Applied Energy 215 (2018) 371–383

Contents lists available at ScienceDirect

Applied Energy
journal homepage: www.elsevier.com/locate/apenergy

Exergy analysis of an integrated solid oxide electrolysis cell-methanation T

reactor for renewable energy storage
⁎
Yu Luoa,b, Xiao-yu Wub, Yixiang Shia, , Ahmed F. Ghoniemb, Ningsheng Caia
a
Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, Department of Energy and Power Engineering, Tsinghua University, Beijing 100084,
China
b
Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139, USA

H I G H L I G H T S

• Compare various electrolysis cells based on the Second Law of Thermodynamics.

• Validated system model for the optimization of power-to-methane route.
• Integrating SOEC and methanation reactor into a single reactor.
• Higher pressure improves the thermal SOEC-methanation performance.

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

Keywords: Renewable power intermittency requires storage for load matching. A system combining a solid oxide electro-
Solid oxide electrolysis cell lysis cell (SOEC) and a methanation reactor (MR) could be an efficient way to convert excess electricity into
Methanation methane, which can be integrated with the existing natural-gas network. In this paper, a comprehensive exergy
Exergy analysis analysis is performed for three methane production systems: (i) water electrolysis + Sabatier reactor (SR, CO2
H2O/CO2 co-electrolysis
MR), (ii) H2O/CO2 co-electrolysis + MR, and (iii) a single SOEC-MR reactor, is performed. First, we find that in
Intermediate temperature
Pressurizing
the case of the water electrolysis + SR system, upon replacing the low-temperature electrolysis cell with SOEC,
the exergy efficiency is dramatically increased by 11% points of percentage at current densities higher that
8000 A m−2, owing to lower electricity consumption. Second, the type of SOEC, operating mode, and operating
conditions are optimized for this system. Results show that H2O/CO2 co-electrolysis + MR performs more ef-
ficiently than water electrolysis + SR at high current density, especially when using an intermediate-tempera-
ture SOEC. The optimal H/C ratio and temperature are found to be 10.54 and 650 °C, respectively. A pressurized
intermediate-temperature SOEC enables the system to achieve better thermal integration and improves the
exergy efficiency to over 77.43% at 6 bar. Finally, the single SOEC-MR reactor with a spatial temperature
gradient has the potential to improve the exergy efficiency to 81.34% while utilizing a compact system.

1. Introduction also serious in China, especially in rainy seasons. Over 25.4 TW h of

Renewable energy, especially wind power, solar power and hydro-
power, plays an increasingly significant role in the energy roadmap
[1,2]. The International Energy Agency (IEA) [3] predicts that renew-
ables will account for nearly 60% of newly installed power generation
capacity by 2040, of which wind power and photovoltaics (PV) will
make up almost half. The intermittency of renewable energy can lead to
the curtailment of large amounts of renewable energy. The curtailment
is more serious in China. According to the National Energy Adminis-
tration of China, 10% of PV power (5.0 TW h) and 15% of wind power
(33.9 TW h) were curtailed in 2015 [4,5]. Hydropower curtailment is
hydropower was curtailed in two provinces of Southwest China (i.e.,
Sichuan and Yunnan Provinces) in 2015 [6,7] due to insufficient grid
capacity and poor planning [8]. As a result, storage technologies are
required. The requirement for future storage capacity has been esti-
mated to be 15–20% of the annual load in order to meet 2–3 months of
storage needs [1,9].
Various technologies have been previously studied and developed
[1,10–15]. Some are feasible for seasonal storage, such as pumped
hydro, compressed air, or fuel storage [11]. Pumped hydro and com-
pressed air storage are both suitable at the large scale, and depend
largely on location. Power-to-gas (PtG), using high temperature solid

⁎
Corresponding author.
E-mail address: shyx@tsinghua.edu.cn (Y. Shi).

https://doi.org/10.1016/j.apenergy.2018.02.022
Received 8 November 2017; Received in revised form 24 January 2018; Accepted 5 February 2018
Available online 20 February 2018
0306-2619/ © 2018 Elsevier Ltd. All rights reserved.
Y. Luo et al. Applied Energy 215 (2018) 371–383

Nomenclature V volume of flow channel (m3) or voltage (V)

Vi diffusion volume of gas-phase species i (m3)
Abbreviation u flow velocity (m s−1)
Uf conversion ratio of the reactants
AEC alkaline electrolysis cell x molar fraction or distance variable through a battery
GT gradient temperature component (m)
IEA international Energy Agency
LSGM magnesium doped lanthanum gallate Greek letters
MR methanation reaction/reactor
NG natural gas α total heat transfer area per volume (m2 m−3) or transfer
OCV open-circuit voltage coefficient for charge transfer
PEMEC proton exchange membrane electrolysis cell δ thickness (mm)
PEN positive-electrolyte-negative δt gap distance of tubes (mm)
PtG power-to-gas ε porosity
PtM power-to-methane η exergy efficiency or overpotential (V)
SOEC solid oxide electrolysis cell ρ density (kg m−3)
SOFC solid oxide fuel cell σ electric conductivity (S m−1)
SR sabatier reaction λ heat conductivity (W m−3 K−1)
TNV thermal neutral voltage τ tortuosity
UT uniform temperature ϕ electric potential (V)
WGS water-gas shift reaction φs,i active material volume fraction
YSZ yttria stabilized Zirconia φ volumetric fraction
ηi overpotential (V)
English letter ω ratio of current related to H2-H2O electrochemical reac-
tion
A effective area of mass transfer (m2) ΔG change of gibbs free energy (J mol−1 K−1)
c concentration (mol m−3)
cp specific heat capacity (J kg−1 K−1) Subscript/ Superscripts
D diffusion coefficient (m2 s−1)
E activation energy for electrochemical reaction (J mol−1) 0 environmental state
Ex exergy (W) act activation polarization
ex exergy per unit mass (W kg−1) AIT air injector tube
F / ṁ flow rate (mol s−1 m−2) or Faraday constant an anode
(96,384 C mol−1) an_surf anode surface
G transfer function or gas flowrate (m3 s−1) ca cathode
h convective heat transfer coefficient (W m−2 K−1) ca_surf cathode surface
H enthalpy (J) cell solid oxide electrolysis cell
i current density (A m−2) ch channel
j0,i exchange current density (A m−2) conc concentration polarization
jloc,i local current density (A m−2) eff effective
J current (A) ex exergy
k pre-exponent factor related to activation resistance el/ca interface of electrolyte and cathode
kf kinetic constant el/an interface of electrolyte and anode
Kp reaction equilibrium constant ic interconnector
L length (m) or circumference (m) in inlet parameter
Li thickness (μm) inner inner side of the layer
Mi molar mass of gas-phase species i (kg mol−1) k kinetic processes of battery
ṅ gas molar flowrate (mol s−1) Kn Knudsen diffusion
n reaction order n relative to activation resistance leak leakage loss
N molar flux (mol m−2 s−1) max maximum exergy efficiency
No number MR methanation reaction
p pressure (Pa) OC open circuit
P power (kW) opt optimal
Q heat loss rate (W) ohm ohmic polarization
r radius (m) out outlet parameter
R resistance (Ω m or Ω) or universal gas constant outer outer side of the layer
(8.314 J mol−1 K−1) parallel in parallel
Ri source term of species i for mass balance equation reac reaction heat
(mol m−2 s−1) s solid
Ro O removal rate per active reaction area of electrolysis cells series in series
(mol s−1 m−2) tot total
Rt reaction rate per area (mol m−2 s−1) TPB triple-phase boundary
Sa,i specific surface area (m−1) WGS water-gas shift reaction
T temperature (K)

372
Y. Luo et al.
oxide electrolysis cell (SOEC), offers an attractive pathway for storage
by converting renewable energy into hydrogen, syngas, methane or
other hydrocarbon fuels [16]. This technology can achieve high con-
version efficiency and can operate over a range of scales with poten-
tially lower costs [16]. IEA has reported that methane-rich natural gas
fares best among the fossil fuels, and its consumption increased by 50%
in 2016 [3]. Power-to-methane (PtM) can integrate with the existing
natural gas pipeline.
Currently, there are two options available for converting electricity
into methane. The conventional PtM process couples a water electro-
lyzer and a Sabatier reactor (SR: CO2 + 4H2 = CH4 + 2H2O) to gen-
erate high-purity methane, which is shown as Route 1 in Fig. 1 [17].
Another novel process, shown as Route 2, couples H2O/CO2 co-elec-
trolysis using an SOEC with a methanation reactor (MR, including both
CO2 + 4H2 = CH4 + 2H2O and CO + 3H2 = CH4 + H2O). Stempien
et al. [18] performed a thermodynamic analysis to evaluate the effi-
ciency of the Route 2 system. They found that this system can achieve a
maximum energy-to-gas efficiency of 60.87%, maximum electricity-to-
gas efficiency of 81.08%, and maximum methane yield of around
1.52 Nm3 h−1 m−2. Nevertheless, the more efficient option remains
uncertain. Besides, it is inappropriate to treat electricity, heat at various
temperatures, and chemical energy equivalently.
Furthermore, studies have shown the potential to combine H2O/CO2
co-electrolysis and an MR into a single reactor, as Route 3 in Fig. 1
[19–24]. At atmospheric pressure, a temperature gradient can be
achieved by active control of the heat sources in the furnace, or by the
counter flow design in an SOEC reactor. This temperature gradient
enhances CH4 production in a tubular SOECs, achieving a higher CH4
yield (9.9–23.1%) [21–23]. Studies on the operating conditions also
revealed that increasing the operating voltage and inlet hydrogen
concentration as well as lowering the inlet steam content can improve
Fig. 1. Three pathways for power to methane.
373
Applied Energy 215 (2018) 371–383
the CH4 yield [1,20–25]. However, as less than 15 vol% hydrogen is
allowed in the existing natural gas pipeline network [26], atmospheric
pressure operation is insufficient for producing high purity methane
unless further purification is performed. Chen et al. [24,25] built a
detailed two-dimensional tubular SOEC model to couple the transport
and reaction phenomena in the pressurized tubular SOEC. They ana-
lyzed the effects of the operating condition on CH4 production. The
inlet gas composition was optimized to be H2O:H2:CO2 = 1.142:3.566:1
for the H2-feeding case or H2O:CO2 = 2.804:1 for the H2-lacking case,
and the operating pressure was optimized to be 3 bar for the H2-feeding
case at 1.3 V. CH4 production at the optimal pressure is 2.5 times of that
at atmospheric pressure. Jensen et al. [1] further showed that the
equilibrium CH4 yield can be enhanced to over 50% at 650 °C with an
inlet H/C atomic ratio of 5.67 and 4% inlet oxygen (O) when the
pressure is above 15 bar. They also suggested an operating pres-
sure > 15 bar and an operating temperature < 650 °C to avoid coking.
This operating condition implies that an intermediate temperature
SOEC should be more suitable for methane synthesis. Based on this
analysis, an energy storage approach integrating the reversible opera-
tion of SOEC with sub-surface storage of CO2 and CH4 has been pro-
posed. The round-trip efficiency was calculated to exceed 70%, and
would have a storage cost of approximately 3 € kW−1 h−1, which is
competitive with the efficiency and cost of the pumped hydro tech-
nology [1].
In this paper, a comprehensive exergy analysis is presented, to
evaluate the quantities and qualities of heat, electricity, and gaseous
fuels, based on a simulation platform built on PSE:gPROMS software.
Three systems, Routes 1–3, shown in Fig. 1, are analyzed. The exergy
efficiency in the SOEC + MR system (Route 2) was calculated and
compared with the Route 1 system, integrating other low-temperature
electrolyzers. The effects of different operating conditions, as well as
Y. Luo et al.
different materials suitable for high-temperature or intermediate-tem-
perature on the exergy efficiency of the SOEC-methanation system
(Routes 1 and 2), were estimated. Finally, the potential of integrating
the SOEC and methanation process into one reactor (Route 3) was
discussed and assessed.
2. System description and exergy calculation
A methane synthesis system consists of an electrolysis cell, an MR,
and auxiliary equipment, i.e., a compressor, heat exchanger, heater,
mixer, gas purifier, condenser, and so on. Information about the aux-
iliary equipment can be found in our previous papers [27–31]. In this
section, the electrolysis cells and the MR are described in detail.
2.1. Electrolysis cells
A detailed model of an SOEC was developed to predict the perfor-
mance of H2O electrolysis and H2O/CO2 co-electrolysis at different
temperatures and voltages. Moreover, simplified models of a proton
exchange membrane electrolysis cell (PEMEC) and an alkaline elec-
trolysis cell (AEC) were developed as well for comparison.
2.1.1. Solid oxide electrolysis cells
A multiscale SOEC model, including positive electrode-electrolyte-
negative electrode (PEN)-level, tubular cell-level, and system-level sub-
models has been described in our previous paper [27] for intermediate
and high temperature electrolysis. In this study, the unknown para-
meters in the multiscale SOEC model were obtained by fitting experi-
ments using either the YSZ-electrolyte [32] or the LSGM-electrolyte
SOECs [33]. The governing equations are expressed in Table 1. The
symbols in the governing equations are defined in Nomenclature.
The polarization curves from Ebbesen et al. [32] and Wendel et al.
[33] were used to validate the SOEC model and obtain several fitting
parameters. The former [32] used an YSZ-electrolyte SOEC (Ni-YS-
Z|YSZ|LSM) for high temperature operation (> 700 °C), while the latter
[33] used an LSGM-electrolyte SOEC (Ni-SLT|Ni-LSGM|LSGM|
LSCF|GDC|LSCF) for intermediate temperature operation (500–650 °C).
Most parameters were calibrated using data from existing literature,
and only the pre-exponential factors kca,H2 O/ kca,CO2/ kan,O2 in the activa-
tion resistance expressions were respectively tuned for SOECs with
different materials. The calibrating and tuning parameters are listed in
Table 2. The same kinetic parameters (i.e., pre-exponential factors and
activation energy in the activation resistance expressions, and ohmic
resistance) were used for both the H2O electrolysis and the H2O/CO2
co-electrolysis modes in the same SOEC. The difference in cell perfor-
mance between these two modes is represented by the ratio of the
current related to H2O electrolysis to the total current, ω. ω = 1 in the
H2O electrolysis mode, ω = 0 in the CO2 electrolysis mode, and
0 < ω < 1 in the H2O/CO2 co-electrolysis modes. The comparison
between experiment and simulation is shown in Fig. 2, indicating the
model is capable of predicting the SOEC performance. Nickel is used as
both the electronic conductor and the catalyst in the fuel electrodes in
both the YSZ-electrolyte and LSGM-electrolyte SOECs. Although the
support in the fuel electrodes is different, the kinetics of heterogeneous
chemical reactions on the nickel surface in the fuel electrodes of both
SOECs are assumed to be the same, as an estimation.
2.1.2. Proton exchange membrane electrolysis cell and alkaline electrolysis
cell
In order to compare the intermediate and high temperature opera-
tion of SOEC with low temperature electrolysis, a model was con-
structed for PEMEC [35] and AEC [36,37] at 80 °C. The current density
and hydrogen yield were kept the same as the corresponding case using
SOEC, so that the capacity, electrical consumption and released/ad-
sorbed heat of the PEMEC or AEC could be calculated. The operating
pressure given in the reference [35,36] was 13.6 atm for PEMEC, and
374
Applied Energy 215 (2018) 371–383
4.3 atm for AEC. The typical polarization curves of PEMEC, AEC, and
SOEC are shown in Fig. 3.
2.2. Methanation reactor
A methanation reactor generally uses a fixed bed reactor with Ni-
based catalyst operating in the range of 200–550 °C and at 20–25 bar
[38]. A series of reactions occurs including methanation reaction, Sa-
batier reaction (SR), WGS reaction, carbon deposition as well as me-
thanol and C2+ hydrocarbon formation. Here, a thermodynamic equi-
librium model coupling MR, SR and WGS is used, which has been used
in the literature to model the processes in the methanation reactor,
especially at high H/C ratios [18,38]. These three reactions are listed as
below:
MR: CO + 3H2 ⇔ CH 4 + H2 O,ΔH623K = −218.53 kJ mol−1 (1)
SR: CO2 + 4H2 ⇔ CH 4 + 2H2 O,ΔH623K = −179.90 kJ mol−1 (2)
WGS: CO + H2 O⇔ CO2 + H2,ΔH623K = −38.63 kJ mol−1 (3)
The equilibrium constant of any two reactions and element balance
are used to calculate the outlet gas composition. The reaction heat is
obtained by calculating the enthalpy change between inlet gas and
outlet gas. Here, the equilibriums of MR and WGS were coupled and the
equilibrium constants are listed as follows [27,39]:
− 0.2513Z 4 + 0.3665Z 3
K pMR = 1.0267 × 1010 × exp ⎛ 2
⎞
⎝ 0.5810Z −27.134Z + 3.277 ⎠
+ (4)
K pWGS = exp(−0.2935Z 3 + 0.6351Z 2 + 4.1788Z + 0.3169) (5)
1000
Z= −1
T (K ) (6)
2.3. Auxiliary equipment
Isentropic efficiency models were used to describe the compressor/
turbine and the ε-NTU method was used for heat exchanger model.
These two models as well as the gas purifier, mixer and the connection
pipeline are described in reference [28]. The gas properties are ob-
tained from PMLMaterial in gPROMS.
2.4. System schematic diagram
The system schematics are shown in Fig. 1. As previously men-
tioned, Route 1 integrates water electrolysis using an AEC, a PEMCE or
an SOEC with a Sabatier reactor. Route 2 integrates H2O/CO2 co-elec-
trolysis using SOEC with a methanation reactor. Route 3 is one single
reactor combining H2O/CO2 co-electrolysis and methanation, it will be
called SOEC-MR reactor. Route 1–3 were implemented in the simula-
tion platform built by commercial software gPROMS shown in Fig. 4.
The MR operated at 350 °C and 23 bar [38]. The heat exchange effec-
tiveness, denoting the ratio of the actual heat transfer rate to the
maximum possible heat transfer rate, was set to 0.9, and efficiency of
the compressors and turbines was set to 0.8. The purifier only separated
CO2 and H2O from the product gas, with 99.5% CO2 removal. The
outlet temperature of the purifier was 25 °C.
2.5. Exergy calculation
For heat flow at T, the exergy is [40]:
T
Exheat = Qheat ⎛1− 0 ⎞
⎝ T⎠ (7)
where Qheat denotes the heat transfer (Qheat > 0, into the system) and
T0 the ambient temperature that is 298.15 K. It should be noted that the
heat exergy that can’t be utilized in this system was considered as lost.
Y. Luo et al. Applied Energy 215 (2018) 371–383

Table 1
Governing equations for tubular SOEC stack[27,30]

PEN-scale sub-model
Average density of PEN ρPEN (kg m−3) ρPEN = φca ρca,s + φel ρel,s + φan ρan,s
Heat conductivity of PEN λPEN (W m−1 K−1) λPEN = φca λ ca,s + φel λ el,s + φan λ an,s
Mass specific heat of PEN cp,PEN (J kg−1 K−1) φca ρca,s cp,ca,s + φel ρel,s cp,el,s + φan ρan,s cp,an,s
cp,PEN =
ρPEN
Total mass balance of gas species in cathode related to the electrochemical flux ∇·(xNi ) = 0,i = H2,H2 O,CO,CO2,N2,CH 4
ωJ
NH2 |el / ca = −NH2 O |el / ca = ,ω = JH2 O / J
2FAel / ca
(1 − ω) J
NCO |el / ca = −NCO2 |el / ca =
2FAel / ca
NCH4 |el / ca = NN2 |el / ca = 0
Total mass balance of gas species in anode related to the electrochemical flux ∇·(xNi) = 0,i = O2,N2
ψJ
NO2 |el / an = − ,N | =0
4FAel / an N2 el / an
Diffusion of the gas species i in the porous electrodes xi Nj − xj Ni Ni
∇·(xi ) = ∑j ≠ i eff
− eff
ct Dij ct Dkn,i
2 −1
Effective binary molecular diffusion coefficient (m s ) 1/2
⎛ 1 1 ⎞
0.00143ε T1.75 ⎜ +
M Mj ⎟
ε ⎝ i
Dijeff = D = ⎠
τ ij τp [Vi1/3 + V1/3 2
j ]
2 −1
Effective Knudsen diffusion coefficient (m s ) eff = D ε εr T
D kn,i kn,i = 97.0
τ τ Mi
Open circuit voltage (V) / Nernst potential (V) VOC = EN =
1
{−(x H2 ΔG H2,ox + x CO ΔGCO,ox + x CH4 ΔGCH4,ox )p = p0
2F (x H2 + xCO + 4x CH4 )

+ RT [x H2lnp H2 + x CO lnpCO + x CH4 lnpCH4 −(x CO + x CH4 )lnpCO2

−(x H2 + x CH4 )lnp H2 O + (0.5x H2 + 0.5x CO + 2x CH4 )lnpO2,an ]}
Activation resistance (Ω) 1 2F pj 0.25 Eca,j
= k ⎛ ⎞ exp(− ),j = H2 O,CO2
Ract ,j RT ca,j ⎝ p0 ⎠ RT

pO 0.25
1 4F E
= k ⎛ 2⎞ exp(− an )
Ract ,O2 RT an ⎝ p0 ⎠ RT

Activation polarization (V) ηact ,ca =

J
(1 / Ract ,H2 + 1 / Ract ,CO) Ael / ca
JRact ,O2
ηact ,an =
Ael / an
Concentration polarization (V) ηconc,ca = |EN |ca surf −EN |el / ca |
ηconc,an = |EN |an surf −EN |el / an |
Ratio of current related to H2-H2O electrochemical reaction 1 / Ract ,H2
ω=
1 / Ract ,H2 + 1 / Ract ,CO
Ohmic polarization (V) and ohmic resistance (Ω m) JRohm R1 σca δic / σic δca
ηohm = ,Rohm = +
L R2 2tanh(Jic )
2 2
R1 = ⎡
⎢
⎣
( 1
σca δca ) +( 1
σan δan ⎥ )
⎤ cosh(Je )
⎦
+
2 + Jesinh(Je )
σca δca σan δan
1/2
σ 1 1
R2 = 2 ⎛ el ⎞ ( + )3/2sinh(Je )
⎝ δel ⎠ σca δca σan δan

Jic =
Lic
2
σic
σan δan δic
,Je =
π (rel / ca + rel / an)
2
σel
δel ( 1
σca δca
+
1
σan δan )
Operating voltage for SOEC modes (V) Vcell = VOC + ηact ,ca + ηact ,an + ηconc,ca + ηconc,an + ηohm + ηleak

Tubular cell-scale sub-model

Volumetric fraction of each layer in PEN (cathode, electrolyte and anode). (rk2,outer − rk2,inner )(1 − εk )
φk =
∑k = ca,el,an (rk2,outer − rk2,inner )(1 − εk )

Heat capacity of the solid phase cp,s in tubular SOEC (J kg−1 K−1) 2
L [(router 2
− rinner ) ρPEN cp,PEN + (r AIT2 2
,outer − r AIT ,inner ) ρAIT cp,AIT ]
cp,s =
ρs Vs
Effective area of mass transfer and the volume of the cathode and anode flow channel for a 2
Aca = 2πrouter L,Vca = [(δt + 2router )2−πrouter ]L
single tubular SOEC 2 2 2
Aan = 2πrinner L,Van = π (rinner −r AIT ,outer + r AIT ,inner ) L
Current for a single tubular SOEC (A) Icell = Ael / ca J
Mass balance equation for both cathode and anode dct ,k
Vk = nk̇ ,in−nk̇ ,out −Ak / ch ∑i (ψNi,k |k / ch −Ri )(k = ca,an)
dt
Species balance equation for both cathode and anode dxi,k
Vk ct ,k = nk̇ ,in (xi,k,in−xi,k ) + Ak / ch (Ri−xi,k ∑i Ri )−
dt
Ak / ch (Ni,k |k / ch −xi,k ∑i Ni,k |k / ch ) (k = ca,an)
Reaction rate of the reversible water-gas shift reaction (WGS) (mol m−3 s−1) p H2 pCO2 ⎞
RtWGS = kWGS ⎛p
f ⎜ H O pCO − ⎟
2 KWGS
p
⎝ ⎠
Reaction equilibrium constant of WGS KpWGS = exp(−0.2935Z3 + 0.6351Z2 + 4.1788Z + 0.3169)
1000
Z= −1
T (K )
−3 −1
Reaction rate of the reversible methanation reaction (MR) (mol m s ) pca 2 x 3 xCO
pca
82,000 ⎛ ⎞
RtMR = −4274
p0
exp (− RT
⎝
)
⎜x CH4 − 2 MR
H2
p0 Kp x H2 O
⎟
⎠
Reaction equilibrium constant of MR (Pa2) − 0.2513Z 4 + 0.3665Z3
KpMR = 1.0267 × 1010 × exp ⎛ 2
⎞ ⎜ ⎟

⎝+ 0.5810Z −27.134Z + 3.277 ⎠

Energy balance equation for both cathode and anode ρs cp,s Vs
dT
= ∑k = ca,an (nk̇ ,in Hk,in)− ∑k = ca,an (nk̇ ,out Hk,out )−JVcell−Qloss
dt
(continued on next page)

375
Y. Luo et al. Applied Energy 215 (2018) 371–383

Table 1 (continued)

Conversion ratio of the reactants Uf Uf =

J
[2Fṅca,in (x H2 O,in + x CO2,in)]

System-scale sub-model
Stack current (A), stack voltage (V) and stack power (W) Jtot = Noparallel Jcell
Vtot = Noseries Vcell
Ptot = Jtot Vtot
Stack inlet molar flow rate (mol s−1) nitot
,in = Noseries Noparellel ni,in

Therefore, the exergy of heat was expressed as:

heater SOEC MR
Exheat = MAX (0,Exheat + Exheat + Exheat ) (8)
For the electricity or work P, the exergy input is equivalent to −P
(P < 0, power into the system) [40–42]. This part of exergy includes
the electricity input of electrolysis, work input of compressors and work
output of turbines. For a specific gas flow at a temperature of T and a
pressure of p, its exergy contains two parts (neglecting the kinetic and
potential exergies): the physical exergy, exphys, and chemical exergy,
exchem, [40]:
Ex flow = ṁ ·ex flow = ṁ (ex phys + ex chem) (9)

ex phys = (h−h 0)−T0 (s−s0) (10)

ex chem = ∑ x i ex i,chem
i = H2/ H2 O/ CO
/ CO2/ CH4/ O2/ N2

0
⎧− RT lnx i |i = H2 O/CO2 /O2 /N2
⎪ 0 0 b 0
⎪⎡gf ,i−agf ,CO2 − 2 gf ,H2 O (g ) ⎤+
⎣ ⎦
ex i,chem =
⎨ ⎡
b
( x O0 2 )a + 4 ⎤ i = H2 /CO/CH 4
⎪ RT0ln ⎢
b⎥
⎪ ⎢ (x 0 a 0
) ( x H2 O (g ) ) 2 ⎥
⎩ ⎣ CO2 ⎦ (11)
where ṁ denotes the mass of gas flow and ex the specific exergy, h,s the
enthalpy at T and p, h0, s0 the enthalpy at T0 and p0 (101,325 Pa). μi,0 is
the chemical potential of species i at T0 and p0, and μi,00 is the chemical
potential of species i in the air. a,b denote the number of carbon and
hydrogen atoms in the molecule i (H2, CO or CH4), respectively.Ther-
efore, the exergy efficiency is calculated as:
CH4
Ex flow
ηex = in out
Ex elec + Exheat + ∑ Ex flow − ∑ Ex flow
excludeCH4 (12)

Fig. 2. Comparison between experimental and simulated polarization curves for (a) YSZ-
3. Results and discussion
electrolyte [32] and (b) LSGM-electrolyte [33] SOECs in both H2O electrolysis and H2O/
CO2 co-electrolysis.
3.1. Route 1: Comparison among different electrolysis cells

and YSZ-SOEC at 800 °C. The base case has a pure H2O inlet flow rate of
For an SOEC, high operating temperatures result in lower electricity
0.08 mol s−1 m−2 for water electrolysis cells (5.18 kg h−1 m−2), and a
consumption but higher thermal energy input. Here, we present a
corresponding CO2 inlet flow rate of 0.029 mol s−1 m−2 for SR, to de-
comparison among Route 1 systems with different water electrolysis
liver a H/C atomic ratio of 5.5 (a H2O/CO2 molar ratio of 2.75) [1]. The
cells including AEC at 80 °C, PEMEC at 80 °C, LSGM-SOEC at 600 °C,

Table 2
Calibrated parameters and tuning parameters[27,33,34]

Parameters YSZ-electrolyte LSGM-electrolyte

Cathode porosity εca /tortuosityτca 0.5/5 [27] 0.26/3 [33]

Anode porosity εan /tortuosityτan 0.5/5 [27] 0.3/3 [33]
Ionic conductivity of electrolyte σel ion (S m−1) 5.17 × 108
3.34 × 10 4exp − ( 85634
RT ) [34] T
exp (− 93,800
RT ) [33]
−1
Electronic conductivity of cathode and anode σca elec / σan elec (S m ) 4.2 × 107 1 × 10 /3 × 104 [33]
3
3.27 × 106−1065.3T /
T
exp ( −
1150
T ) [34]
Activation Energy Eca/Ean (kJ mol−1) 60/162 [33]
Pre-exponent factors kk of H2/CO/O2 in the equations of activation resistance (tuned) 6.0 × 1010/1.2 × 1010 /4.0 × 1012 6.0 × 106/1.2 × 106 /5.0 × 1014

376
Y. Luo et al. Applied Energy 215 (2018) 371–383

higher heat and lower electrical inputs. Yet the ηex of SOEC exceeds that
of PEMEC at higher current densities because of the lower electrical
input. AEC has the lowest ηex. The exergy efficiency of the LSGM-SOEC
is slightly lower than that of the YSZ-SOEC at i < 8000 A m−2, but
surpasses it by about 1% at i < 8500 A m−2, due to the better heat
integration of the LSGM-SOEC and MR at high current density. The
peak exergy efficiency of the SOEC is at i = 8000–8500 A m−2, where
the overall heat exergy in the system turns from positive to negative. To
evaluate the main energy consumption, the detailed exergy flow at
i = 8500 A m−2 (corresponding to O removal rate
RO = 0.044 mol s−1 m−2) is shown in Fig. 5b. At H/C = 5.5 and
RO = 0.044 mol s−1 m−2, the CH4 production rate in the four cases is
almost the same. The electricity consumption for water electrolysis
accounts for at least 86% of the total energy consumption. The value
exceeds 93% in the cases of the two low temperature electrolysis cells.
Therefore, the use of high temperature electrolysis cells is crucial for
lowering the energy consumption of the methanation system. Next, we
focus on optimizing the operation of the SOEC based system.
Fig. 3. Typical polarization curves of PEMEC, AEC, LSGM-electrolyte SOEC, and YSZ-
electrolyte SOEC [32,33,35,36] for H2O electrolysis.
3.2. Route 1: Effect of H/C ratio
flow rate is evaluated per unit of active reaction area. The exergy ef-
In the previous case, Route 1 at a H/C ratio of 5.5 was analyzed, and
ficiency, ηex, and the methane production rate, RCH4, obtained at the
excess carbon dioxide was found. Here, we examine the effect of the H/
outlet of the CO2 splitter at varying current densities (proportional to
C ratio on ηex in order to optimize the system operation. An optimal
the hydrogen spillover rate/oxygen removal rate ratio) are shown in
current density of 8500 A m−2 (RO = 0.044 mol s−1 m−2) is chosen
Fig. 5a. Results indicate that the PEMEC has the highest exergy effi-
based on the LSGM-electrolyte SOEC at 600 °C, and the H/C ratio is
ciency at a low current density, i (< 3500 A m−2), resulting from
varied between 5.5 and 14.6. The H2O inlet flow rate is fixed at

Fig. 4. System schematic diagram in gPROMS.

377
Y. Luo et al.
Fig. 5. (a) Exergy efficiency and CH4 production rate in the Route 1 system with various
water electrolysis cells at varying current densities; (b) exergy input/output rate for
different water electrolysis cells at 8500 A m−2.
0.055 mol s−1 m−2 to ensure 80% H2O conversion in the electrolysis
cells. The effects of the H/C ratio on the ηex, production rate, and vo-
lume fraction of CH4 and H2 are shown in Fig. 6. The exergy efficiency
improves slightly at H/C ratios below 10 and rises once the H/C ratio
exceeds 10. The change in the CH4 production rate shows that the RCH4
stays almost constant at H/C ratios below 10 but drops when the H/C
ratio exceeds 10. The opposite trend is observed for the H2 production
rate at the system outlet. A H/C ratio of 10 is the critical point when
CO2 and H2 are at a stoichiometric ratio. As the H/C ratio exceeds 10,
H2 becomes excessive, and its volume fraction rises in the system outlet.
In this case, ηex is higher because the chemical exergy of hydrogen is
higher than that of CH4. Nevertheless, the gas in the pipeline network
can only contain up to 15 vol% hydrogen [26]. Therefore, in the fol-
lowing analysis, an optimal H/C ratio of 10.54 (H2O/CO2 = 5.27) is
used for further analysis to limit the outlet to 15 vol% of H2. In this
case, the system reaches an exergy efficiency of 71.00% and produces
0.81 Nm3 h−1 m−2 of CH4.
3.3. Comparison between Route 1 and Route 2
H2O/CO2 co-electrolysis has been considered as an alternative
technology for CO2 conversion, one-step syngas production, and higher
hydrocarbon synthesis [43,44]. A tradeoff has to be considered for
optimizing the efficiency of H2O/CO2 co-electrolysis. On the one hand,
it is difficult to break C]O bonds, hence H2O/CO2 co-electrolysis re-
quires more electricity than H2O electrolysis, which lowers the effi-
ciency of the SOEC reactor. On the other hand, methanation from CO
(Eq. (1)) releases more heat than the Sabatier reaction (methanation
starting with CO2, Eq. (2)). Thus, a greater amount of heat could be
378
Applied Energy 215 (2018) 371–383
recovered and used to support the endothermic H2O/CO2 co-electro-
lysis when the voltage is below the thermal-neutral voltage (TNV). The
value of TNV can be calculated by the following equation [45]:
Hout −Hin
VTNV =
2F (13)
Generally, the TNV is between 1.28 (for H2O electrolysis) and 1.47
(TNV for CO2 electrolysis) [46,47]. In this section, we compare Route 1
(H2O electrolysis + SR) and Route 2 (H2O/CO2 co-electrolysis + MR)
in detail. In Route 1, CH4 can only be produced in the SR. In Route 2,
CH4 can be produced in both the H2O/CO2 co-electrolysis in the SOEC
and the MR. Fig. 7a shows the exergy efficiency of a Sabatier reactor
integrated with H2O electrolysis (Route 1) and an MR integrated with
H2O/CO2 co-electrolysis (Route 2), using LSGM-SOEC at 600 °C and
YSZ-SOEC at 800 °C, respectively, with an optimized H/C ratio of 10.54.
Assuming no current efficiency loss, the outlet gas composition and
flow rate are constant at the specified inlet gas and current density in
both Route 1 and Route 2, because of the equilibrium gas obtained from
the MR/SR outlet. A peak efficiency is achieved when the current
density is within the range from −5500 A m−2 to −6000 A m−2, as the
heat demand of the total system turns from positive to negative. The
highest exergy efficiency of Route 1 is 70.08% for the YSZ-SOEC at
800 °C, and 70.78% for the LSGM-SOEC at 600 °C at −5500 A m−2, and
that of Route 2 is 69.46% for YSZ-SOEC at 800 °C, and 70.61% for
LSGM-SOEC at 600 °C at −6000 A m−2. High current density means
that the heat requirement can be supplied by the MR and self-heating
from the polarization losses, which results in a drop in the exergy ef-
ficiency. Generally, a heat pump is required to transfer the low-tem-
perature heat from the MR to the SOEC when operating at a higher
temperature. In this analysis, we use the values of the heat exergy
among the SOEC, the MR, and the heaters to describe the heat exchange
among them instead of performing a detailed analysis. When the cur-
rent density is higher than 8500 A m−2, the lack of CO2 limits CH4
production but enhances the H2 content in the product gases, so that the
ηex drops more slowly and recovers at a current density over 8500 A
m−2. A comparison between LSGM-SOEC at 600 °C and YSZ-SOEC at
800 °C reveals that an LSGM-SOEC operating at an intermediate tem-
perature has a higher efficiency than a YSZ-SOEC operating at a high
temperature. Despite the lower electrical consumption of the high
temperature SOEC, higher quality of heat input and compressor work
demand outweighs the saved electricity.
Route 2 has lower exergy efficiency at low current densities due to
the higher overpotential of H2O/CO2 co-electrolysis, but its efficiency
rises with current density owing to the availability of more reactants. In
addition, the current density of H2O/CO2 co-electrolysis for the
Fig. 6. Effects of the H/C ratio on the exergy efficiency, production rate, and volume
fraction of CH4 and H2 at the Route 1 system outlet, with a constant O removal rate
(current density).
Y. Luo et al. Applied Energy 215 (2018) 371–383

met with decreasing inlet reactant flow rate.

3.4. Effect of temperature on H2O/CO2 co-electrolysis

A higher operating temperature for H2O/CO2 co-electrolysis re-

quires higher heat demand to preheat the inlet gas, but lowers the
electrical consumption for accelerating the reaction and reducing the
reversible potential. The effects of the SOEC operating temperature on
the system exergy efficiency are evaluated at a H/C ratio of 10.54, and a
current density of −6000 A m−2, within the validated temperature
range, with the results depicted in Fig. 8. In these cases, H2O/CO2 co-
electrolysis in the LSGM-SOEC performs more efficiently than H2O
electrolysis in the LSGM-YSZ, while H2O electrolysis in the YSZ-SOEC
has a higher efficiency than H2O/CO2 co-electrolysis in the YSZ-SOEC,
corresponding to the exergy efficiency at the same current density in
Fig. 7a. Moreover, for all four cases considered here, the exergy effi-
ciency first rises, then slows down, and stabilizes at approximately 75%
with increasing operating temperature. The temperature enhancement
leads to a significant drop in electricity consumption, but higher heat
demand owing to the drop in the operating voltage. The initial rise of
ηex is the result of the heat supply from the MR. As the heat supply from
the MR is insufficient, additional heat is needed, hence the exergy ef-
ficiency improves more slowly and finally stabilizes. When the oper-
ating temperature of the LSGM-SOEC rises from 600 °C to 650 °C, the
exergy efficiency rises by almost 5%. At 650 °C, the LSGM-SOEC has an
exergy efficiency of 75%, which is the same as the efficiency of the YSZ-
SOEC operating at 850 °C, thanks to its better electrochemical perfor-
mance at intermediate temperatures (See Fig. 3). The ηex of the LSGM-
SOEC at 700 °C is approximately 75%, almost 18% higher than that of
the YSZ-SOEC at the same temperature. Consequently, the optimal
operating temperature is found to be 650 °C (923.15 K) when an LSGM-
electrolyte SOEC was used.

Fig. 7. (a) Exergy efficiency and corresponding CH4 production rate of a Sabatier reactor 3.5. Effect of pressure on H2O/CO2 co-electrolysis
integrated with H2O electrolysis (Route 1) and a methanation reactor integrated with
H2O/CO2 co-electrolysis (Route 2), driven by an LSGM-SOEC at 600 °C and a YSZ-SOEC at
800 °C, respectively, with an H/C Ratio of 10.54. (b) The maximum exergy efficiency and
The base cases presented the coupling between the pressurized MR
the corresponding operating current density variation with the inlet steam flow rate ratio at 23 bar and the SOEC reactor operating at 2 bar. Operating at higher
at H/C = 10.54 for Route 1 and Route 2. pressure lowers the internal resistance of the SOEC [48,49]. Moreover,
the direct methane synthesis in the SOEC stacks is favored at higher
pressure, which offers the potential for a more compact reactor and
maximum exergy efficiency in Route 2 is higher than that of H2O
system design. If the SOEC reactor is also pressurized, the additional
electrolysis in Route 1 because of the extra heat demand for heating
energy consumption and economic cost for this system is lower, in
CO2 to the operating temperature of the SOEC. Therefore, Route 2 has
addition to being beneficial for the methane synthesis system. Thus, the
high efficiency at the high current density required to achieve high-
effect of the higher pressurize operation on the exergy efficiency is
yield methane. In the following sections, the optimal current density is
analyzed.
set at −6000 A m−2 for Route 2 and Route 3. Further, the maximum ηex
Fig. 8b shows the effect of the operating pressure on the exergy
and its corresponding current density of both H2O electrolysis and H2O/
efficiency of the methane synthesis system using H2O electrolysis at
CO2 co-electrolysis of the LSGM-SOEC at 600 °C are investigated.
600 °C, and H2O/CO2 co-electrolysis at 600 °C and 650 °C, operating at
Fig. 7b shows that the maximum efficiency and corresponding current
−6000 A m−2 and at a H/C ratio = 10.54. Pressure increase sig-
density are sensitive to the inlet H2O flow rate at a constant H/C ratio.
nificantly improves the exergy efficiency at operating pressures below
As the inlet reactant rate increases, the thermal demand for gas heating
6 bar, beyond which the efficiency rises more slowly. The improvement
increases, but more steam remains unreacted in the SOEC, causing a
is attributed to the decrease of the internal resistance as well as the rise
lower ηex, and a higher current density. To further examine the impact
of the operating conditions, variation of the maximum ηex, and corre-
Table 3
sponding current densities are shown in order to evaluate the optimal
Fitting equations for the maximum exergy efficiency and optimal current density of H2O
operating current density for a certain inlet flow rate at the optimal H/C electrolysis and H2O/CO2 co-electrolysis.
ratio of 10.54. The fitting equations of H2O electrolysis and H2O/CO2
co-electrolysis are displayed in Table 3. The theoretical limiting current H2O electrolysis*
2
density for complete conversion of reactant is drawn as a short-dashed max
ηex in
= 9.7167FH in
2 O−2.7589FH2 O + 0.84747
line in Fig. 7b. This indicates that the smallest applicable inlet steam in
Iopt (A m−2) = −634290FH
2 in
2 O + 26474FH2 O−4840.2
flow rate should be over 0.237 mol s−1 m−2 for H2O electrolysis, and
0.187 mol s−1 m−2 for H2O/CO2 co-electrolysis, in order to achieve the H2O/CO2 co-electrolysis*
2
maximum ηex at a certain inlet flow rate. The fitting curves in Fig. 7b max
ηex in
= 22.466FH in
2 O−4.7716FH2 O + 0.9014
2
(short dotted lines) indicate that the highest exergy efficiency ap- in
Iopt (A m−2) = −74313FH in
2 O−37972FH2 O−3561.4
proaches but cannot reach 77.43% for H2O electrolysis, and 82.00% for
−1
H2O/CO2 co-electrolysis, due to the increasing concentration resistance in
* Unit of FH 2 O : mol s m−2.

379
Y. Luo et al.
Fig. 8. (a) Effects of the SOEC temperature on the exergy efficiency of the methanation process, coupled with H2O electrolysis or H2O/CO2 co-electrolysis, respectively driven by an
LSGM-SOEC and a YSZ-SOEC at a H/C ratio of 10.54 and a current density of −6000 A m−2; (b) effect of the operating pressure on the exergy efficiency of the methane synthesis system
using H2O electrolysis at 600 °C, and H2O/CO2 co-electrolysis at 600 °C and 650 °C, operated at −6000 A m−2 and a H/C ratio of 10.54.
Fig. 9. Effects of pressure in the Route 3 system using a uniform-temperature SOEC-MR
reactor at 550 °C and 600 °C on (a) product gas composition; (b) calculated exergy effi-
ciency, ratio of energy stored in CH4, and actual exergy efficiency.
of the exothermic methanation rate within the SOEC stacks. The effi-
ciency improvement of the H2O electrolysis case at higher pressures
(about 4.6% from 2 bar to 6 bar) results from the drop in the
380
Applied Energy 215 (2018) 371–383
electrochemical resistance. For the H2O/CO2 co-electrolysis case, the
additional exergy efficiency improvement (about 1.2%) is due to the
heat supply from the methanation within the SOEC. Calculations also
reveal that the operating voltage is always lower than 1.26 V in all
cases, and hence the SOEC is operating in the endothermic mode, and
extra heat is needed for the system. As the pressure improves the me-
thanation, heat generated by methanation in the SOEC stacks supports
the endothermic electrolysis, and lowers the overall heat input from the
heater. At 600 °C, the pressurized H2O/CO2 co-electrolysis process
(over 6 bar), coupled with the MR, can reach an exergy efficiency of
over 76.20%. A temperature rise of 50 °C to 650 °C further improves the
exergy efficiency to over 77.43% at an SOEC operating pressure of over
6 bar.
3.6. The potential to couple SOEC and methanation in one reactor
The previous section has demonstrated the enhanced exergy effi-
ciency in the system using a pressurized SOEC + MR (Route 1 and 2)
because of the more effective heat integration. However, a considerable
fraction of the heat generated in the MR is utilized by the SOEC in an
inefficient way. By integrating the SOEC and MR into a single reactor,
the local thermal coupling further enhances the exergy efficiency. In the
Introduction (Section 1) we reviewed work on the SOEC-MR at atmo-
spheric pressure, which led to at most a 23.1% methane yield [19–23].
In this section, the potential of an SOEC-MR will be evaluated with
respect to the exergy efficiency and CH4 production ratio.
3.6.1. Route 3: Uniform-temperature SOEC-MR reactor
An SOEC-MR reactor operating at a uniform temperature could be
easier to realize than one operating with a spatial temperature gradient.
The porous cathode is a mixed conductor sintered from Ni and LSGM
particles for the LSGM-SOEC. Ni particles have been shown to be an
excellent catalyst for the methanation [38]. By adding a Ni-based cat-
alyst packed bed in the cathode channel or coating Ni particles on the
wall [50], H2O/CO2 co-electrolysis and methanation processes could be
coupled in a tubular SOEC. In the simulation, a uniform-temperature
SOEC-MR (UT-SOEC-MR) can be modeled by considering the outlet gas
composition as being at equilibrium. In the system, the SOEC stack and
MR are integrated into a uniform-temperature SOEC-MR. The outlet gas
from the UT-SOEC-MR is fed into the CO2 purifier after heat exchange
with the inlet gas. The operating pressure and temperature of the UT-
SOEC-MR varies from 2 bar to 25 bar, and between 550 °C and 600 °C,
Y. Luo et al.
respectively, considering the tradeoff between electrochemical perfor-
mance and methanation potential. Fig. 9 shows the corresponding ex-
ergy efficiency, product gas composition, and ratio of energy stored in
CH4. An exergy efficiency as high as over 85% is obtained at 600 °C and
2 bar, yet such a high efficiency corresponds to a hydrogen content of
almost 90 vol% and methane content less than 5% in the product gas.
Pressure rise and temperature drop both promote CH4 production ratio
while decreases the exergy efficiency due to the decrease in the hy-
drogen content. However, even if the operating pressure rises to 25 bar,
and the temperature drops to 550 °C, the CH4 volume fraction only just
reaches 43.30 vol%, the CO volume fraction is below 10%, and H2
volume fraction still stays at over 54 vol%. The volume fraction of
hydrogen is much higher than the maximum limit of H2 for natural gas
network pipeline (15 vol%). Thus, an additional purifier for hydrogen
removal is needed in the case of the system using the UT-SOEC-MR. The
actual exergy efficiency that considers H2 removal should be equal to
the product of the calculated exergy efficiency and the ratio of energy
stored in CH4 to that in the outlet gas of the system shown in Fig. 9b.
The actual exergy efficiency at 550 °C is higher, but it approaches that
at 600 °C with reactor pressurization. The actual exergy efficiency is
about 50% when the UT-SOEC-MR operates at 25 bar, which shows that
the UT-SOEC-MR is not efficient when compared with the separate
SOEC + MR system.
3.6.2. Route 3: An SOEC-MR reactor with a temperature gradient
The optimal temperature mismatch between the SOEC and the MR
is a key issue in the UT-SOEC-MR reactor. To examine this issue, an
SOEC-MR reactor with a spatial temperature gradient (GT-SOEC-MR) as
Fig. 10. (a) Schematic of a GT-SOEC-MR reactor; (b) effect of the operating pressure on the exergy efficiency and product gas composition in the Route 3 system using a GT-SOEC-MR
reactor.
381
Applied Energy 215 (2018) 371–383
shown in Fig. 10a, that enables H2O/CO2 co-electrolysis and metha-
nation processes to operate at their respective favored temperature
[19,21–23] should be analyzed. The gradient-temperature reactor has
two main zones: the SOEC zone and the MR zone. The SOEC zone lo-
cated at the upstream position operates uniformly at the SOEC-favored
temperature TSOEC to promote the reaction of H2O/CO2 co-electrolysis.
The MR zone is located at the downstream position to further process
the produced syngas. The MR zone has a temperature gradient from
TSOEC to MR-favored temperature TMR. As the equilibrium of the me-
thanation shifts forward, heat is gradually generated along with the
cathode gas flow. Compared with the separate SOEC-MR system
(Routes 1/2), the generated heat from the MR has a much higher
average temperature (between TMR and TSOEC), hence a higher exergy.
Our previous research [22], combining experiments and numerical si-
mulation, found that the counter flow mode is able to form a natural
temperature gradient to promote the H2O/CO2 co-electrolysis at the
upstream and methanation at the downstream. Meanwhile, the counter
flow can also lower the temperature difference in the heat transfer
processes to reduce the exergy losses. Chen and Lei et al. [21,23] re-
spectively used the tubular SOEC to achieve a CH4 yield of 11.84% and
23.1% by locating the high temperature zone around the effective SOEC
region and the low temperature zone at the downstream position. Chen
et al. [24] further estimated the effect of pressurization on the reactor.
Based on numerical modeling, it was found that 3 bar was the optimal
operating pressure and the methanation temperature was suggested to
be lower than 488 °C for ensuring at least 95% of maximum CH4 pro-
duction.
Such a GT-SOEC-MR reactor has been demonstrated and proved to
Y. Luo et al.
have the potential to produce high-yield methane. Given efficient heat
transfer between the MR and the SOEC, heat recovery from the outlet
gas and adequate methanation can be realized to mitigate the exergy
losses caused by temperature mismatch, and an ideal system based on
the GT-SOEC-MR reactor will be able to operate more efficiently. Such
an ideal GT-SOEC-MR can be simulated by linking the SOEC operating
at TSOEC directly to the MR at TMR. Results of the system calculation,
based on the ideal GT-SOEC-MR, are shown in Fig. 10b, where
TMR = 350 °C, and the SOEC is operating at −6000 A m−2, the H/C
ratio = 10.54, with TSOEC = 600 °C and 650 °C for comparison with the
previous cases in Fig. 8b. Different from the separate SOEC + MR
system, the GT-SOEC-MR system has a negative correlation with the
operating pressure due to the different pressure dependency of the
SOEC and the MR. The separate SOEC + MR system in Fig. 8b operates
the MR at a constant pressure of 23 bar, yet the GT-SOEC-MR system
operates the SOEC and the MR at a similar pressure. The exergy effi-
ciency curves of these two systems intersect at 23 bar at 600 °C and
27 bar at 650 °C, showing that the GT-SOEC-MR system is superior
when operated at < 23 bar for 600 °C, and < 27 bar for 650 °C. At low
pressures, the GT-SOEC-MR system performs at a higher exergy effi-
ciency than the separate SOEC + MR system because of the higher
hydrogen content in the product gas. To satisfy the hydrogen tolerance
(15 vol%) of natural gas network pipeline, the operating pressure of
over 8.15 bar is required. Nevertheless, the pressurized operation low-
ered the exergy efficiency. As a consequence, the optimal operating
pressure is 8.15 bar based on the exergy efficiency and a tolerable hy-
drogen volume fraction of natural gas network. At the optimal pressure,
the GT-SOEC-MR system can achieve an exergy efficiency of 79.30% at
600 °C, and 81.34% at 650 °C, which is about 3% higher than the se-
parate SOEC + MR system.
4. Conclusion
In this paper, a comprehensive exergy analysis of three different
methane synthesis systems, designated as Routes 1–3, was performed.
Route 1 integrates water electrolysis with a Sabatier reactor, in which
different water electrolyzers were compared. An SOEC was compared
with low temperature electrolysis, i.e., an alkaline electrolysis cell and a
proton exchange membrane electrolysis cell. The Route 1 system, using
an SOEC, becomes more efficient at a current density below −3400 A
m−2 and has over 11 points of percentage higher exergy efficiency at or
below -8000 A m−2 due to the lower electrical consumption.
In Route 2, H2O/CO2 co-electrolysis by an SOEC is used to replace
the H2O electrolysis. H2O electrolysis and H2O/CO2 co-electrolysis,
based on an intermediate temperature LSGM-electrolyte SOEC and a
high temperature YSZ-electrolyte SOEC, respectively, were considered
in Route 2 systems. Results revealed that H2O/CO2 co-electrolysis has a
higher current density at the maximum efficiency, hence is more sui-
table for higher-yield methane production. Furthermore, the inter-
mediate-temperature SOEC enabled the methane synthesis system to
run more efficiently and achieve a more compact design. The effects of
current density, inlet H/C ratio, flow rate, and temperature on the ex-
ergy efficiency were estimated. The optimal system operation was
found at a current density of −6000 A m−2, a H/C ratio of 10.54, and
an SOEC operating temperature of 650 °C. Based on the optimal oper-
ating conditions and the intermediate temperature LSGM-electrolyte
SOEC, the operating pressure of the SOEC was varied while the tem-
perature and pressure of the MR was kept at 350 °C and 23 bar. Results
indicated that the intermediate-temperature SOEC enables the Route 2
system to realize better thermal integration and improves the exergy
efficiency to over 77.43% at over 6 bar.
Finally, the potential of a one-reactor operation to integrate the
SOEC with the MR (Route 3) was discussed. The temperature mismatch
of the SOEC and the MR makes it hard to use a uniform-temperature
single SOEC-MR reactor to produce CH4-rich product gas (at most
43.40% at 550 °C and 25 bar), as it requires further hydrogen removal.
382
Applied Energy 215 (2018) 371–383
When only considering methane production, the energy-to-methane
efficiency using a uniform-temperature SOEC-MR reactor only reaches
50.30%. Therefore, a single SOEC-MR reactor with an internal tem-
perature gradient was studied. Calculations revealed that an ideal
temperature-gradient SOEC-MR reactor had significant potential to
enhancing efficiency. When the operating pressure was lower than
19 bar, the temperature-gradient supported single reactor performed
more efficiently. Limited to the 15 vol% hydrogen tolerance of the
natural gas network, the optimal operating pressure of 8.15 bar can
realize an exergy efficiency as high as 81.34%, at a H/C ratio of 10.54,
SOEC temperature of 650 °C, MR temperature of 350 °C, and operating
current density of −6000 A m−2. For the realization of such an effec-
tive single SOEC-MR reactor, effective heat transfer between the MR
and SOEC zones, heat recovery from the outlet gas and adequate me-
thanation should be possible to lower the exergy losses caused by
temperature mismatch.
Acknowledgement
The study support from Project 2014CB249201 supported by the
National Basic Research Program of China (973 Program), Projects
51476092 (National Natural Science Foundation of China, NSFC) and
Youth Foundation Program for Fundamental Scientific Research in
Tsinghua University (221 Program) are appreciated.
References
[1] Jensen SH, Graves C, Mogensen M, Wendel C, Braun R, Hughes G, et al. Large-scale
electricity storage utilizing reversible solid oxide cells combined with underground
storage of CO2 and CH4. Energy Environ Sci 2015;8:2471–9.
[2] Connolly D, Lund H, Mathiesen BV, Leahy M. A review of computer tools for ana-
lysing the integration of renewable energy into various energy systems. Appl Energy
2010;87:1059–82.
[3] IEA. World energy outlook 2016. IEA Publications. Paris; 2016. < http://paper.
people.com.cn/zgnyb/html/2017-01/09/content_1742525.htm > .
[4] National Energy Administration of China. Related data statistics of the photovoltaic
power in 2015; 2016. < http://www.nea.gov.cn/2016-02/05/c_135076636.
htm > .
[5] National Energy Administration of China. Development status of the wind power
industry in 2015; 2016. < http://www.nea.gov.cn/2016-02/02/c_135066586.
htm > .
[6] Li Y. The hydropower curtailment reaches a new peak in 2016. China Energy News;
2017. < http://paper.people.com.cn/zgnyb/html/2017-01/09/content_1742525.
htm > .
[7] Yang D, Hu Z. Seriou hydropower curtailment in Sichuan requires some measures.
Xinhua News; 2016. < http://news.xinhuanet.com/fortune/2016-02/26/c_
1118167572.htm > .
[8] Stanway D. China’s Sichuan to restrict new hydro projects over 2016-2020. Reuters;
2016. < < http://sustainability.thomsonreuters.com/2016/10/19/chinas-
sichuan-to-restrict-new-hydro-projects-over-2016-2020/ > .
[9] Becker S, Frew BA, Andresen GB, Zeyer T, Schramm S, Greiner M, et al. Features of a
fully renewable US electricity system: Optimized mixes of wind and solar PV and
transmission grid extensions. Energy 2014;72:443–58.
[10] Wang H, Yin W, Abdollahi E, Lahdelma R, Jiao W. Modelling and optimization of
CHP based district heating system with renewable energy production and energy
storage. Appl Energy 2015;159:401–21.
[11] Beaudin M, Zareipour H, Schellenberglabe A, Rosehart W. Energy storage for mi-
tigating the variability of renewable electricity sources: an updated review. Energy
Sustain Dev 2010;14:302–14.
[12] Zhao H, Wu Q, Hu S, Xu H, Rasmussen CN. Review of energy storage system for
wind power integration support. Appl Energy 2015;137:545–53.
[13] Ma T, Yang H, Lu L. Development of hybrid battery-supercapacitor energy storage
for remote area renewable energy systems. Appl Energy 2015;153:56–62.
[14] Zhang Z, Zhang X, Chen W, Rasim Y, Salman W, Pan H, et al. A high-efficiency
energy regenerative shock absorber using supercapacitors for renewable energy
applications in range extended electric vehicle. Appl Energy 2016;178:177–88.
[15] Zhu J, Yuan W, Qiu M, Wei B, Zhang H, Chen P, et al. Experimental demonstration
and application planning of high temperature superconducting energy storage
system for renewable power grids. Appl Energy 2015;137:692–8.
[16] Foit SR, Vinke IC, de Haart LG, Eichel R. Power-to-syngas-an enabling technology
for the transition of the energy system? Production of tailored synfuels and che-
micals using renewably generated electricity. Angew Chem Int Ed
2017;129:5488–98.
[17] Iskov H, Rasmussen CN. Global screening of projects and technologies for Power-to-
Gas and Bio-SNG. Hørsholm: Danish Gas Technology Centre; 2013. < https://www.
energinet.dk/SiteCollectionDocuments/Engelske%20dokumenter/Forskning/
global_screening_08112013_final.pdf > .
Y. Luo et al.
[18] Stempien JP, Ni M, Sun Q, Chan SH. Production of sustainable methane from re-
newable energy and captured carbon dioxide with the use of Solid Oxide
Electrolyzer: a thermodynamic assessment. Energy 2015;82:714–21.
[19] Xie K, Zhang Y, Meng G, Irvine JTS. Direct synthesis of methane from CO2/H2O in
an oxygen-ion conducting solid oxide electrolyser. Energy Environ Sci 2011;4:2218.
[20] Li W, Wang H, Shi Y, Cai N. Performance and methane production characteristics of
H2O-CO2 co-electrolysis in solid oxide electrolysis cells. Int J Hydrogen Energy
2013;38:11104–9.
[21] Chen L, Chen F, Xia C. Direct synthesis of methane from CO2-H2O co-electrolysis in
tubular solid oxide electrolysis cells. Energy Environ Sci 2014;7:4018–22.
[22] Luo Y, Li W, Shi Y, Cao T, Ye X, Wang S, et al. Experimental characterization and
theoretical modeling of methane production by H2O/CO2 co-electrolysis in a tub-
ular solid oxide electrolysis cell. J Electrochem Soc 2015;162:F1129–34.
[23] Lei L, Liu T, Fang S, Lemmon JP, Chen F. The co-electrolysis of CO2-H2O to methane
via a novel micro-tubular electrochemical reactor. J Mater Chem A 2017.
[24] Chen B, Xu H, Ni M. Modelling of SOEC-FT reactor: pressure effects on methanation
process. Appl Energy 2017;185:814–24.
[25] Chen B, Xu H, Chen L, Li Y, Xia C, Ni M. Modelling of one-step methanation process
combining SOECs and Fischer-Tropsch-like reactor. J Electrochem Soc
2016;163:F3001–8.
[26] Melaina MW, Antonia O, Penev M. Blending hydrogen into natural gas pipeline
networks: a review of key issues. NREL; 2013. < http://www.nrel.gov/docs/
fy13osti/51995.pdf > .
[27] Luo Y, Shi Y, Zheng Y, Cai N. Reversible solid oxide fuel cell for natural gas/re-
newable hybrid power generation systems. J Power Sources 2017;340:60–70.
[28] Bao C, Cai N, Croiset E. A multi-level simulation platform of natural gas internal
reforming solid oxide fuel cell–gas turbine hybrid generation system-Part II.
Balancing units model library and system simulation. J Power Sources
2011;196:8424–34.
[29] Bao C, Shi Y, Li C, Cai N, Su Q. Multi-level simulation platform of SOFC-GT hybrid
generation system. Int J Hydrogen Energy 2010;35:2894–9.
[30] Bao C, Shi Y, Croiset E, Li C, Cai N. A multi-level simulation platform of natural gas
internal reforming solid oxide fuel cell–gas turbine hybrid generation system: Part I.
Solid oxide fuel cell model library. J Power Sources 2010;195:4871–92.
[31] Wang Y, Shi Y, Ni M, Cai N. A micro tri-generation system based on direct flame fuel
cells for residential applications. Int J Hydrogen Energy 2014;39:5996–6005.
[32] Ebbesen SD, Graves C, Mogensen M. Production of synthetic fuels by co-electrolysis
of steam and carbon dioxide. Int J Green Energy 2009;6:646–60.
[33] Wendel CH, Gao Z, Barnett SA, Braun RJ. Modeling and experimental performance
of an intermediate temperature reversible solid oxide cell for high-efficiency, dis-
tributed-scale electrical energy storage. J Power Sources 2015;283:329–42.
[34] Shi Y, Luo Y, Cai N, Qian J, Wang S, Li W, et al. Experimental characterization and
modeling of the electrochemical reduction of CO2 in solid oxide electrolysis cells.
Electrochim Acta 2013;88:644–53.
383
Applied Energy 215 (2018) 371–383
[35] Han B, Mo J, Kang Z, Zhang FY. Effects of membrane electrode assembly properties
on two-phase transport and performance in proton exchange membrane electro-
lyzer cells. Electrochim Acta 2016;188:317–26.
[36] Miles MH, Kissel G, Lu PWT, Srinivasan S. Effect of temperature on electrode kinetic
parameters for hydrogen and oxygen evolution reactions on nickel electrodes in
alkaline solutions. J Electrochem Soc 1976;123:332–6.
[37] Ebbesen SD, Jensen SH, Hauch A, Mogensen M. High temperature electrolysis in
alkaline cells, solid proton conducting cells and solid oxide cells. Chem Rev
2014;114:10697–734.
[38] Schaaf T, Grünig J, Schuster MR, Rothenfluh T, Orth A. Methanation of CO2-storage
of renewable energy in a gas distribution system. Energy Sustain Soc 2014;4:1–14.
[39] Luo Y, Shi Y, Li W, Cai N. Comprehensive modeling of tubular solid oxide elec-
trolysis cell for co-electrolysis of steam and carbon dioxide. Energy
2014;70:420–34.
[40] Cownden R, Nahon M, Rosen MA. Exergy analysis of a fuel cell power system for
transportation applications. Exergy Int J 2001.
[41] Ni M, Leung MKH, Leung DYC. Energy and exergy analysis of hydrogen production
by a proton exchange membrane (PEM) electrolyzer plant. Energy Convers Manage
2008;49:2748–56.
[42] Ong KM, Ghoniem AF. Modeling of indirect carbon fuel cell systems with steam and
dry gasification. J Power Sources 2016;313:51–64.
[43] Stoots CM. High-Temperature Co-Electrolysis of H2O and CO2 for Syngas
Production. 2006 Fuel Cell Seminar. Hawaii; 2006.
[44] Graves C, Ebbesen SD, Mogensen M, Lackner KS. Sustainable hydrocarbon fuels by
recycling CO2 and H2O with renewable or nuclear energy. Renew Sust Energy Rev
2011;15:1–23.
[45] Bierschenk DM, Wilson JR, Barnett SA. High efficiency electrical energy storage
using a methane–oxygen solid oxide cell. Energy Environ Sci 2011;4:944–51.
[46] Ni M. Modeling of a solid oxide electrolysis cell for carbon dioxide electrolysis.
Chem Eng J 2010;164:246–54.
[47] Luo Y, Shi Y, Li W, Cai N. Dynamic electro-thermal modeling of co-electrolysis of
steam and carbon dioxide in a tubular solid oxide electrolysis cell. Energy
2015;89:637–47.
[48] Jensen SH, Sun X, Ebbesen SD, Knibbe R, Mogensen M. Hydrogen and synthetic fuel
production using pressurized solid oxide electrolysis cells. Int J Hydrogen Energy
2010;35:9544–9.
[49] Momma A, Takano K, Tanaka Y, Kato T, Yamamoto A. Experimental Investigation
of the effect of operating pressure on the performance of SOFC and SOEC. ECS Trans
2013;57:699–708.
[50] Dokamaingam P, Assabumrungrat S, Soottitantawat A, Sramala I, Laosiripojana N.
Modeling of SOFC with indirect internal reforming operation: comparison of con-
ventional packed-bed and catalytic coated-wall internal reformer. Int J Hydrogen
Energy 2009;34:410–21.