Plasma catalytic steam methane reforming for distributed hydrogen production

Article abstract of DOI:10.1016/j.cattod.2019.05.015

Steam methane reforming (SMR) via thermal catalytic approach is one of the dominant sources of industrial hydrogen, however it proceeds with slow response and low specific productivity. Here we demonstrate a plasma catalytic SMR for distributed hydrogen p

Full text of DOI:10.1016/j.cattod.2019.05.015

Accepted Manuscript
Title: Plasma Catalytic Steam Methane Reforming for
Distributed Hydrogen Production
Authors: Xiaobing Zhu, Xiaoyu Liu, Hao-Yu Lian, Jing-Lin
Liu, Xiao-Song Li
PII:
S0920-5861(19)30232-9
DOI:
Reference:
https://doi.org/10.1016/j.cattod.2019.05.015
CATTOD 12191
To appear in:
Catalysis Today
Received date:
Revised date:
Accepted date:
23 January 2019
14 April 2019
7 May 2019
Please cite this article as: Zhu X, Liu X, Lian H-Yu, Liu J-Lin, Li X-Song, Plasma
Catalytic Steam Methane Reforming for Distributed Hydrogen Production, Catalysis
Today (2019), https://doi.org/10.1016/j.cattod.2019.05.015
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Plasma Catalytic Steam Methane Reforming for Distrib-
uted Hydrogen Production
Xiaobing Zhu , Xiaoyu Liu, Hao-Yu Lian, Jing-Lin Liu, Xiao-Song Li*
*
Center for Hydrogen Energy and Environmental Catalysis, Dalian University of Technology, Dalian
1
16024, China.
*
Corresponding authors. Email: xzhu@dlut.edu.cn (X.Z.), lixsong@dlut.edu.cn (X.L.)
Graphical abstract
Highlights:
•
•
•
•
Warm plasma catalytic steam methane reforming (SMR) for distributed hydrogen production is demonstrated.
The methane conversion of 90% is achieved at total hydrogen (t-H ) production rate of 2.7 SLM.
The energy efficiency (of CH to t-H ) of 75% and the low energy cost of 1.5 kWh/Nm is achieved.
can be suppressed by SEI and S/C at plasma zone, and completely dismissed at catalyst
2
3
4
2
The formation of C
2
H
x
bed zone.
1

ABSTRACT
Steam methane reforming (SMR) via thermal catalytic approach is one of the dominant sources of
industrial hydrogen, however it proceeds with slow response and low specific productivity. Here we
demonstrate a plasma catalytic SMR for distributed hydrogen production, for which warm plasma by
gliding arc discharge initiates the reaction, followed by Ni-based catalyst in a heat-insulated reactor
without extra heating. In terms of the plasma alone process, specific energy input (SEI), steam/CH
S/C) and total inlet flow rate (F ) contribute to the methane conversion. In parallel, SEI and S/C account
for the decrease in C selectivity hence the increase in selectivity of CO and CO , while with F all
the selectivity is approximately constant. The reaction pathway represented by the selectivity can be in-
fluenced by SEI and S/C rather F . To utilize the heat and active species with the reaction in plasma
zone, Ni/CeO /Al catalyst bed is coupled. For the coupled process, the conversion approaches the
thermodynamic equilibrium values, with the favorable dismissed C selectivity thus the complete se-
lectivity to CO and CO . The coupled process was maintained steady for six hours, and the methane
conversion of 90% at total hydrogen (t-H ) production rate of 2.7 SLM is achieved under optimum con-
4
ratio
(
t
2
H
x
2
t
t
2
2
O
3
2
H
x
2
2
ditions of SEI, S/C, F
t
and gas hourly space velocity (GHSV) of 110 kJ/mol, 3, 3 SLM and 18000 ml·g-
1·
h . Compared to 59% and 2.3 kWh/Nm of the plasma alone process, such a coupled process achieves
-1
3
3
the energy efficiency (of methane to t-H ) of 75% and the low energy cost of 1.5 kWh/Nm . Conse-
2
quently, our approach of plasma catalytic SMR features the merits of rapid response, compact system
and high specific productivity, which can be anticipated for the emerging needs of distributed hydrogen
generation.
KEYWORDS: hydrogen production; steam methane reforming; plasma catalysis; gliding arc
1
. INTRODUCTION
Hydrogen energy in combination with fuel cell electric vehicles (FCEVs) can be well deployed
when the safety concern on the use of hydrogen and the on-demand accessibility are fully addressed.
2

Distributed hydrogen production is crucial to address (to minimize) the safety issue related to hydrogen
storage and delivery. Currently, the blooming establishment of hydrogen station in California, U.S. (35
open and 29 in development) is a sign for the needs of distributed hydrogen production [1].
Fossil fuels are the dominant sources of industrial hydrogen. Recently, methane reforming for hy-
drogen production has been intensively investigated due to the abundance, ease of liquefaction, the
highest hydrogen storage capacity (i.e., hydrogen/carbon ratio), and no carbon-carbon bond [2-6]. Steam
methane reforming (SMR) reaction (R1) possesses theoretically the largest mole fraction of hydrogen in
product gas, compared with partial oxidation [7]and autothermal reforming [8].
⊝
CH + H O → 3H + CO ∆퐻
= ꢀ06 kJ/mol
(R1)
4
2
2
298.15 퐾
Thermocatatlytic approach for SMR for industrial hydrogen production exhibits high selectivity of
target product [9, 10]. Of strong endothermic reaction, thermocatalytic SMR occurs with necessary
external heat supply, and thus large volumeric size and slow response, for which it is usually conducted
for continuous stationary production of hydrogen at large scale.
Non-thermal plasmas [11], e.g., cold plasma and warm plasma, possess the merits of compact,
rapid response, which can be suitable for distributed hydrogen production at small scale. In terms of the
rate and efficiency of hydrogen production, the cold plasmas including dielectric barrier discharge
(DBD) [12-15] and corona discharge [16], are far inferior to the warm plasmas represented by
microwave discharge [17-19] and gliding arc discharge [20-26]. As a consequence, the warm plasma for
methane reforming is a very active subject [17-19, 23, 26]. Ni-based catalyst of a recognized industrial
catalyst exhibits high catalytic performance with low cost [27-29]. So far, no such a work on the SMR
using (gliding arc) plasma catalytic approach is reported.
In the present work, the plasma catalytic SMR for distributed hydrogen production is performed,
for which warm plasma by gliding arc discharge initiates the reaction, followed by Ni-based
(Ni/CeO
2
/Al
2
3
O ) catalyst in a heat-insulated reactor without extra heating. Under the optimal conditions
of plasma alone process, the coupled process performs well in terms of conversion, selectivity and effi-
ciency. The catalyst before and after stability test is observed by techniques of X-ray diffraction (XRD)
3

and X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM) and energy-
dispersive X-ray spectroscopy (EDS).
2
. EXPERIMENTAL PROCEDURES
As described in our previous work [30], the experimental setup is shown in Figure S1. After vapor-
ization, a gas mixture of steam and methane (99.999% purity) is fed into a stainless steel reactor, which
is insulated by ceramic fiber cotton. An input of DC power is applied for the generation of gliding arc
discharge at atmospheric pressure. For gliding arc discharge, the discharge current (I) equals the output
current of the power supply, which is measured by a built-in ampere meter. The discharge voltage (U) is
equal to the voltage of the current limiting resistance subtracted from the output voltage of the DC pow-
er. The plasma power (P) is obtained by the product of discharge voltage and discharge current.
Catalyst is packed and located at post plasma zone in the reactor if applicable. Ten grams catalyst
-1
-1
(
3.5 cm height) is set at gas hourly space velocity (GHSV) of 18000 ml·g ·h after plasma zone. A gap
of 4 cm between high voltage electrode and the entry of catalyst bed is set. The axial temperature profile
of central catalyst bed in the reactor and the reactor wall is measured using two thermocouples. By
incipient wetness impregnation method, the precursors of aqueous Ce(NO
subsequently dispersed with γ-Al support (1-2.5 mm in diameter) for overnight, followed by drying
at 110 C for 6 h and 500 C for 6 h. The catalyst of Ni/CeO /Al with the loading of 11 wt.% Ni and
wt.% Ce is achieved.
As shown in Figure S1, two gas chromatographs (GC) are used in sequence, to analyze gasous
product. For the accuracy, internal standard gas comes with sample gas. Specifically, helium gas as
internal standard gas is used to quantify H , and nitrogen as internal standard gas for CH , CO, CO
hydrocarbon. The first GC (Agilent 1790T) consists of a thermal conductivity detector (TCD) and
a TDX-01 column (2 mm inner diameter, 1.5 m length), which uses H as carrier gas to quantitatively
determine CH , CO, CO . The second GC (Agilent 6890N) consists of a TCD unit (with a column of
3
)
3
and Ni(NO
3
)
2
are
2
O
3
o
o
2
2
O
3
8
2
4
2
,
C
2
H
x
2
4
2
carbon molecular sieve 601), and a flame ionization detector (FID) unit (with a porapak-N column),
4

which uses nitrogen as carrier gas. The TCD unit is used to detect H
analysis of hydrocarbons, e.g., CH and C (C , C and C
fied based on the molar ratio of C /CH , for which CH is measured by the first GC. Using Gibbs free
2
, and the FID unit is conducted for
4
2
H
x
H
2 2
2
H
4
2
H
6
). The C hydrocarbon is quanti-
2
H
x
2
H
x
4
4
energy minimization by HSC Chemistry 7.0 software the thermodynamic equilibrium value (e.g., con-
version) is calculated.
Conversion of CH
4
(푋ꢁꢂꢃ) and H
2
O (푋 ) is calculated using equations E1-E3,
ꢂ ꢅ
ꢄ
푖푛
표푢푡
퐹
−퐹
ꢆꢇ
ꢆꢇ
ꢃ
ꢃ
푋ꢁꢂꢃ
=
× ꢈ00%
× ꢈ00%
(E1)
푖푛
퐹
ꢆꢇ
ꢃ
푖푛
표푢푡
퐹
ꢉ−퐹
ꢇ ꢉ
ꢇ
ꢄ
ꢄ
푋
ꢂ ꢅ
ꢄ
=
(E2)
(E3)
푖푛
퐹
ꢇ
ꢄ
ꢉ
ꢋꢌꢍ
ꢎꢏ
ꢋꢌꢍ
ꢋꢌꢍ
ꢊ
= ꢊ
ꢐ ꢊ
ꢐ ꢀꢊ
ꢂ ꢅ
ꢂ ꢅ
ꢁꢅ ꢁꢅ
ꢄ
ꢄ
ꢄ
ꢎꢏ
ꢋꢌꢍ
where ꢊ and ꢊ
(Subscript i = CH
4
, H
2
O, CO, CO ) represent the gaseous inlet and outlet flow
2
i
i
rates, respectively.
Selectivity and balance is calculated on the basis of carbon atom or hydrogen atom. For the carbon-
퐶
based, the selectivity of CO (S ), CO
2
(푆ꢁꢅꢄ) and C
2
H
x
(푆 ) and the carbon balance (퐵 ) are calculat-
ꢁ
ꢅ
ꢁ
ꢁ
ꢄ
ed using equations E4-E7,
표푢푡
퐹
ꢆꢉ
푆ꢁꢅ = 퐹푖푛
× ꢈ00%
× ꢈ00%
(E4)
(E5)
·ꢑ
ꢆꢇ
ꢆꢇꢃ
ꢃ
표푢푡
퐹
ꢆꢉ
ꢄ
푆ꢁꢅꢄ = 퐹푖푛
·ꢑ
ꢆꢇ
ꢆꢇꢃ
ꢃ
표푢푡
표푢푡
표푢푡
2
(퐹
ꢒ퐹
ꢒ퐹
)
퐶
ꢆꢄꢇꢄ
ꢆꢄꢇꢃ
ꢆꢄꢇꢓ
푆 =
× ꢈ00%
(E6)
(E7)
ꢁ
푖푛
ꢄ
퐹
·ꢑ
ꢆꢇꢃ
ꢆꢇ
ꢃ
퐵ꢁ = 푆ꢁꢅ + 푆ꢁꢅꢄ + 푆ꢁ퐶
ꢄ
where ꢊiꢋꢌꢍ (i = C
H
2
, C
H
4
, C
H
6
) denotes the gaseous outlet flow rate of C
H
. For the hydrogen-
2
2
2
2
x
ꢔ
based, the selectivity of H
2
(푆 ) and C
2
H
x
(푆 ), and the hydrogen balance (퐵 ), are calculated using
ꢂ
ꢁ
ꢂ
ꢄ
ꢄ
equations E8-E10,
표푢푡
퐹
ꢇ
ꢄ
푆ꢂꢄ = 2
× ꢈ00%
(E8)
푖푛
푖푛
퐹
·ꢑ
ꢒ퐹 ꢉ·ꢑ
ꢇ
ꢆꢇꢃ ꢇꢄꢉ
ꢄ
ꢆꢇ
ꢃ
5

표푢푡
표푢푡
표푢푡
2
퐹
ꢒ4퐹
ꢒꢕ퐹
ꢆꢄꢇꢓ
ꢔ
ꢆꢄꢇꢄ
푖푛
ꢆꢄꢇꢃ
푖푛
푆 =
× ꢈ00%
(E9)
ꢁ
ꢄ
4퐹
·ꢑ
ꢒ2퐹
·ꢑ
ꢇ ꢉ ꢇꢄꢉ
ꢄ
ꢆꢇ
ꢆꢇꢃ
ꢃ
퐵ꢂ = 푆ꢂꢄ + 푆ꢁꢔ
(E10)
ꢄ
ꢋꢌꢍ
where ꢊꢂꢄ denotes the outlet flow rate of H
The exothermicity of water-gas shift (WGS) reaction allows CO (with H
but without any extra energy input. Hence, the converted H (from CO via WGS), would be counted
to the present H , for which the sum of the present CO (representing the converted H ) and the present
is called total H (t-H ).
From methane to total H
ed according to equations E11-E12,
2
.
2
O) converted to CO and
2
H
2
2
2
2
H
2
2
2
2
(t-H ), the energy efficiency (η) of and the energy cost (EC) are calculat-
2
표푢푡
표푢푡
ꢖ퐹
푖푛
ꢒ퐹
ꢗ·퐿ꢔ푉
ꢇ
ꢆꢉ
ꢇꢄ
ꢄ
휂 = 퐹
× ꢈ00%
ꢒ푃
ꢆꢇꢃ
(E11)
(E12)
·ꢑ
·퐿ꢔ푉
ꢆꢇ
ꢆꢇꢃ
ꢃ
푃
퐸ꢘ = 퐹
표푢푡
표푢푡
ꢒ퐹
ꢇ
ꢆꢉ
ꢄ
where ꢙ represents plasma power, and ꢚ퐻ꢛ , ꢚ퐻ꢛ
for the lower heating values of H
2
, CH ,
4
ꢂ
ꢁꢂꢃ
ꢄ
respectively.
For the product, dry-basis concentration of H
equations E13-E15,
푑푟푦
ꢄ
푑푟푦
푑푟푦
ꢍ−ꢂꢄ
2
(ꢘꢂ ), CO (ꢘꢁꢅ ), t-H
2
(ꢘ
) is calculated using
표푢푡
퐹
푑푟푦
ꢂꢄ
ꢇꢄ
표푢푡
ꢘ
= 퐹
× ꢈ00%
× ꢈ00%
× ꢈ00%
(E13)
(E14)
(E15)
표푢푡
표푢푡
표푢푡
표푢푡
표푢푡
ꢒ퐹ꢆ
표푢푡
ꢒ퐹ꢆ
ꢇ
ꢄ
ꢒ퐹
ꢒ퐹
ꢒ퐹
ꢒ퐹
ꢆꢉ ꢆ ꢇ
ꢄ ꢄ
ꢇ
ꢆꢇ
ꢆꢉ
ꢇ
ꢄ
ꢃ
ꢄ
ꢄ
ꢃ
ꢓ
표푢푡
퐹ꢆꢉ
푑푟푦
ꢅ
ꢘꢁ = 퐹
표푢푡
표푢푡
표푢푡
표푢푡
표푢푡
표푢푡
ꢒ퐹ꢆ
표푢푡
ꢒ퐹ꢆ
ꢇ
ꢄ
ꢒ퐹
ꢒ퐹
ꢒ퐹
ꢒ퐹
ꢆꢉ ꢆ ꢇ
ꢄ ꢄ
ꢇ
ꢆꢇ
ꢆꢉ
ꢇ
ꢄ
ꢃ
ꢄ
ꢄ
ꢃ
ꢓ
표푢푡
표푢푡
퐹
ꢒ퐹
푑푟푦
ꢍ−ꢂꢄ
ꢇꢄ
표푢푡
ꢆꢉ
표푢푡
ꢘ
= 퐹
표푢푡
표푢푡
표푢푡
표푢푡
ꢒ퐹ꢆ
표푢푡
ꢒ퐹ꢆ
ꢇ
ꢄ ꢓ
ꢒ퐹
ꢒ2퐹
ꢒ퐹
ꢒ퐹
ꢆꢉ ꢆ ꢇ
ꢄ ꢄ
ꢇ
ꢄ
ꢆꢇ
ꢆꢉ
ꢇ
ꢃ
ꢃ
ꢄ
ꢄ
X-ray diffraction (XRD) measurement on the catalysts before and after six-hour stability test is
conducted by an X-ray diffractometer (XRD-6000, Rigaku Smartlab 9) using a Cu Kα radiation (λ=1.54
o
o
o
Å) at 45 kV and 200 mA in the 2θ ranging from 10 to 80 at 6 /min. X-ray photoelectron spectroscopy
(XPS, ESCALAB 250Xi, Thermofisher, U.S.) measurement on the catalysts before and after the stabil-
6

ity test, is conducted using an Al Kα X-ray source (1486.6 eV) operated at 15 kV and 10.8 mA. The
binding energy is calibrated according to the XPS peak of carbon 1s at 284.6 eV. Energy dispersive X-
ray spectroscopy (EDS) for the observation of element mapping of the catalyst before and after stability
test, is carried out on EDAX (AMETEK, U.S.).
Supporting Information describes experimental setup (Figure S1), stability test (Figure S2), and
morphology and element mapping (Figure S3).
3
. RESULTS AND DISCUSSION
For the plasma alone process, the effects of energy density (SEI), steam/CH
inlet flow rate (F ) on SMR reaction were examined. The experimental conditions are listed in Table 1.
4
(S/C) ratio and total
t
With the reaction by plasma, the heat and the active species can be efficiently utilized by the subsequent
catalyst, for which our warm plasma catalytic SMR was developed and performed well. The stability
test was conducted, followed by the observations on the catalysts before and after the stability by XRD,
XPS, and EDS. The energy efficiency and energy cost was analyzed compared to various plasma-related
approaches.
Table 1. Experimental conditions for SMR in plasma zone.
Experiments
SEI
S/C ratio
F
t
I
U
P
/
kJ/mol
/ SLM
2.5
/ mA
/ kV
/ W
SEI effect
S/C effect
80-110
100
2.0
1.5-3.0
2.0
53-77
75
2.8-2.7
2.5
149-206
187
2.5
F
t
effect
110
2.2-3.0
73-95
2.6
188-244
3
.1 Gliding arc plasma
.1.1 Energy density
3
7

With the increase in SEI, the formation of C
2
x
H hydrocarbon (that may cause coking) is suppressed
thus favorably the selectivity is enhanced. SEI contributes to the endothermic SMR reaction (R1),
however not much to water-gas shift (WGS, R2) reaction due to its exothermicity. The gaseous product
consists of mostly CO and H
2
, with small amount of CO
2
and C
2
H
x
(mainly C
2
2
H ).
⊝
ꢀꢜꢝ.ꢈꢞ ꢟ
CO + H O → H + CO ∆퐻
= ꢐꢠꢈ kJ/mol
(R2)
2
2
2
Figure 1. Effect of energy density, SEI, for SMR by warm plasma on (a) conversion, (b) carbon-
based selectivity and balance and (c) hydrogen-based selectivity and balance, under conditions of total
flow rate F of 2.5 SLM, and steam/methane (S/C) ratio of 2. The energy density of plasma, represented
t
by SEI, equals the plasma power over the moles of reactant molecules.
In Figure 1a, with SEI, the methane conversion increases from 40% to 56% and water does from
1
5% to 24%, which is consistent with literatures [24, 31]. With this non-volumetric work (electric,
comparable to heat), the thermodynamic equilibrium of endothermic SMR reaction shifts forward,
hence with the increase in conversion. In Figures 1b and 1c, the mass balance of reaction is shown
8

approximate 100%(±3%). For the carbon-based selectivity, with SEI the decrease in C
approximately equals the increase in CO and CO (69 to 74% and 4 to 7%), which exactly means the
conversion of C to CO and CO . With SEI the increase in CO selectivity is a little larger than that of
CO . For the hydrogen-based selectivity, it has no obvious changes with SEI due to the very low
concentration of C in product gas. In Figure 1c, the hydrogen-based selectivity of C drops 2%
from 5%) thus that of H rises slightly 2% (to 95%).
.1.2 S/C ratio
With S/C ratio the formation of C
2
H
x
(26 to 17%)
2
2
H
x
2
2
2
H
x
2
H
x
(
2
3
2
x
H is inhibited (thus the selectivity increases) and the WGS is
promoted. Figure 2a shows that with the increase in S/C ratio from 1.5 to 3, the methane conversion in-
creases from 46% to 55%, while the water conversion decreases from 27% to 17%. It is attributed to
that the SMR reaction equilibrium shifts forward with the increase in water concentration of reactant.
Figure 2b shows that under the approximate 100% (±3%) carbon balance, with S/C ratio the carbon-
based selectivity of C
the CO selectivity keeps almost steady at 77%. The decrease in C
the increase in CO selectivity, while CO selectivity is constant. Interestingly, with the increase in S/C
2
H
x
reduces from 20% to 13%, the CO
2
selectivity increases from 4% to 10%, and
2
H
x
selectivity approximately equals
2
ratio, the CO selectivity increases because the SMR equilibrium shifts forward, however the CO conver-
sion is enhanced since the WGS equilibrium shifts forward. Consequently, with S/C ratio the CO selec-
tivity goes up for SMR (CO formation), and however the CO conversion rises for WGS (CO consump-
tion), which results in the steady CO selectivity for the total reaction. As shown in Figure 2c, there is no
obvious change in the hydrogen-based selectivity due to the very low concentration of C
2
x
H in product
gas, e.g., the selectivity of C drops 2% (from 4%) thus that of H rises slightly 2% (to 98%).
2
H
x
2
9

Figure 2. Effect of S/C ratio for SMR by warm plasma on (a) conversion, (b) carbon-based
selectivity and balance and (c) hydrogen-based selectivity and balance, under conditions of total flow
ꢎꢏ
rate F
t
of 2.5 SLM and SEI of 100 kJ/mol (the corresponding ꢙ of 188 W).
3
.1.3 Total flow rate
Total flow rate, F
on the inhibition of C
pathway (of methane reforming) herein. Figure 3a shows that with F
t
contributes to the increase in methane conversion, however has no contribution
formation (thus no changes in the selectivity), indicating the identical reaction
from 2.2 to 3 SLM the methane
2
H
x
t
conversion increases from 53% to 60% and that of water from 22% to 27%. Under conditions of the
same SEI and S/C ratio, total flow rate does not influence the chemical equilibrium of SMR. Interesting-
ly, with total flow rate the conversion of methane and water increases, which is attributed to the elonga-
tion of arc and the increase in rotation frequency of arc. In Figures 3b and 3c, the selectivity (of carbon-
based and hydrogen-based) of CO, CO
2
and C
2
x
H keeps basically constant with total flow rate.
1
0

Figure 3. Effect of total inlet flow rate for SMR by warm plasma on (a) conversion, (b) carbon-
based selectivity and balance and (c) hydrogen-based selectivity and balance, under conditions of SEI of
110 kJ/mol, S/C ratio of 2 (the corresponding ꢙ of 188-244 W).
So far, for the plasma alone process, two factors of SEI and S/C ratio contribute to the inhibition of
formation (thus the increase in the selectivity), however F has no such contribution. In terms of
methane conversion three factors of SEI, S/C ratio and F contribute. More interestingly, for the plasma
catalytic process, the methane conversion is further enhanced favorably with the complete
disappearance of C
C
2
H
x
t
t
2
x
H .
3
.2 Gliding arc plasma catalysis
For the plasma alone process, the stream remains high energy generated by gliding arc discharge
with the reaction). To efficiently utilize the heat and the active species sourced from the plasma zone,
(
the catalyst bed is coupled subsequently. Since there is no extra heating and the only heat source from
1
1

the former plasma zone, the catalyst-bed temperature profile drops dramatically for the endothermicity
of SMR reaction. Symbols T and T , and 푥ꢁB represent the axial temperature of catalyst bed and reactor
c
w
wall, and the axial distance from the entry of the catalyst bed, respectively. In Figure 4a, the catalyst-
o
bed temperature, T
c
drops dramatically from 977 to 726 C at the first 2 cm in catalyst-bed height, and
o
slowly down to 651 C at the end of catalyst bed. The temperature of reactor wall, T
w
keeps steady at
o
o
around 605 C. For consistency, the end temperature of catalyst bed, 651 C is set for that of thermody-
namic equilibrium state for SMR.
Figure 4. (a) Axial temperature profile of Ni/CeO
by approaches of warm plasma, warm plasma catalysis, and the thermodynamic equilibrium state in
terms of (b) the conversion, and (c) the selectivity, under optimum conditions of SEI, S/C ratio, F , ꢙ,
2
/Al
2
O
3
catalyst bed, and the comparison of SMR
t
-1
-1
and GHSV of 110 kJ/mol, 3, 3 SLM, 246 W, 18000 ml·g ·h , respectively.
In Figure 4b, compared with that of plasma alone process, the remarkable increase in the conver-
sion of methane from 65 to 90% and water from 21 to 42% is observed, which approaches the thermo-
1
2

o
dynamic equilibrium state value at 651 C of the end temperature of catalyst bed. Apparently, the endo-
thermic SMR reaction performs well at catalyst-bed zone (after plasma zone) with efficient utilization of
the energy sourced from warm plasma, which is evidenced by this remarkable increase in methane con-
version. The strong endothermicity of SMR accounts for the significant drop in catalyst-bed tempera-
ture, i.e., the higher temperature represents the faster reaction kinetics hence the much steeper drop in
temperature (at the first 2 cm). Moreover, Figure 4c shows the decrease in CO selectivity from 77 to
5
2% and the increase in CO
CO , is the evidence for the WGS (R2) over the catalyst. Clearly, at the catalyst bed zone the WGS is
more promoted than the SMR. Inspiringly, the complete conversion of the C (that was generated at
plasma zone) occurs with the hydrogen selectivity rising from 96% to around 100% in Figure 4c.
2
selectivity from 10 to 46%. Hence, the changes in selectivity of CO and
2
2
H
x
Figure 5. (a) Energy efficiency and energy cost, and (b) dry-basis concentration of H
2
, CO, t-H for
2
SMR by approaches of warm plasma and warm plasma catalysis under the same conditions as Figure 4.
1
3

Figure 5a shows the energy efficiency (of methane to hydrogen) of 75% and the energy cost of 1.5
3
3
kWh/Nm for warm plasma catalysis, compared to 59% and 2.3 kWh/Nm for warm plasma. Figure 5b
shows the dry concentration of H (also t-H ) up to 76% (79%) and CO down to 11% for SMR by warm
plasma catalysis. The increase in H (also t-H ) concentration and the decrease in CO concentration
arise from the remarkable increase in methane conversion in Figure 4b, thus the less remained methane,
and the complete conversion of C in Figure 4c.
2
2
2
2
2
H
x
3
.3 Stability test and the characterization on catalysts
Figure S2 shows six hours stability test for warm plasma catalytic SMR. Methane conversion and
water conversion keep steady at 90% and 42%, and the selectivity of CO, CO and H does at 52%, 46%
2
2
and approximate 100%, respectively. The stability test manifests that warm plasma zone and Ni-based
catalyst bed zone work well under the optimum conditions of Figure 4.
Figure 6. XRD measurements on (a) the fresh and (b) the used Ni/CeO
2
/Al
2
3
O catalysts before and after
six-hour stability test under the same conditions of Figure 4.
1
4

Figure 6 shows XRD spectra of the fresh and the used Ni/CeO
six-hour stability test. The assigned peaks of Al , CeO , NiO were observed for the fresh sample. For
the used sample, the weaker peaks of CeO appear, meanwhile the peaks of NiO disappear completely
2
/Al
2
O
3
catalysts before and after
2
O
3
2
2
with the appearance of peaks of metallic Ni at 44.5° (strong), 51.9° and 76.4° (weak). It is evidence that
Ni-based catalyst can be auto-reduced during the reaction, which allows no pre-reduction process [21].
Using Scherrer equation, the calculated Ni nanoparticles are around 17.2 nm in average size.
Figure 7. XPS spectra of Ce 3d for (a) the fresh and (b) the used Ni/CeO
2
/Al
2
3
O catalysts before
and after six-hour stability test under the same conditions of Figure 4.
Figure 7 shows XPS spectra of Ce 3d of the fresh and used catalyst samples. Three peaks are dis-
tinguished by the deconvolution of Ce 3d spectra for both samples. Consistent with literature [32], the
peaks of Ce 3d3/2 are marked as u, and that of Ce 3d5/2 as v1, v
1
', v
2
',v
3
and v '. For the fresh sample, the
3
9
2
9
1
4+
9
2
peaks of v
1
(881.9 eV, 3d 4f ) and v
3
(887 eV, 3d 4f ) are assigned to Ce and u (880.1 eV, 3d 4f ) to
3
+
9
2
Ce [32, 33]. For the used sample after the reaction, the peaks of v
1
'(881.8 eV, 3d 4f ) and v '(887.2 eV,
3
1
5

9
1
4+
9
1
3+
4+
3
d 4f ) are assigned to Ce , and v
2
'(884.9 eV, 3d 4f ) to Ce [34]. During the reaction, the Ce on the
3
+
surface of catalysts can be easily reduced. Before the reaction a very small amount of Ce (indicated by
3
+
the weak peak u) appears in the fresh catalyst, while after the reaction a large amount of Ce (stronger
4
+
3+
peak v
2
') appears in the used catalyst, which is attributed to the reduction of Ce to Ce during the re-
3
+
action. More interestingly, after the reaction, the remarkable increase in the amount of Ce is consistent
4
+
with and accounts for the decrease in the peak intensity of CeO
2
(representing Ce ) of XRD spectrum
in Figure 6.
In Figure S3 EDS mapping is conducted on the fresh and used catalysts. It shows that the distri-
bution of Ce and Ni elements on catalyst surface appears no obvious changes. It is quite consistent with
the results of stability test, i.e., the conversion and the selectivity keep steady for six hours. Also it indi-
cates the good stability of the reaction of warm plasma catalytic SMR.
3
.4 Plasma-based approaches for SMR
Figure 8. Energy efficiency with production rate of H
2
and CO for SMR reaction by various non-
thermal plasma-based reforming approaches under conditions in Table 2.
1
6

In Figure 8 it shows the comparision among plasma-based approaches for SMR. Although total H
2
(t-H
2
) is used anywhere in the present work, herein the term of the production rate of H and CO is used
2
for consistency with references. In general, warm plasmas possess much higher energy efficieny (also
with larger conversion in Table 2) than cold plasmas. The energy efficiency of gliding arc plasma can be
up to 35 times that of DBD (excluding DBD of literature [14] due to the calculation). The calculated
energy efficiency of literature [14] excluded the extra heating (that was used), which needs to be in-
formed. Worth noting is that, in this work warm plasma catalytic SMR is conducted, achieving high
energy efficiency of 75% at large production rate of H and CO of 2.7 SLM, which far exceeds the re-
2
sults of non-thermal plasma-based approaches for SMR for hydrogen production.
Table 2. Conditions and conversions for various non-thermal plasma-based approaches for SMR
for hydrogen production.
Plasma
Catalyst
Ni/CeO /Al
S/C
3
푋ꢁꢂꢃ(%)
90
reference
this work
this work
gliding arc
gliding arc
2
2
O
3
3
65
Ni/Al
2
O
3
[14]
DBD
DBD
2
0.25
1
46
19
4.8
95
80
[
12]
[
[
15]
19]
DBD
microwave
microwave
3
Ni/Al
2
O
3
[18]
3
CONCLUSIONS
In this work plasma catalytic SMR for distributed hydrogen production was conducted, for which
warm plasma by gliding arc discharge initiates the reaction, followed by Ni-based (Ni/CeO
alyst in a heat-insulated reactor without extra heating.
2
/Al
2
3
O ) cat-
1
7

For plasma alone process, energy density SEI and S/C ratio contribute to the methane conversion
due to the forward shift of chemical equilibrium, and inhibit the formation of C resulting in the in-
crease in the selectivity of H , CO, CO . Moreover, total flow rate F contributes to the methane conver-
2
H
x
2
2
t
sion, however has no influence on the selectivity, indicating the identical reaction pathway (of methane
reforming).
To efficiently utilize the heat and the active species sourced from the plasma zone, such a coupled
process of warm plasma catalytic SMR is performed. For the plasma catalytic process, the methane
conversion is further enhanced favorably with the complete disappearance of C
lectivity of CO and CO are the evidence for the WGS (R2) over the catalyst.
Warm plasma catalytic SMR was performed steady for six hours under optimum conditions of SEI,
2
H
x
. The changes in se-
2
-1
-1
S/C ratio, F
t
, and GHSV of 110 kJ/mol, 3, 3 SLM, 18000 ml·g ·h , respectively. It achieved methane
conversion of 90% at production rate of total H
2
(t-H
2
) of 2.7 SLM. Compared with warm plasma alone
3
process, the energy cost of warm plasma catalytic process reduces from 2.3 to 1.5 kWh/Nm , and the
energy efficiency (of methane to t-H
2
) from 59% to 75%.
XRD and XPS measurements were conducted on the catalysts before and after stability test, and it
was observed that Ni/CeO
2
/Al
2
3
O catalysts can be auto-reduced during the reaction.
References
[
1] https://cafcp.org/sites/default/files/h2_station_list.pdf;
https://www.californiahydrogen.org/wp-content/uploads/2017/10/1345-01-Fukumoto_FCE.pdf.
2] B. Bej, N.C. Pradhan, S. Neogi, Production of hydrogen by steam reforming of methane over alumina supported nano-
NiO/SiO catalyst, Catalysis Today, 207 (2013) 28-35.
3] C.-H. Kim, J.-Y. Han, H. Lim, K.-Y. Lee, S.-K. Ryi, Hydrogen production by steam methane reforming in membrane
[
2
[
reactor equipped with Pd membrane deposited on NiO/YSZ/NiO multilayer-treated porous stainless steel, Journal of
Membrane Science, 563 (2018) 75-82.
[
4] Z. Xie, B. Yan, S. Kattel, J.H. Lee, S. Yao, Q. Wu, N. Rui, E. Gomez, Z. Liu, W. Xu, L. Zhang, J.G. Chen, Dry
reforming of methane over CeO -supported Pt-Co catalysts with enhanced activity, Applied Catalysis B: Environmental, 236
2018) 280-293.
5] S.K. Cheah, L. Massin, M. Aouine, M.C. Steil, J. Fouletier, P. Gélin, Methane steam reforming in water deficient
conditions on Ir/Ce0.9Gd0.1 2-x catalyst: Metal-support interactions and catalytic activity enhancement, Applied Catalysis B:
Environmental, 234 (2018) 279-289.
2 3
6] S.M. Sajjadi, M. Haghighi, Impregnation vs. sol-gel and sol-gel-plasma dispersion of nickel nanoparticles over Al O
2
(
[
O
[
employed in combined dry reforming and partial oxidation of greenhouse gases to syngas, International Journal of Hydrogen
Energy, 43 (2018) 15014-15029.
[
7] N. Edwards, S.R. Ellis, J.C. Frost, S.E. Golunski, A.N.J. van Keulen, N.G. Lindewald, J.G. Reinkingh, On-board
hydrogen generation for transport applications: the HotSpot (TM) methanol processor, Journal of Power Sources, 71 (1998)
23-128.
8] T. Aicher, J. Full, A. Schaadt, A portable fuel processor for hydrogen production from ethanol in a 250 W-el fuel cell
system, International Journal of Hydrogen Energy, 34 (2009) 8006-8015.
1
[
1
8

[
9] S.D. Angeli, G. Monteleone, A. Giaconia, A.A. Lemonidou, State-of-the-art catalysts for CH
temperature, International Journal of Hydrogen Energy, 39 (2014) 1979-1997.
10] M.A. Nieva, M.M. Villaverde, A. Monzón, T.F. Garetto, A.J. Marchi, Steam-methane reforming at low temperature on
nickel-based catalysts, Chemical Engineering Journal, 235 (2014) 158-166.
11] T. Nozaki, K. Okazaki, Non-thermal plasma catalysis of methane: Principles, energy efficiency, and applications,
Catalysis Today, 211 (2013) 29-38.
12] X. Zhang, B. Wang, Y. Liu, G. Xu, Conversion of Methane by Steam Reforming Using Dielectric-barrier Discharge,
Chinese Journal of Chemical Engineering, 17 (2009) 625-629.
13] A.M. Montoro-Damas, A. Gómez-Ramírez, A.R. Gonzalez-Elipe, J. Cotrino, Isotope labelling to study molecular
fragmentation during the dielectric barrier discharge wet reforming of methane, Journal of Power Sources, 325 (2016) 501-
05.
14] Q. Liu, H. Zheng, R. Yang, G. Pan, Experimental study on chemical recuperation using hybrid dielectric barrier
4
steam reforming at low
[
[
[
[
5
[
discharge-catalytic methane-steam reforming, Proceedings of the Institution of Mechanical Engineers, Part A: Journal of
Power and Energy, 228 (2014) 451-461.
[
15] J.-L. Liu, R. Snoeckx, M.S. Cha, Steam reforming of methane in a temperature-controlled dielectric barrier discharge
reactor: the role of electroninduced chemistry versus thermochemistry, Journal of Physics D: Applied Physics, 51 (2018)
85201-338510.
16] M.-w. Li, G.-h. Xu, Y.-l. Tian, L. Chen, H.-f. Fu, Carbon Dioxide Reforming of Methane Using DC Corona Discharge
Plasma Reaction, The Journal of Physical Chemistry A, 108 (2004) 1687-1693.
17] D. Czylkowski, B. Hrycak, M. Jasiński, M. Dors, J. Mizeraczyk, Microwave plasma-based method of hydrogen
production via combined steam reforming of methane, Energy, 113 (2016) 653-661.
18] Y.-F. Wang, C.-H. Tsai, W.-Y. Chang, Y.-M. Kuo, Methane steam reforming for producing hydrogen in an
atmospheric-pressure microwave plasma reactor, International Journal of Hydrogen Energy, 35 (2010) 135-140.
19] D.H. Choi, S.M. Chun, S.H. Ma, Y.C. Hong, Production of hydrogen-rich syngas from methane reforming by steam
microwave plasma, Journal of Industrial and Engineering Chemistry, 34 (2016) 286-291.
20] K. Li, J.-L. Liu, X.-S. Li, X.-B. Zhu, A.-M. Zhu, Post-plasma catalytic oxidative CO
based catalysts, Catalysis Today, 256 (2015) 96-101.
21] K. Li, J.-L. Liu, X.-S. Li, X. Zhu, A.-M. Zhu, Warm plasma catalytic reforming of biogas in a heat-insulated reactor:
Dramatic energy efficiency and catalyst auto-reduction, Chemical Engineering Journal, 288 (2016) 671-679.
22] J.-L. Liu, Z. Li, J.-H. Liu, K. Li, H.-Y. Lian, X.-S. Li, X. Zhu, A.-M. Zhu, Warm-plasma catalytic reduction of CO
with CH , Catalysis Today, (2018) doi: 10.1016/j.cattod.2018.1005.1046.
23] Y.N. Chun, Y.C. Yang, K. Yoshikawa, Hydrogen generation from biogas reforming using a gliding arc plasma-catalyst
reformer, Catalysis Today, 148 (2009) 283-289.
24] N. Rueangjitt, T. Sreethawong, S. Chavadej, H. Sekiguchi, Plasma-catalytic reforming of methane in AC microsized
gliding arc discharge: Effects of input power, reactor thickness, and catalyst existence, Chemical Engineering Journal, 155
2009) 874-880.
25] F. Zhu, H. Zhang, X. Yan, J. Yan, M. Ni, X. Li, X. Tu, Plasma-catalytic reforming of CO
catalysts in a rotating gliding arc reactor, Fuel, 199 (2017) 430-437.
26] K. Pornmai, S. Suvachitanont, S. Chavadej, Reforming of CO
discharge for hydrogen production, International Journal of Green Energy, 15 (2018) 441-453.
27] B. Li, Z. Xu, F. Jing, S. Luo, N. Wang, W. Chu, Improvement of catalytic stability for CO
copper promoted Ni-based catalyst derived from layered-double hydroxides, Journal of Energy Chemistry, 25 (2016) 1078-
085.
28] F. Wang, B. Han, L. Zhang, L. Xu, H. Yu, W. Shi, CO2 reforming with methane over small-sized Ni@SiO
with unique features of sintering-free and low carbon, Applied Catalysis B: Environmental, 235 (2018) 26-35.
29] E.-h. Yang, Y.S. Noh, G.H. Hong, D.J. Moon, Combined steam and CO reforming of methane over La1-xSr
perovskite oxides, Catalysis Today, 299 (2018) 242-250.
30] H.-Y. Lian, X.-S. Li, J.-L. Liu, X. Zhu, A.-M. Zhu, Oxidative pyrolysis reforming of methanol in warm plasma for an
on-board hydrogen production, International Journal of Hydrogen Energy, 42 (2017) 13617-13624.
31] W. Piavis, S. Turn, An experimental investigation of reverse vortex flow plasma reforming of methane, International
Journal of Hydrogen Energy, 37 (2012) 17078-17092.
32] X. Liao, Y. Zhang, M. Hill, X. Xia, Y. Zhao, Z. Jiang, Highly efficient Ni/CeO
hydrogenation of maleic anhydride, Applied Catalysis A: General, 488 (2014) 256-264.
33] A. Löfberg, J. Guerrero-Caballero, T. Kane, A. Rubbens, L. Jalowiecki-Duhamel, Ni/CeO
vectors for the chemical looping dry reforming of methane for syngas production, Applied Catalysis B: Environmental, 212
2017) 159-174.
34] L.-Y. Lin, H. Bai, Promotional effects of manganese on the structure and activity of Ce–Al–Si based catalysts for low-
3
[
[
[
[
[
2
reforming of methane over Ni-
[
[
2
4
[
[
(
[
2
-rich biogas over Ni/γ-Al
-containing natural gas with steam in AC gliding arc
reforming of methane by
2 3
O
[
2
[
2
1
[
2
catalysts
NiO
3
[
2
x
[
[
[
2
catalyst for the liquid phase
based catalysts as oxygen
[
2
(
[
temperature oxidation of acetone, Chemical Engineering Journal, 291 (2016) 94-105.
ACKNOWLEDGMENT
1
9

This work is supported by National Natural Science Foundation of China (11705019).
Additional information
Supporting Information.
Additional data, Figures S1-S3. The Supporting Information is available free of charge via the Internet at
2
0