very short reaction synthesis, 2) feed into the natural gas grid to increase its heating
time (102 s), stopped by cooling and therefore quenching the value, 3) combustion to generate electricity, or 4) storage and
mixture. The methane pyrolysis toward higher hydrocarbon and later use. The basic principle is shown in Figure 1.
hydrogen formation was investigated several times,[27–30] and it
is known that the endothermal reactions starts at temperatures
of 1300 K, producing mainly benzene at these temperature. At
higher temperatures (>1500 K), mainly acetylene and ethylene
are formed according to the net reactions (Equation (3)–(5)).
CH4 ⇆ 0.5C2 H2 þ 1.5H2 Δg 0 ¼ 155 103 J mol1 (3)
CH4 ⇆ 0.5 C2 H4 þ H2 Δg 0 ¼ 85 103 J mol1 (4)
CH4 ⇆ 0.167 C6 H6 þ 1.5 H2 Δg 0 ¼ 72 103 J mol1 (5)
Depending on reaction time and temperature, the main

species are acetylene, ethylene, and benzene always together with
a high amount of hydrogen but also species such as propene,
propane, 1,3-butadiene but polycyclic aromatic hydrocarbons
(PAHs) and soot are formed as well. Experimental investigations
of methane pyrolysis by Nativel et al.[31] in a single-pulse shock
tube and by Keramiotis et al.[32] in a flow reactor showed the same Figure 1. Basic principle of the process.
" #
Atakan[38] optimized a compression–chemical equilibrium– sP π cosðθÞ
¼ sinðθÞ 1 þ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (9)
expansion process for pyrolysis of methane, ethane, and hydro- s̄P 2 R2 sin2 ðθÞ
gen to defined expected species within initial temperature and
pressure, compression ratio, and argon dilution. The target where sp is the mean piston velocity and R is the connecting rod
variable of the optimization was, e.g., a maximum stored exergy. length to crank radius ratio. The rotation speed converts the
He only used thermodynamic equilibration, without regarding crank angle dependency into a time dependency. While piston
the chemical kinetics. The optimization predicts a maximum speed and volume change, pressure and temperature also
stored exergy at the upper bounds of inlet temperature, compres- increase as a function of time. Ideal gas behavior is assumed
sion ratio and argon dilution, and the lower bounds of inlet for all changes of state. For every time step, the balances for
pressure. Thus, the pyrolysis process appeared to be a feasible species and energy conservation are solved. The time integration
approach. An increase in exergy of more than 11% with exergetic for the species and energy balances is solved using a stiff ordinary
efficiencies of up to 92% were predicted producing chemicals, differential equation (ODE) solver, namely SUNDIALS.[41]
such as acetylene, ethylene, and benzene. However, ethane in The used absolute and relative tolerances are 1021. The equation
the equilibrium study seemed not to be very promising as in for the conservation of species k (Equation (10)) is
equilibrium it reacts exothermically to methane. A recent review
dyk :
contains experimental results for methane pyrolysis in a rapid m ¼ V ωk M k (10)
compression machine.[39] Methane was diluted with 90% argon dt
and 5% helium, so the mixture contains also only 5% methane. for a constant total mass m and without surface reactions, where yk
:
For a compression pressure of 18 bar and compression temper- is the mass fraction, ω is the molar production rate, and Mk is the
atures of 1450–1750 K, the conversion of methane was between molecular weight of each species k.
2.5% and 20%. The measured species were H2, C2H2, and C2H4 The energy conservation (11) calculates the change in internal
with yields of up to 12% (H2), 5% (C2H2), and 2% (C2H4), energy due to transfer of work and heat for a closed system.
respectively.
The kinetics of the conversion of methane and ethane to dU dV :
¼ p þQ (11)
higher energy fluids through pyrolysis in a piston engine is inves- dt dt
tigated theoretically in the present work with respect to energy :
p is the pressure and Q the heat (loss) rate. The heat transfer
storage, as they are the main components of natural gas.
through the cylinder walls is calculated from Newton’s law of cooling
The endothermal reactions are initiated at high temperatures
(Equation (14)), using the Woschni correlation (Equation (12)),[42]
at the TDC due to compression stroke and argon dilution.
to estimate the convective heat transfer coefficient α
A simplified time-dependent piston engine is modeled, only
regarding the compression and expansion strokes. A detailed α ¼ C d0.2 p0.8 T 0.53
chemical reaction mechanism is used, as described in the fol- " #0.8
lowing section. A comprehensive parameter study is performed VD Tr (12)
C 1 s̄P þ C2 ðp pm Þ
to find feasible operating points within a reasonable amount of V r pr
storable exergy and efficiency. In addition to exergy, the kinetics
are also investigated, analyzing the educt conversion, the prod- where VD is the displacement volume, Tr, pr, and Vr are tem-
uct gas composition, and the crank-angle dependent reaction perature, pressure, and volume at a reference state, and C and
path. The main aim is to analyze, whether the in-cylinder C1 are constants according to Woschni. C2 is set to zero as no
process is thermodynamically controlled or if the kinetics combustion occurs and thus the actual pressure p and the
can lead to different conclusions with respect to the product motored pressure pm are similar to each other. The resistances
formation and the thermodynamics, than those made by for heat conduction through the cylinder walls and the convec-
Atakan.[38] tive heat transfer on the cooling water side were neglected, as
their heat-transfer resistances are significantly smaller than the
one for convection to the inner cylinder wall.
2. Modeling The chemical reactions are simulated with the complete
Polimi elementary reaction mechanism,[43–46] including 484 spe-
The piston engine is simulated as a time-dependent single- cies and 19 341 reactions including species up to C20, lumped
zone model with detailed chemical kinetics in Cantera within species and chemistry of PAHs, because the formation of soot
Python.[40] The engine is modeled as a closed system, because or soot precursors were expected. The mechanism is validated
only the compression and expansion strokes are considered. for pyrolysis, partial oxidation and combustion of hydrocarbons
The complete engine cycle consists of one rotation of the crank, at low and high temperatures, and soot formation in (laminar)
360 , starting with the piston movement from bottom dead flames and in jet-stirred and flow reactors. Other reaction mech-
center (BDC) to TDC (compression stroke) and back from anisms for this purpose could be the Aramco 3.0[47] or the
TDC to BDC (expansion stroke). A resolution of a tenth of a PolyMech[48] but were not chosen here, because the Polimi is
degree is selected. The cylinder volume V is determined from more detailed with respect to the PAH-submechanism. The
the crank-angle (θ) dependent position of the piston. The PolyMech, in particular, contains only species up to C6 and does
change in the cylinder volume is therefore dependent on the not contain C6 decomposition reactions. Although the choice of
piston velocity sp which is calculated from Equation (9) mechanism is important for this investigation, the main
Table 1. Engine parameters used for the simulation. engines, it is already used in multi-stage piston compressors.
Thus, in principle, such a high inlet pressure can be realized,
Parameter Specification Unit even if the design of such a piston engine has to be adapted.
Displacement VD 4 10 4
m3 In this study, it is shown to what extent these unusual conditions
3 prove to be favorable in terms of thermodynamics and kinetics.
Bore d 79.5 10 m
Methane and ethane are used as fuels, or better as educts as
Stroke st 80.5 103 m
they are the main components of natural gas. The educts are
Compression ratio ε 22 – diluted with an atomic inert gas to reduce the heat capacity of
Connection rod/crank radius R 3.5 – the mixture, which leads to a higher temperature at the end
Rotation speed 3000 RPM of compression stroke, and thus, to a higher educt conversion.
Coolant temperature TC 373 K
To describe and evaluate the process, a thermodynamic and a
kinetic analysis is carried out. The thermodynamic analysis
includes the calculation of the supplied work W and the heat loss
outcome is not influenced as tested for some conditions. With Q according to Equation (13) and (14)
different mechanism, the mole fractions and the product distri- Z
bution change to some extent, but the main tendencies remain W ¼ pdV (13)
unchanged, also the thermodynamic aspects and, thus, the
study’s outcome. :
Q ¼ α A ðT c TÞ (14)
The chosen engine parameters are listed in Table 1 and cor-
respond to typical stationary engine parameters, but with larger and the calculation of the irreversible entropy production Sirr and
rotation speed and compression ratio, because this is more favor- the exergy loss Eloss according to Equation (15) and (16). The
able for the storage process. A high compression ratio leads to a European sign convention is used for the work, added work
higher temperature increase and thus a higher temperature at has a positive value.
TDC. In addition, the compression ratio is a low-cost method
to increase TDC temperatures and pressures, compared to Q
Sirr ¼ m ðsProduct sEduct Þ (15)
increased inlet conditions that required an additional energy TC
input. Thus, a realistically high compression ratio of 22:1 was
E loss ¼ T u Sirr (16)
chosen, similar to recent HCCI engine investigations,[49,50]
and was not varied. In the present work, an effective compression As a result of the endothermal reaction, the product gas com-
ratio is assumed as the single-zone model does not consider position changes, which leads to a change in the internal energy.
uncertainties due to crevices, blow-by, or cold cylinder walls. The chemical and physical exergies of the mixture are chosen to
A high rotation speed leads to shorter reaction times, which describe the energetic value of the product gas mixture and to
reduces the production of carbon due to quenching effects determine the storable exergy. The chemical exergy is calculated
during the expansion stroke. This effect is explained in detail according to Atakan.[38] As described there, the gas mixture,
in the Section 3.1. The coolant of the cylinder walls is water. whose exergy is to be determined, is chemically equilibrated with
As no combustion takes place in this process, a reduction in surrounding air in the ratio 1/f with f ¼ 106. Assuming, the sur-
the cooling power using air as a coolant would also be conceiv- rounding air is at standard conditions (298.15 K and 1.01325 bar)
able. This could lead to a more homogeneous temperature dis- and has a relative humidity of 70%, according to ref. [52], the
tribution and an increased reactant conversion. chemical specific exergy echem is calculated using
In addition to the geometric parameters, the varied process
parameter ranges for initial temperature T0, initial pressure echem ¼ ðhsur þ f hair ðf þ 1Þ heq Þ T u
p0, and Ar mole fraction are listed in Table 2. The inlet tempera- (17)
ðssur þ f sair ðf þ 1Þ seq Þ
ture was chosen in a range of 323–573 K and it is assumed that
the temperature between inlet valve opening (IVO) and BDC where h is the enthalpy and s is the entropy, the subscript sur is
remains almost constant as it is not a fired operation. used for the mixture at surrounding (standard) conditions; the
Especially, the upper boundary of the inlet temperature seems subscript air is used for air, which is described earlier and the
practicable according to Lim et al.,[51] where spark-ignition subscript eq is used for the chemically equilibrated mixture.
engine experiments were performed with inlet temperatures up The physical specific exergy for an open system is calculated using
to 750 K. The inlet pressure was chosen in a range of 1–10 bar.
Although an inlet pressure of up to 10 bar is unusual for IC ephys ¼ ðh hsur Þ T u ðs ssur Þ (18)
Thus, the sum of the specific chemical and the physical exergy
Table 2. Process parameters used for the simulation.
describes the total specific exergy of the mixture.
Parameter Range Unit emixture ¼ echem þ ephys (19)
Initial temperature T0 323–573 K
Initial pressure p0 1–10 bar
The total increase in the exergy ΔE, and consequently, the total
amount of energy which can be stored is the difference of prod-
Argon dilution 85–99 mol%
uct and educt-specific exergies multiplied with the mass.
ΔE ¼ m ðeProduct eEduct Þ (20)
The storage power for a four-stroke engine is then calculated

using
N
P ¼ ΔE s : (21)
2 60 min
To describe the efficiency of the process, the storage efficiency

η is calculated from the increase of exergy to supplied work ratio.
ΔE
η¼ (22)
W
The kinetic analysis is done by analyzing the product gas com-

position and the reaction path. Furthermore, the conversion of
the educt is calculated using Equation (23) and yields of species
Figure 2. Temperature (left axis, solid lines) and methane conversion
of interest are calculated using Equation (24) (right axis, dashed lines) as a function of crank angle for different inlet
nE,0 nE temperatures and argon dilution: 323 K inlet temperature and without
XE ¼ (23) argon (black lines), 373 K inlet temperature and 95% argon (blue lines),
nE,0
and 423 K and 95% argon (red lines).
nP nP,0 νP
YP ¼ (24)
nE,0 νE conversion. To achieve a reasonable conversion, e.g. 80%, the
needed TDC temperature can be reached by varying the inlet
where n is the molar amount of substance, the indices E and P
temperature or by varying the inlet argon mole fraction.
describe the educt or the product and the index 0 describes the
The argon dilution affects directly the fuel content and the heat
inlet state. The parameter ν is the number of C or H atoms of the
capacity of the mixture and therefore the temperature increases
species to be calculated. The yields of hydrocarbons are referred
in the compression stroke and at the end of the compression
to the C-atoms and the yield of the hydrogen is referred to the
stroke. The inlet temperature affects the complete charge in
H-atoms.
the cylinder and the supplied work, and as well as the inlet argon
In addition, a reaction path analysis is performed using
mole fraction, the TDC temperature. The results of the variation
Cantera[40] for methane and ethane pyrolysis to identify the
of the inlet argon mole fraction and the inlet temperature on the
most important reactions responsible for the products and the
conversion are shown in Figure 3 for methane (left) and ethane
differences in product gas composition. For each reaction and
(right). The conversion of both, methane and ethane, increase
species involved, the net production rate, consisting of forward
with inlet temperature and inlet argon mole fraction.
and reverse reaction rates, is calculated for a specified time during
However, to keep the input temperatures within a realistic range,
the engine cycle with the Cantera module. Based on the net
argon dilution is necessary. Thus, an increasing inlet argon mole
production rates, the carbon fluxes between two or more species
are then graphically displayed, with the largest carbon fluxes being fraction shifts the inlet temperature to lower values for a wanted
the most important reactions for the selected time. conversion due to the achievable TDC temperature and even the
lowest inlet temperature of 323 K leads to a range of the conver-
sion between <10% and 100% depending on the inlet argon
3. Results and Discussion mole fraction. In addition, the TDC temperatures for different
combinations of inlet argon mole fractions and inlet tempera-
First, the effect and the necessity of the argon dilution on the gas tures are shown as isothermal curves. The conversion of the
temperature and the methane conversion during compression educts increases with the TDC temperature. The comparison
and expansion stroke is shown in Figure 2. If only methane with of methane and ethane shows that the needed TDC temperatures
a common inlet temperature of 323 K is compressed, the temper- are 400 K lower for ethane than for methane at the same con-
ature at TDC reaches 650 K and no methane is converted version. For a conversion of 12%, a TDC temperature of 1600 K
(black dashed line). If the educt is now highly diluted with argon, (323 K < T0 < 450 K) is necessary for methane, but only 1200 K
the heat capacity of the mixture is strongly decreased and (323 K < T0 < 473 K) for ethane. A similar temperature differ-
the temperature after compression increases. For the same inlet ence is also observed for a conversion of e.g. 80%: for methane,
temperature of 323 K, the initial mixture consists of 95% argon a TDC temperature of at least 1800 K is needed, whereas for eth-
and 5% methane (blue lines), a temperature of 1650 K is reached ane 1400 K is sufficient. The reason for the lower TDC tempera-
at the end of compression stroke and 37% of the methane is con- ture is explained in a later section. In addition, the TDC
verted. An additional small increase in the inlet temperature to temperatures and conversion rates at these operating conditions
423 K for the diluted mixture (red curves) leads then to a temper- are shifted toward lower inlet argon mole fractions due to the
ature of 1800 K at TDC and to a methane conversion of 77%. higher density of ethane.
The conversion of the educts mainly depends on the TDC tem- Although only the influence of inlet temperature and argon
perature and the highest TDC temperatures result in the highest dilution on the educt conversion is shown here, there are also
Figure 3. Left: CH4 conversion (black lines, colored area) and TDC temperature (white lines) for methane pyrolysis as a function of inlet temperature and
inlet argon mole fraction for an inlet pressure of 1 bar. Right: C2H6 conversion (black lines, colored area) and TDC temperature (white lines) for ethane
pyrolysis as a function of inlet temperature and inlet argon mole fraction for an inlet pressure of 1 bar.
effects on the product gas composition and the stored exergy, as be seen in Figure 4, that physical-exergy losses (calculated from
described in Section 3.4. Equation (15)) are negligible in this process as they are less than
3% of the expended exergy (work and chemical exergy of the
3.1. Process Details mixture). This is due to a seven times higher chemical exergy
compared to physical exergy and supplied work in the system.
To get a deeper insight into the process, in the following section, Although the physical exergy decreases during the compression
both the course of energetic quantities related to the crank angle stroke due to entropy production, the chemical exergy in the mix-
and the conversion of the educt to the target species are analyzed. ture increases due to the endothermal reactions. The physical
Eloss,phys
As it is known from typical combustion engines, work is added exergetic efficiencies calculated from ηex,phys ¼ 1 ðWþE Edukt Þ
for
during the compression stroke and removed in the expansion every operating point are >97%, which once again illustrates
stroke and overall; the work removed is higher due to the com- how small the physical exergy losses are in this process. As
bustion and the integral to calculate the total work is negative; the physical exergetic efficiency is not very revealing with respect
work is transferred from the system. This process is now to exergy storage, the efficiency in further sections is calculated
completely different. During the compression stroke, work is still according to Equation (22), which shows how much of the
added, but the amount of work that is added is higher than the expended work is converted into chemical exergy of the mixture.
work removed in the expansion stroke due to the endothermal The more of the educt is converted toward higher exergy chem-
reactions; this is shown in Figure 4. Both, heat losses and exergy icals per supplied amount of work, the higher is the efficiency.
losses are mainly high in the range of the TDC, and as soon as Compared to the physical exergetic efficiency, the efficiency
the mixture is cooled down through the expansion stroke, the according to Equation (22) is noticeably lower (here, 58%)
heat-transfer rate drops and so do the exergy losses. It can also because of chemical exergy losses due to the increased specific
entropy after reaction. In comparison, the compression of meth-
ane requires, due to the needed higher temperatures, more work
and results in larger heat losses than the compression of ethane,
although there is a larger mass of ethane in the cylinder due to its
higher density and higher molecular weight. The smaller amount
of work addition in ethane pyrolysis is mainly caused by the
lower isobaric and isochoric specific-heat capacities and their
lower ratio for ethane, leading to a lower temperature and pres-
sure increase. To give a numerical example of the operating point
at T0 ¼ 573 K, p0 ¼ 1 bar, and xAR ¼ 0.93, the compression and
expansion of methane requires 69 J of work and 33 J of heat
is transferred; the compression of ethane requires only 59 J of
work, whereas 25 J of heat is transferred.
The conversion of methane and the formation of target species
as well as the temperature trace are shown in Figure 5 (left) as a
function of crank angle for the same exemplary operating point
Figure 4. Work (black lines), heat loss (red lines), and physical exergy loss at T0 ¼ 573 K, p0 ¼ 1 bar, and xAR ¼ 0.93. The recognizable
(blue lines) for methane (solid lines) and ethane (dashed lines) as a func- consumption of methane starts 28 before TDC at 1620 K and
tion of crank angle for an inlet temperature of 573 K, an inlet pressure of 25 bar (1% CH4 conversion). Thereafter, at 25 before TDC, at
1 bar, and an inlet argon mole fraction of 0.93. 1680 K, 29 bar, and 5% CH4 conversion, a reduction in the
Figure 5. Left: Mole fractions (left axis, solid lines, colored), temperature (right axis, dashed lines, black) and temperature of the nonreacting mixture
(right axis, dashed lines, gray) as a function of crank angle for an inlet temperature of 573 K, an inlet pressure of 1 bar, and an inlet argon mole fraction of
0.93 for methane as reactant. Right: Mole fractions (left axis, solid lines, colored), temperature (right axis, dashed lines, black) and temperature of the
nonreacting mixture (right axis, dashed lines, gray) as a function of crank angle for an inlet temperature of 573 K, an inlet pressure of 1 bar, and an inlet
argon mole fraction of 0.93 for ethane as reactant.
temperature rise is observed due to the endothermal reactions, TDC (1270 K and 18 bar), 12% of the ethane is consumed and
compared to a nonreacting mixture. This reduction in the a smaller temperature increase is observed, because of the fast
temperature increase is also seen for the following crank angles. formation of ethylene at this point. Nevertheless, the reduction in
Up to this point, mainly ethylene together with a high amount of temperature increase, compared to a nonreacting mixture, is less
hydrogen is produced. At 10 before top dead center (BTDC), due pronounced in comparison to methane, because endothermal
to the increase in temperature to 1800 K, ethylene is first decom- reactions are counterbalanced by exothermal methane
posed to acetylene, which is then further converted to benzene formation. At higher temperatures, 20 BTDC to 20 ATDC, acet-
and light PAHs such as naphthalene and acenaphthylene. At ylene and benzene are formed: a part of the ethylene is con-
TDC (75 bar), the temperature reaches 1850 K, with a maximum sumed to form acetylene, which in turn to a small extent is
of 86% methane-consumption to acetylene and benzene. further converted to benzene. The maximum temperature and
Immediately after the TDC, while temperature is still high, pressure at the TDC are 1465 K and 60 bar, which are comparably
acetylene and benzene decompose and form heavy PAHs, low and cause the absence of PAH and soot particle formation,
e.g., C20H10 and C20H16 and soot particles. At 20 after TDC, although their precursors, acetylene and benzene, are present.
the temperature and pressure decrease to 1350 K and 30 bar, At 20 after TDC, temperature and pressure drop to 1110 K
respectively. Now all reactions are relatively slow, and the mix- and 25 bar and the mixture is frozen again. At the end of the
ture is frozen and products such as acetylene survive. At the expansion stroke, nearly all the ethane is consumed (97%)
end of one cycle, at BDC, both, temperature and pressure are and temperature and pressure decrease to even lower values
below (325 K, 0.6 bar) the initial values (573 K, 1 bar) due to of 303 K and 0.56 bar due to the endothermal reactions and
the endothermal reactions and the conversion of physical exergy the large educt conversion. The main products are not only
to chemical exergy. Overall, 85% of methane is converted. This is hydrogen and ethylene but also smaller amounts of acetylene
below the maximum conversion, due to backreactions to meth- and benzene are observed with yields of 27%, 40%, 14%, and
ane. The main products are hydrogen, acetylene, ethylene, and 5%, respectively. Although there is no PAH and soot particle
benzene with yields of 63%, 29%, 7%, and 5%, respectively, formation, 20% of the carbon of ethane remains in methane,
together with high amounts of light and heavy PAHs with yields which is energetically less valuable. Although the chemical exergy
of up to 32%. Yields of species that are represented in only very has increased by 128.5 kJ kg1, the physical exergy has decreased
small quantities and unconverted educts as well were not by 35.9 kJ kg1. In total, the storage power and the efficiency are
calculated here. The total specific exergy of the mixture is 0.79 kW and 54%, which is lower than the storage power of meth-
increased by 118.7 kJ kg1, with the chemical exergy increased ane in the same operating point and even the educt conversion is
by 149.9 kJ kg1 and the physical exergy decreased by much higher. The lower storage power for ethane is caused by
31.2 kJ kg1. Thus, the storage power for this process is the lower work, that is supplied and by a different distribution
0.99 kW, and if one relates this to the work expended, the of the main products: less hydrogen and acetylene but more
efficiency is 58%, meaning that 58% of the work is stored in ethylene. This will be discussed in more detail in Section 3.4.
the exergy of the mixture.
If ethane is used as educt, the process proceeds basically sim-
ilar and is shown in Figure 5 (right) for the operating point at 3.2. Heat-Transfer Models
T0 ¼ 573 K, p0 ¼ 1 bar, and xAR ¼ 0.93. One percent conversion
of ethane is found at 37 before TDC at 1180 K and 12 bar; this The heat-transfer model has a significant influence on the dissi-
is at an earlier crank angle and at a lower temperature, due to the pated heat, temperatures, and pressures in the cylinder. Thus,
lower activation energy of the ethane pyrolysis. At 30 before also the educt conversion and the temperature/pressure
dependent product formation is also affected. To show the influ- temperature is 1930 K, obtained by the model of Chang and
ence on such parameters, the heat-transfer models from the minimum TDC temperatures are 1790 and 1800 K, obtained
Woschni,[42] Hohenberg,[53] Annand,[54] and Chang et al.[55] were by the models of Hohenberg and Annand, respectively. Although
compared. The heat-transfer model from Woschni, Hohenberg, the relative deviation from the average is also 5%, the absolute
and Annand are developed for spark ignition and diesel engines, deviation is up to 80 K. Such temperature differences affect
whereas the one from Chang is applicable for HCCI engines. As the endothermal reactions and the temperature traces during
none of these heat-transfer models has been validated for the the expansion stroke. The CH4 and H2 mole fractions are
present process, the results of such heat-transfer models are shown in Figure 6 (right) and are representative for educt con-
compared qualitatively. Although the correlations of Woschni, version and product generation. According to the maximum tem-
Hohenberg, and Chang are largely based on the temperature-/ perature and pressure, the CH4 conversion and H2 production,
pressure and volume curves, the correlation of Annand also obtained using the model from Chang, are highest, whereas
includes the dependence of the heat transfer on fluid properties lower values are predicted using the models from Hohenberg
such as the thermal conductivity and viscosity. The heat-transfer and Annand.
coefficients at BDC, e.g., for Woschni and Annand, are The storage power as well as the storage efficiency (Table 3)
60 W (m2 K)1 and 230 W (m2 K)1, respectively at T0 ¼ 573 K, are higher relative to the reference prediction from the Woschni
p0 ¼ 1 bar, and 93 mol% argon. At TDC, the heat-transfer model in accordance with the product yields for the model of
coefficients increase to values of 800 W (m2 K)1 (Woschni) Chang and lower for the models of Hohenberg and Annand.
and 1500 W (m2 K)1 (Annand). Although these differences The deviation from the average is up to 15% for the storage
are considerable, it turns out that the influence on the power and up to 30% for the efficiency because of a changed
process are not as large. Figure 6 (left) shows the pressure product spectrum. Although the CH4 conversion using
and temperature trace for an inlet temperature of 573 K; an inlet Hohenberg and Annand is significantly lower, benzene yields
pressure of 1 bar and 93 mol% argon. The pressure traces deviate have more than doubled due to the lower temperatures.
by maximum 5% in the TDC from the average. The maximum Overall, the heat-transfer models affect not only the temperatures
pressure in TDC (78 bar) is obtained using the model from and the educt conversion but also the product distribution. The
Chang and the minimum pressures in TDC (71 and 72 bar) results obtained with the different heat-transfer models are sum-
are obtained using the models from Hohenberg and Annand, marized in Table 3. As the results obtained by the heat-transfer
respectively. The deviation of the temperatures traces is more model from Woschni deviate least from the average regarding
distinct. Similar to the pressure curves, the maximum TDC temperature, pressure, CH4 conversion and storage power and
Figure 6. Left: Pressure (left axis, solid lines) and temperature (right axis, dashed lines. Right: CH4 mole fraction (left axis, solid lines) and H2 mole
fraction (right axis, dashed lines). Both as a function of crank angle for an inlet temperature of 573 K, an inlet pressure of 1 bar, and an inlet argon mole
fraction of 0.93. The simulations were performed with the heat-transfer model from Woschni[42] (black), Hohenberg[53] (red), Annand[54] (green), and
Chang[55] (blue).
Table 3. TDC temperature, TDC pressure, heat flux, CH4 conversion, yields of H2, C2H2 and C6H6, storage power and storage efficiency for different
heat-transfer models.
TDC TDC Heat CH4 H2 C2H2 C6H6 Storage Storage

temperature [K] pressure [bar] flux [kW] conversion [%] yield [%] yield [%] yield [%] power [kW] efficiency [%]
Woschni 1850 75 0.83 85 63 29 5 0.99 58

Hohenberg 1790 71 1.25 77 59 26 10 0.84 45
Annand 1800 72 1.11 79 61 27 9 0.88 49
Chang 1930 78 0.43 91 71 34 3 1.12 72
efficiency, we chose the Woschni correlation in the present reverse), which then leads to the production from ethylene.
investigation. The acetylene formation is preceded by the formation of C2H3
in the reaction of ethylene with methyl radicals (Equation (26))
3.3. Reaction Path Analysis C2 H4 þ CH3 ⇆ C2 H3 þ CH4 (26)
The pyrolysis of methane and ethane leads to different main Methyl radicals are produced by the H-abstraction of methane,
products. In this section, the formation paths of the different which is the start reaction of methane pyrolysis or by the uni-
carbon-containing products are discussed. Carbon reaction path molecular decomposition of ethane (Equation (27) and (28))
analysis is shown in Figure 7, for methane and ethane pyrolysis,
performed at 50% conversion, respectively. This point was cho- CH4 þ H ⇆ CH3 þ H2 (27)
sen for the reaction path analysis because all relevant reactions to
C2 H6 ðþMÞ ⇆ 2CH3 ðþMÞ (28)
interesting products take place there. For CH4, the conditions are
165 CA, 1750 K, and 47 bar, and for C2H6 they are 160 CA,
As the reaction of ethane to ethyl after hydrogen abstraction is
1350 K, and 29 bar. Both, methane and ethane, mainly produce
favored, methyl is produced less in ethane pyrolysis than in
hydrogen, whereas methane produces considerably more
methane pyrolysis, where it is part of the start reaction. Thus,
hydrogen due to the larger H/C ratio. For methane pyrolysis,
less methyl radicals are available to produce the intermediate
the mainly produced hydrocarbons are acetylene, ethylene,
C2H3, which is necessary for the path toward acetylene
benzene and soot precursors, whereas for ethane, it is methane.
(Equation (26)). A small amount of ethylene is still reacting to
The differences in the main product distributions are mainly due
C2H3 and methane. The temperature in the ethane pyrolysis
to different intermediates. The first main product formed is
is too low to consume the methane again, so that it remains
ethylene. In case of ethane pyrolysis, up to 55% of the carbon
as a product. Although less ethylene is present in methane pyrol-
remains in the product ethylene; for methane pyrolysis, it is sig-
ysis, significantly more of it is decomposed, so that significantly
nificantly less. Ethylene is formed from ethyl radical (C2H5)
more acetylene is also formed. In ethane pyrolysis, the path
decomposition (Equation (25))
toward benzene goes along C4H4 radicals, which are formed
C2 H5 ðþMÞ ⇆ C2 H4 þ HðþMÞ (25) from acetylene. C4H4 further reacts again with acetylene to
produce benzene. For methane pyrolysis, propyne (p-C3H4)
For ethane pyrolysis, ethyl is directly formed after and C3H3 radicals are formed fast, and benzene is then produced
H-abstraction from methane. In methane pyrolysis in turn, ethyl from C3H3 radicals. The consumption of benzene toward light
is formed from methyl (CH3) recombination (Equation (28), and heavy PAHs occurs at high temperatures in reactions with
(a) (b)
Figure 7. Reaction path for a) methane and b) ethane for an inlet temperature of 573 K, an inlet pressure of 1 bar, and an inlet argon mole fraction of 0.93,
performed at 50% conversion, respectively. The colors of the lines are related to the carbon fluxes according to the net reaction rates.
acetylene, C3H3 radicals, and CH3 radicals. Thus, due to the For ethane pyrolysis, on the other hand, the reaction
higher temperatures and the increased presence of C2H2, (Equation (32)) is the most important reaction with respect to
C3H3 and CH3, naphthalene (C10H8) and acenaphthylene C6H6 formation.
(C12H8) as well as the lumped species such as C20H10 and
C20H16 are products of the methane pyrolysis but not in the eth- C2 H4 þ C4 H5 ⇆ C6 H6 þ H þ H2 (32)
ane pyrolysis.
The formation of the lumped species is mainly sensitive to the
To identify the most sensitive reactions for the products of
reactions (Equation (33)) and (Equation (34)).
interest, a sensitivity analysis was performed and can be found
in the electronic Supporting Information. The most important C16 H10 þ H ⇆ C16 H9 þ H2 (33)
reaction for CH4 and C2H4 formation for methane pyrolysis is
the recombination of CH3 to C2H5 (Equation (29)). C16 H9 þ C2 H2 ⇆ 0.5C16 H10 þ H þ 0.5C20 H10 (34)
2CH3 ⇆ C2 H5 þ H (29)
3.4. Storage Power and Product Gas
The H-radical released in (Equation (29)) increases the conver-
sion of CH4 or C2H6. The ethyl radical C2H5 decomposes to
The influence of initial temperature, argon mole fraction, and
C2H4 and H. In contrast, the most important reaction for
pressure are investigated to find favorable operating points.
the decomposition of C2H4 and formation of C2H2 is the
Storage power and efficiency are shown in Figure 8 (left) for
H-abstraction from C2H4 by H radicals, forming C2H3 and
methane and ethane as a function of the inlet argon mole fraction
H2 (Equation (30)), for methane and ethane pyrolysis. The fact
for an inlet temperature of 573 K and an inlet pressure of 1 bar.
that the same reaction is responsible for both the formation
This condition was chosen with respect to the maximum storage
of C2H2 and the decomposition of C2H4 explains the trends
power at which the most useful products for methane pyrolysis
of the mole fractions in Figure 5.
are produced. For a comparison, most parameters should remain
C2 H4 þ H ⇆ C2 H3 þ H2 (30) unchanged; thus, this condition was also chosen for ethane
pyrolysis. Overall, methane and ethane have a similar course
The higher amount of H-radicals available in methane pyrol- of storage power and efficiency: for lower inlet argon mole frac-
ysis, due to the favorable H/C ratio, could support the formation tions (xAr < 0.88) both, storage power and efficiency are also low,
of C2H2. Regarding the formation of C6H6, the most important as the temperature increase during the compression stroke is
reaction with respect to methane pyrolysis is the recombination insufficient for a substantial educt consumption. With increasing
of propargyl radicals (C3H3). inlet argon mole fractions, the storage power of methane
increases strongly, whereas the storage power of ethane
2C3 H3 ðþMÞ ⇆ C6 H6 ðþMÞ (31) increases only slightly, which is a result of the overall higher con-
version of ethane due to the higher reactivity. Therefore, the max-
As argon is the main compound in the gas phase, its influence imum storage power of ethane is 0.8 kW with an efficiency of
as collision partner may be important and introduces some 54% and is at 92 mol% argon, while that of methane is at a
uncertainty. To investigate this, the collision efficiency of argon slightly higher argon load of 93 mol% but is significantly larger
in the most important third-body reactions for ethene and acety- (0.99 kW, 58%). The higher maximum is mainly due to the lower
lene formation were varied by a factor of 3, and the influence exergy of methane, compared to ethane and the distribution of
on the respective product gas was negligible; they changed the product species, which are also shown in Figure 8 (right).
by 0.4–2%. For the operating conditions with the maximum storage power,
Figure 8. Left: Storage power (black lines, left axis) and storage efficiency (red lines, right axis) as a function of inlet argon mole fraction for an inlet
temperature of 573 K and an inlet pressure of 1 bar for methane (solid lines) and ethane (dash lines). Right: Yields of main products as a function of inlet
argon mole fraction for an inlet temperature of 573 K and an inlet pressure of 1 bar for methane (solid lines) and ethane (dash lines).
the two main products are hydrogen and acetylene for methane increasing inlet pressures, with respect to the maximum storage
as educt, and ethylene and methane for ethane as educt, respec- power (compare Figure 8). First, the main products of methane
tively. Especially the high amount of methane in the product gas pyrolysis: hydrogen, acetylene, ethylene, and benzene are shown
of ethane pyrolysis explains the overall lower storage power com- in Figure 9a,b as a function of inlet temperature and inlet pres-
pared to methane pyrolysis. But not only the lower chemical sure. Figure 9c shows the species C20H10, representing PAHs, as
exergy due to the product gas composition affects the lower stor- a function of inlet temperature and inlet pressure. Regarding the
age power, also the lower temperatures and pressures over the temperature, the maximum yields for hydrogen and acetylene are
complete cycle and at the end of the cycle reduces the physical achieved at the maximum inlet temperatures and the maximum
exergy, which also influences the storage power. At high inlet yields for ethylene and benzene are attained at the lowest and
argon mole fractions, the storage power and the efficiency of medium inlet temperatures, as was already discussed in the pre-
both is decreasing again, first due to the small amount of educt vious section. An increasing inlet pressure leads to a higher
in the mixture and second due to change in the product gas amount of supplied work and a higher charge in the cylinder.
composition. Although hydrogen and acetylene yields increase Also, the TDC pressures and TDC temperatures rise up to
significantly, the yields of benzene and ethylene decrease. In 1100 bar and 2100 K, and the reaction rates rise, so that less
addition, heavy PAHs and soot precursors are formed at these hydrocarbons are formed but instead the equilibrium composi-
high temperatures. The storage power and the efficiency of tion according to reaction (Equation (2)) is progressively estab-
ethane has a local minimum between 0.96 < xAR < 0.98: lished. If such TDC pressures are reached, the technical
A second local maximum can be seen in xAR ¼ 0.97 as a result feasibility is not given. In a realistic piston engine, this is a limit-
of the sharp increase in the acetylene yield and simultaneous ing factor. Thus, yields for ethylene and benzene are reduced and
reduction in the methane yield, which compensates the lower shifted toward lower temperatures. Acetylene, even at low tem-
ethylene yield. peratures, is converted completely to C20H16, which is the largest
Considering now the course of the yields of the products as a hydrocarbon in the mechanism prior to solid carbon. At high
function of the inlet argon mole fraction, it turns out that the inlet pressures and low inlet temperatures, in addition to the
maxima of 1) ethylene and 2) benzene are at low and medium formation of C20H16 from acetylene, the exothermal reaction
inlet argon mole fractions (and thus, low and medium temper- back to methane is favored, due to the principle of Le
atures) and the maxima of hydrogen and acetylene are at the Chatelier. The carbon of methane, which is not in the acetylene
highest inlet argon mole fractions and the highest temperatures. path, remains in a pathway, where methane is converted to eth-
As the formation of ethylene is the first step in the reaction paths ane and back to methane via CH3 and is therefore not available
of both, methane and ethane, ethylene is predominantly formed for other reactions. These reactions are pressure-dependent
at low temperatures and low conversion rates, where not enough unimolecular reactions and reactions with H2, both favored at
H and CH3 radicals are available for a further decomposition. high pressure, which is also needed for the formation of acety-
With increasing inlet argon mole fractions, temperatures during lene. The maximum hydrogen yield is up to 70% for the maxi-
compression increase and ethylene is decomposed to acetylene, mum inlet temperature (573 K) and inlet pressure (10 bar),
which is unstable at these temperatures and further reacts to where the maximum yield for acetylene is up to 30% for the max-
benzene via C3H3 for the methane case and C4H4 for the ethane imum inlet temperature and the lowest inlet pressure (1 bar).
case. Another product of ethane pyrolysis, which has its Maximum yield for ethylene is up to 10% for the lowest inlet
maximum in the same range (0.93 < xAr < 0.95) is methane. temperatures (323 K) and inlet pressures of 4 bar < p0 < 10 bar
Methane is also produced by the decomposition of ethylene, and for benzene up to 16% for inlet temperatures of
but at medium temperatures it is not pyrolyzed itself. At high 373 K < T0 < 423 K and inlet pressures of 2 bar < p0 < 10 bar.
inlet argon mole fractions, several effects are superimposed The storage power and the efficiency as a function of inlet tem-
due to the high temperatures, in both, methane and ethane pyrol- perature and inlet pressure is shown in Figure 9d for methane
ysis. Significantly more acetylene and hydrogen are produced pyrolysis. If only the temperature is increased at a constant pres-
while the yields of ethylene and benzene continue to decrease. sure, a local maximum is observed. For an inlet pressure of 1 bar,
One reason for this is that methane as educt and as intermediate it is at an inlet temperature of 423 K. Charge and reaction extent
of ethane pyrolysis is almost completely converted and more rad- have opposite temperature dependencies. For higher inlet pres-
icals are available to form intermediates, such as CH3 and C2H3, sures, the storage power increases due to the cylinder charge and
and thus, more acetylene is formed not only via C2H3 but also by the faster reactions and is shifted to smaller inlet temperatures.
the decomposition of benzene via C6H5. At these high temper- The product gas contains mainly hydrogen and C20H16 as well as
atures in addition to PAHs and soot precursors, acetylene is methane from the back reaction and less higher hydrocarbons.
retained in larger quantities. The decomposition of benzene The maximum storage power for the engine of this study is
and the increased presence of acetylene at higher temperatures 7.5 kW at the maximum inlet temperature and the maximum
in turn, leads to the unwanted formation of light and heavy inlet pressure due to the high cylinder charge and the high
PAHs, which require acetylene, benzene, or intermediates based chemical exergy of hydrogen compared to methane. At lower
on them as reaction partners. temperatures for the maximum inlet pressure, storage power
As there is a maximum storage power for inlet argon mole is decreased again, due to the exothermal reaction back to meth-
fractions near xAR ¼ 0.93, in the following, the variation of inlet ane, even there is the highest cylinder charge.
temperature and inlet pressure are shown for both reactants at The efficiency is between 55% < η < 75%, depending on inlet
this fixed argon mole fraction. However, the optimum inlet temperature and pressure. The temperature dependence is simi-
argon mole fraction is slightly shifted toward lower values with lar to the storage power dependence; the efficiency has a local
(a) (b)
(c) (d)
Figure 9. Methane pyrolysis as a function of inlet temperature and inlet pressure for an inlet argon mole fraction of 0.93. a) H2 yield (black lines, colored
area) and C2H2 yield (white lines). b) C2H4 yield (black lines, colored area) and C6H6 yield (white lines). c) C20H10 yield. d) Storage power (black lines,
colored area) and storage efficiency (white lines).
maximum between 423 and 323 K. As the efficiency is the ratio of For ethane pyrolysis, similar effects occur overall as ethane
stored exergy to supplied work, also the supplied work has to be pyrolysis contains partly methane pyrolysis due to the decompo-
inspected. In general, the supplied work decreases with increas- sition from ethane to methane. In addition, the effect of
ing temperatures, if no chemical reactions occur. If endothermal increased reaction rates due to higher pressures also occurs.
reactions occur, the supplied work in the compression stroke Figure 10a–c show the yields of the main products hydrogen,
does not change noticeably until the reactions start. The resulting acetylene, ethylene, and benzene and C20H10, representative
reduced temperature and pressure levels (see Section 3.1) results for PAHs, as a function of inlet temperature and inlet pressure.
in a significant reduction of the work released in the expansion Due to the overall lower temperatures and pressures, which are
stroke. Thus, more work is required overall when endothermal up to 1700 K and 850 bar in the TDC, the yields in hydrogen and
reactions occur. Thus, two effects with opposite temperature acetylene as well as the mole fractions of light and heavy PAHs
dependencies influence the supplied work, and consequently, are significantly lower, whereas the yields of ethylene and ben-
the maximum efficiency is at medium inlet temperatures zene are higher. The maximum acetylene yield is only 14% and
because the increase in the exergy due to the endothermal reac- found for the highest inlet temperatures (573 K) and the lowest
tions is significantly higher than the increase in the supplied inlet pressures (1 bar), as the formation of acetylene is based on
work. At low inlet temperatures, the efficiency decreases due the previous formation of methane from ethane. For higher inlet
to low conversion and at high inlet temperatures, the efficiency pressures (6–10 bar), acetylene is mainly decomposed to C20H16,
decreases due a strong increase of supplied work. as it follows the path of methane pyrolysis and the main products
At higher inlet pressures, the efficiency is shifted toward lower are hydrogen (with a maximum yield of 35%) and C20H16 accord-
temperatures and the maximum efficiency is located in the lower ing to the equilibrium composition (hydrogen and solid carbon).
right quadrant of the diagram due to the reaction-accelerating Nevertheless, significantly less heavy PAHs are formed over the
effect of the high pressure and the small decrease in the supplied entire pressure and temperature range. The maximum yields for
work caused by the small amount of exothermal reactions back to ethylene and benzene are 55% and 18% and they are shifted
methane. toward higher temperatures, due to the overall lower temperature
(a) (b)
(c) (d)
Figure 10. Ethane pyrolysis as a function of inlet temperature and inlet pressure for an inlet argon mole fraction of 0.93. a) H2 yield (black lines, colored
area) and C2H2 yield (white lines). b) C2H4 yield (black lines, colored area) and C6H6 yield (white lines). c) C20H10 yield. d) Storage power (black lines,
colored area) and storage efficiency (white lines).
level: The maximum ethylene yields are located at inlet to methane pyrolysis, at an inlet temperature of 373 K and the
temperatures of 423 K < T0 < 523 K and inlet pressures of maximum inlet pressure (10 bar). The lower power is explained
1 bar < p0 < 4 bar and the maximum benzene yields at inlet by the product gas composition: it contains larger quantities of
temperature 500 K < T0 < 573 K and inlet pressures of ethylene together with small amounts of hydrogen and benzene
3 bar < p0 < 7 bar. At higher inlet pressures (>6 bar) and inter- and methane, which is not decomposed to form acetylene or
mediate inlet temperatures (373 K < T0 < 423 K), the product heavy PAHs, whereas in the case of methane pyrolysis at the
gas mainly consists of ethylene and methane, and especially same operation points, methane and heavy PAHs are mainly
methane is not decomposed toward acetylene or PAHs due to present. Furthermore, this operating point (low inlet tempera-
the principle of Le Chatelier. ture and high inlet pressure) provides the highest cylinder
The storage power and the efficiency as a function of inlet tem- charge. However, at inlet temperatures between 473 and
perature and inlet pressure for ethane pyrolysis are shown in 523 K and maximum inlet pressure, mainly C20H16 is produced
Figure 10d. Overall, the storage power and the efficiencies in from acetylene together with high amounts of methane
ethane pyrolysis are lower than in methane pyrolysis. Even at (YCH4 ¼ 85%) and only small amounts of hydrogen. The exother-
conditions, such as T0 ¼ 473 K and 2 < p0 < 3 bar, where meth- mal reaction to methane as well as the hydrogen which is stored
ane and ethane have similar efficiencies of 65%, the storage in methane lead to the minimum storage power at this operating
power in ethane pyrolysis is lower than in methane pyrolysis. point. At the maximum inlet temperature (for the same pres-
Again, the temperature dependence and the pressure depen- sure), methane decomposes; thus, the hydrogen yield increases
dence can be analyzed separately. The storage power as a and the product gas contains mainly C20H16 and hydrogen
function of the temperature at a constant pressure has a maxi- according to chemical equilibrium. Thus, the storage power
mum for an inlet temperature of 323 K < T0 < 423 K, and with increases again slightly. Compared to methane pyrolysis, there
increasing pressure, the maximum value rises and is shifted is considerably less hydrogen due to the lower H/C ratio; thus,
toward lower temperatures, as seen and explained earlier. the storage power at maximum inlet temperature and inlet pres-
The maximum storage power is 6 kW and is found, in contrast sure is lower.
The efficiency for ethane pyrolysis is between 10% < η < 70%
and with a similar dependency as the storage power: The maxi-
mum efficiency of 70% is close to the maximum storage power as
well as the minimum efficiency is close to the minimum storage
power. For a constant inlet pressure, the efficiency has a maxi-
mum at a certain inlet temperature, as discussed for methane
pyrolysis. With increasing pressure, the maximum is shifted
toward lower temperatures due to faster reactions and a more
favorable product gas composition.
The observed maximum yields of interesting species, maxi-
mum storage power and maximum efficiency are summarized
in Figure 11 for methane and ethane, respectively. The maxima
are calculated for an inlet argon mole fraction of 0.93, where the
maximum of the storage power is observed. In a direct compari- Figure 12. Conversion, yields of H2, CH4, C2H2, C2H4, and C6H6 (left axis),
and storage power and efficiency (right axes) at yields of C20H10 and
son of methane and ethane as educt, the difference in both,
C20H10 < 104 for methane (373 K, 10 bar, 89 mol% argon) and ethane
product gas composition on one side and in storage power and (373 K, 10 bar, 93 mol% argon), respectively. The yield of H2 is related to
efficiency on the other side becomes clear. Although methane H-atoms, whereas the yields for the hydrocarbons are related to the C-atoms.
pyrolysis results in significantly more hydrogen and acetylene,
the main products of ethane pyrolysis are methane and ethylene.
Benzene production is moderate for both reactants. Methane methane is relatively low, the storage power of ethane is at its
pyrolysis of methane not only leads to a larger storage power maximum value as already shown in Figure 11. If soot precursor
and higher efficiencies but also to larger yields of soot precursors formation shall be avoided, ethane is the preferable reactant, as
such as C20H10 and C20H16. For methane pyrolysis this is already high educt conversions and high C2H4 yields can still be
observed at relatively low temperatures and a broad pressure achieved, whereas methane has its advantages in the different
range and is accompanied with a moderate conversion, whereas product spectra at different operating conditions and often in
for ethane pyrolysis, soot precursors are mainly formed at the a higher storage power and efficiency.
highest inlet temperatures and pressures.
As the generation of soot and soot precursors is disadvanta-
geous, Figure 12 shows the achievable educt conversion, yields 4. Conclusion
of the most interesting species, as well as storage power and effi-
ciency at which almost no heavy PAHs (YC20H10, YC20H16 < 104) The pyrolysis of methane and ethane in a motored piston engine
are formed. For methane, such conditions are found at with a high but realistic compression ratio and engine speed
T0 ¼ 373 K, p0 ¼ 10 bar, and xAr ¼ 0.89, and for ethane at (22:1, 3000 RPM) at inlet temperatures between 323 and
T0 ¼ 373 K, p0 ¼ 10 bar, and xAr ¼ 0.93. Due to the low temper- 573 K, inlet pressures between 1 and 10 bar and an argon dilution
atures and high pressures, not only the PAH production is sig- between 85 and 99 mol% was investigated theoretically with
nificantly reduced but also the methane conversion. Also, the respect to chemical energy storage. Therefore, the piston engine
yields of H2, C2H2, and C6H6 from both, methane and ethane was modeled as a time-dependent reactor with a compression
pyrolysis are lower. The ethane conversion is reduced to a con- and expansion stroke but without gas-exchange strokes. The
siderably lesser extent. The formation of C2H4 is only slightly main products are hydrogen, acetylene, ethylene, benzene,
affected. Although the storage power and the efficiency of and soot precursors depending on the operating conditions
Figure 11. Maximum yields of H2, CH4, C2H2, C2H4, C6H6 and representative for heavy PAHs C20H10 and C20H16 (left axis), as well as maximum storage
power and maximum efficiency (right axes) for an inlet argon mole fraction of 0.93 and at different T0 and p0 for methane and ethane pyrolysis. The yield of
H2 is related to H-atoms, whereas the yields for the hydrocarbons are related to the C-atoms.
and used reactant. Due to the higher reactivity of ethane, lower preheater and inlet compressor, the efficiency according to
temperatures are required to achieve a conversion, which also Equation (22) is extended by the physical exergy in the denomi-
has the advantage of producing less PAHs and soot precursors. nator. If now the components to reach the inlet conditions are
The formation of methane from ethane is rather disadvantageous included, the efficiency is decreased by 1–6%-points at low inlet
for the process as it reduces the exergy of the mixture. The com- pressures (1 bar < p0 < 4 bar). At very high inlet pressures
parison with a previous equilibrium study[38] is interesting. (10 bar), the efficiency strongly decreases (by up to 20% points)
There, it was found that, if equilibrium is reached, the usage due to the increased exergy input. The product gas could be
of ethane is not useful for energy storage in piston engines. separated using a membrane, pressure swing adsorption, or
In the present work, it turns out that the kinetics of the process condensation as this is already state-of-the-art for comparable
can be stopped in piston engines before the equilibrium is processes.[35] The separation and recirculation of argon and
reached. In this case, ethane pyrolysis has several advantages unconverted methane could influence the efficiency of the pro-
compared to methane pyrolysis. For ethane–argon mixtures, sig- cess. The development of an entire process concept based on
nificantly less methane and benzene as well as significantly more these estimations and approaches is part of our future work.
ethylene is produced if equilibration is not reached, as assumed
in ref. [38]. In addition, PAH formation is only observed at very
high temperatures and pressures and can be avoided more easily Supporting Information
when ethane is used as reactant. In contrast, in methane pyroly-
Supporting Information is available from the Wiley Online Library or from
sis, PAH formation can only be avoided at low initial tempera- the author.
tures, which also leads to a lower conversion and lower storage
power. Nevertheless, both educts, can be used, because they lead
to different product spectra, including the expected formation of Acknowledgements
soot due to the presence of heavy PAHs, and the process can run
under different conditions (T0 and p0). Higher pressures lead to This work was supported by the Deutsche Forschungsgemeinschaft within
the framework of the DFG research unit FOR 1993 “Multi-functional con-
higher reactions rates, so that conversion and product formation
version of chemical species and energy”, Project-Number 229243862.
starts at lower temperatures. Open Access funding enabled and organized by Projekt DEAL.
The thermodynamic analysis of this process showed that
physical exergy losses are negligible and that very high exergetic
efficiencies can be reached by the conversion of work to chemical Conflict of Interest
exergy. The maximum storage power (increase in exergy) during
the compression stroke is 7.5 kW for methane and 6 kW for The authors declare no conflict of interest.
ethane which is more than 70% of the work to be stored.
These values for ethane pyrolysis can also be obtained without
soot-precursor formation. Keywords
Overall, this engine process seems promising with respect to
chemical energy storage or PtG conversion, although the theore- energy storage, ethane, hydrogen, methane, pyrolysis
tical investigation should be followed by an experimental one to Received: October 28, 2020
prove the present predictions. As methane and ethane are pres- Revised: December 16, 2020
ent in different natural gas mixtures and sometimes in bio-gases, Published online: January 25, 2021
it is also interesting for future investigations, to study the
influence of the mixture composition on the outcome of such
a piston engine energy storage process. [1] G. L. Kyriakopoulos, G. Arabatzis, Renew. Sustain. Energy Rev. 2016,
Further concepts to reduce the high argon content should be 56, 1044.
analyzed, because the separation from the products could reduce [2] X. Luo, J. Wang, M. Dooner, J. Clarke, Appl. Energy 2015, 137, 511.
the storage efficiency considerably. In addition, the separation of [3] B. Zakeri, S. Syri, Renew. Sustain. Energy Rev. 2015, 42, 569.
the products should be investigated, including the design of an [4] H. Chen, T. N. Cong, W. Yang, C. Tan, Y. Li, Y. Ding, Progr. Nat. Sci.
integrated process concept. 2009, 19, 291.
[5] X. Tan, Q. Li, H. Wang, Int. J. Elec. Power Energy Syst. 2013, 44, 179.
[6] M. M. Rahman, A. O. Oni, E. Gemechu, A. Kumar, Energy Convers.
Manage. 2020, 223, 113295.
5. Outlook
[7] Z. Zhou, M. Benbouzid, J. Frédéric Charpentier, F. Scuiller, T. Tang,
In addition to the main process, some auxiliaries are required. Renew. Sustain. Energy Rev. 2013, 18, 390.
[8] M. Budt, D. Wolf, R. Span, J. Yan, Appl. Energy 2016, 170, 250.
To set the inlet conditions (temperature, pressure), a preheater
[9] S. Karellas, N. Tzouganatos, Renew. Sustain. Energy Rev. 2014, 29,
and an inlet compressor are necessary. After the main process,
865.
the product gas should be separated, and the argon should be [10] J. P. Deane, B. P. Ó Gallachóir, E. J. McKeogh, Renew. Sustain. Energy
recirculated. The exergy input for the inlet conditions is repre- Rev. 2010, 14, 1293.
sented by the physical exergy, which is low compared to the [11] S. Rehman, L. M. Al-Hadhrami, M. M. Alam, Renew. Sustain. Energy
chemical exergy. For example, the physical exergy in the inlet Rev. 2015, 44, 586.
state is 0.4 kW and the chemical exergy 15 kW for 573 K, [12] W. F. Pickard, D. Abbott, Proc. IEEE 2012, 100, 317.
1 bar, and 93 mol% argon. To estimate the influence of the [13] M. Fasihi, D. Bogdanov, C. Breyer, Energy Proc. 2016, 99, 243.
[14] S. Shiva Kumar, V. Himabindu, Mater. Sci. Energy Technol. 2019, C. Rudolph, D. Kaczmarek, H. Gossler, S. Drost, V. Bykov,
2, 442. U. Maas, C. Schulz, Renew. Sustain. Energy Rev. 2020, 133, 110264.
[15] B. Parkinson, M. Tabatabaei, D. C. Upham, B. Ballinger, C. Greig, [40] D. G. Goodwin, H. K. Moffat, R. L. Speth, in Cantera: An Object-
S. Smart, E. McFarland, Int. J. Hydrogen Energy 2018, 43, 2540. Oriented Software Toolkit For Chemical Kinetics, Thermodynamics,
[16] M. Steinberg, Int. J. Hydrogen Energy 1998, 23, 419. And Transport Processes, Version 2.3.0, Zenodo, 2017.
[17] L. Tock, F. Maréchal, Int. J. Hydrogen Energy 2012, 37, 11785. [41] A. C. Hindmarsh, P. N. Brown, K. E. Grant, S. L. Lee, R. Serban,
[18] J. D. Holladay, J. Hu, D. L. King, Y. Wang, Catal. Today 2009, D. E. Shumaker, C. S. Woodward, ACM Trans. Math. Softw. 2005,
139, 244. 31, 363.
[19] M. A. Ancona, G. Antonioni, L. Branchini, A. de Pascale, F. Melino, [42] G. Woschni A Universally Applicable Equation for the Instantaneous
V. Orlandini, V. Antonucci, M. Ferraro, Energy Proc. 2016, 101, 854. Heat Transfer Coefficient in the Internal Combustion Engine, SAE
[20] M. Gruber, P. Weinbrecht, L. Biffar, S. Harth, D. Trimis, J. Brabandt, Technical Paper 670931, SAE International400 Commonwealth
O. Posdziech, R. Blumentritt, Fuel Process. Technol. 2018, 181, 61. Drive, Warrendale, PA, USA 1967.
[21] N. Z. Muradov, Energy Fuels 1998, 12, 41. [43] E. Ranzi, A. Frassoldati, R. Grana, A. Cuoci, T. Faravelli, A. P. Kelley,
[22] M. Karaismailoglu, H. E. Figen, S. Z. Baykara, Int. J. Hydrogen Energy C. K. Law, Progr. Energy Combust. Sci. 2012, 38, 468.
2019, 44, 9922. [44] A. Frassoldati, T. Faravelli, E. Ranzi, Combust. Flame 2003, 135, 97.
[23] E. K. Lee, S. Y. Lee, G. Y. Han, B. K. Lee, T.-J. Lee, J. H. Jun, K. J. Yoon, [45] T. Faravelli, Combust. Flame 2003, 132, 188.
Carbon 2004, 42, 2641. [46] A. Cuoci, A. Frassoldati, T. Faravelli, E. Ranzi, Combust. Flame 2009,
[24] E. Tezel, H. E. Figen, S. Z. Baykara, Int. J. Hydrogen Energy 2019, 44, 156, 2010.
9930. [47] C.-W. Zhou, Y. Li, U. Burke, C. Banyon, K. P. Somers, S. Ding, S. Khan,
[25] N. Muradov, Catal. Commun. 2001, 2, 89. J. W. Hargis, T. Sikes, E. L. Petersen, M. AlAbbad, A. Farooq, Y. Pan,
[26] E. Bartholomé, Chem. Eng. Sci. 1954, 3, 94. Y. Zhang, Z. Huang, J. Lopez, Z. Loparo, S. S. Vasu, H. J. Curran,
[27] G. Fau, N. Gascoin, P. Gillard, J. Steelant, J. Analyt. Appl. Pyrol. 2013, Combust. Flame 2018, 197, 423.
104, 1. [48] S. Porras, D. Kaczmarek, J. Herzler, S. Drost, M. Werler, T. Kasper,
[28] C. Guéret, M. Daroux, F. Billaud, Chem. Eng. Sci. 1997, 52, 815. M. Fikri, R. Schießl, B. Atakan, C. Schulz, U. Mass, Combust. Flame
[29] F. P. Larkins, A. Z. Khan, Aust. J. Chem. 1989, 42, 1655. 2020, 212, 107.
[30] A. Holmen, Catal. Today 2009, 142, 2. [49] M. Pochet, H. Jeanmart, F. Contino, Front. Mech. Eng. 2020, 6, 52.
[31] D. Nativel, B. Shu, J. Herzler, M. Fikri, C. Schulz, Proc. Combust. Inst. [50] Z. Wang, G. Du, Z. Li, X. Wang, D. Wang, Fuel 2019, 255, 115701.
2019, 37, 197. [51] E. G. Lim, E. E. Dames, K. D. Cedrone, A. J. Acocella, T. R. Needham,
[32] C. Keramiotis, G. Vourliotakis, G. Skevis, M. A. Founti, C. Esarte, A. Arce, D. R. Cohn, L. Bromberg, W. K. Cheng, W. H. Green, Can. J.
N. E. Sánchez, A. Millera, R. Bilbao, M. U. Alzueta, Energy 2012, Chem. Eng. 2016, 94, 623.
43, 103. [52] T. J. Kotas, The Exergy Method of Thermal Plant Analysis, Krieger
[33] G. F. Glasier, P. D. Pacey, Carbon 2001, 39, 15. Publishing Company, Malabar, FL, USA, 1995.
[34] C. V. Naik, A. M. Dean, Proc. Combust. Inst. 2009, 32, 437. [53] G. F. Hohenberg, Advanced Approaches for Heat Transfer Calculations,
[35] R. Hegner, B. Atakan, Int. J. Hydrogen Energy 2017, 42, 1287. SAE Technical Paper Series 790825, SAE International400
[36] K. Banke, R. Hegner, D. Schröder, C. Schulz, B. Atakan, S. A. Kaiser, Commonwealth Drive, Warrendale, PA, USA 1979.
Fuel 2019, 243, 97. [54] W. J. D. Annand, Proc. Inst. Mech. Eng. 1963, 177, 973.
[37] S. Wiemann, R. Hegner, B. Atakan, C. Schulz, S. A. Kaiser, Fuel 2018, [55] J. Chang, O. Güralp, Z. Filipi, D. N. Assanis, T.-W. Kuo, P. Najt,
215, 40. R. Rask, New Heat Transfer Correlation for an HCCI Engine Derived
[38] B. Atakan, Energies 2019, 12, 3332. from Measurements of Instantaneous Surface Heat Flux, SAE
[39] B. Atakan, S. A. Kaiser, J. Herzler, S. Porras, K. Banke, Technical Paper Series 2004-01-2996, SAE International400
O. Deutschmann, T. Kasper, M. Fikri, R. Schießl, D. Schröder, Commonwealth Drive, Warrendale, PA, USA 2004.

Compression

Uploaded by

Copyright:

Available Formats

Compression

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Compression

Uploaded by

Copyright:

Available Formats

FULL PAPER

Pyrolysis of Methane and Ethane in a Compression–

storage (EES) in applications such as

CH4 ⇆ 0.5C2 H2 þ 1.5H2 Δg 0 ¼ 155 103 J mol1 (3)

CH4 ⇆ 0.5 C2 H4 þ H2 Δg 0 ¼ 85 103 J mol1 (4)

CH4 ⇆ 0.167 C6 H6 þ 1.5 H2 Δg 0 ¼ 72 103 J mol1 (5)

Depending on reaction time and temperature, the main

ΔE ¼ m ðeProduct eEduct Þ (20)

The storage power for a four-stroke engine is then calculated

To describe the efﬁciency of the process, the storage efﬁciency

The kinetic analysis is done by analyzing the product gas com-

TDC TDC Heat CH4 H2 C2H2 C6H6 Storage Storage

Woschni 1850 75 0.83 85 63 29 5 0.99 58

You might also like