Steam reforming of dimethoxymethane, methanol and dimethyl ether on CuO–ZnO/Γ-Al2O3 catalyst

Article abstract of DOI:10.1134/S0023158417050196

The performance of a СuO–ZnO/γ-Al₂O₃ catalyst for the reactions of methanol, dimethyl ether (DME) and dimethoxymethane (DMM) steam reforming (SR) to hydrogen-rich gas was studied. The catalyst was found to be active and selective for methanol and DMM SR producing hydrogen-rich gas with low content of CO (<1 vol %). It provided complete conversion of methanol and DMM at 300°C, and hydrogen productivity of, respectively, 15 and 16.5 L_H2g_cat^-1h^-1. With the use of physicochemical methods and catalytic experiments, it was shown that the catalyst surface contained the acid sites typical for γ-Al₂O₃, and CuO–ZnO agglomerates, responsible, respectively, for DMM hydration to methanol and formaldehyde, and SR of these compounds to hydrogen-rich gas.

Full text of DOI:10.1134/S0023158417050196

ISSN 0023-1584, Kinetics and Catalysis, 2017, Vol. 58, No. 5, pp. 577–584. © Pleiades Publishing, Ltd., 2017.
Original Russian Text © A.A. Pechenkin, S.D. Badmaev, V.D. Belyaev, E.A. Paukshtis, O.A. Stonkus, V.A. Sobyanin, 2017, published in Kinetika i Kataliz, 2017, Vol. 58, No. 5,
pp. 589–597.
Steam Reforming of Dimethoxymethane, Methanol and Dimethyl
Ether on CuO–ZnO/γ-Al2O3 Catalyst1
A. A. Pechenkina, b, *, S. D. Badmaeva, b, V. D. Belyaeva, E. A. Paukshtisa,
O. A. Stonkusa, b, and V. A. Sobyanina
aBoreskov Institute of Catalysis, Siberian Branch, Russian Academy of Sciences,
pr. Acad. Lavrentieva 5, Novosibirsk, 630090 Russia
bNovosibirsk State University, Pirogova st. 2, Novosibirsk, 630090 Russia
*e-mail: pechenkin@catalysis.ru
Received December 12, 2016
Abstract—The performance of a СuO–ZnO/γ-Al2O3 catalyst for the reactions of methanol, dimethyl ether
(DME) and dimethoxymethane (DMM) steam reforming (SR) to hydrogen-rich gas was studied. The cata-
lyst was found to be active and selective for methanol and DMM SR producing hydrogen-rich gas with low
content of CO (<1 vol %). It provided complete conversion of methanol and DMM at 300°C, and hydrogen
−1
productivity of, respectively, 15 and 16.5
−1. With the use of physicochemical methods and catalytic
h
cat
LH
g
2
experiments, it was shown that the catalyst surface contained the acid sites typical for γ-Al2O3, and CuO–ZnO
agglomerates, responsible, respectively, for DMM hydration to methanol and formaldehyde, and SR of these
compounds to hydrogen-rich gas.
Keywords: hydrogen, dimetoxymethane, methanol, dimethyl ether, steam reforming, copper-zinc catalysts,
fuel cells
DOI: 10.1134/S0023158417050196
According to [18–23], DMM SR (I) proceeds at
250–300°С by consecutive kinetic scheme including
DMM hydration to methanol and formaldehyde (II)
and SR of the formed methanol (III) and formaldehyde
(IV) to hydrogen-rich gas. During DMM SR, a forma-
tion of CO by RWGS reaction (V) can occur as well.
Power units based on low- and high-temperature
proton-exchange membrane fuel cells (PEMFC) are
considered as alternative ecologically benign sources
of electric power for various applications [1–5]. PEM-
FCs are fuelled by hydrogen or hydrogen-rich gas
which can be generated by catalytic steam reforming of
oxygenated organic compounds.
CH3OCH2OCH3 + 4H2O = 8H2 + 3CO2,
(I)
CH3OCH2OCH3 + Н2О = 2СН3ОН + СН2О, (II)
In particular, scientists concentrated considerable
efforts on the synthesis of hydrogen-rich gas for
PEMFC feeding from methanol and DME [6–15]. It
was proved that methanol and DME can be converted
selectively to hydrogen-rich gas at relatively low tem-
perature (250–350°C). Efficient methanol and DME
SR catalysts were developed and used to create “fuel
processors” (i.e., hydrogen-rich gas generators).
СН3ОН + Н2О = 3Н2 + СО2,
СН2О + Н2О = 2Н2 + СО2,
СО2 + Н2 = СО + Н2О.
(III)
(IV)
(V)
It is generally assumed that the solid acid catalysts
(acid sites) are responsible for DMM hydration,
whereas the Cu-based catalysts (Cu-containing sites),
for methanol/formaldehyde SR.
As reported in our recent paper [23], the CuO–
ZnO/γ-Al2O3 catalysts were active and selective for
DMM SR. Comparative analysis of the catalytic per-
formance of γ-Al2O3, ZnO/γ-Al2O3, CuO/γ-Al2O3
and CuO–ZnO/γ-Al2O3 under DMM SR conditions
allowed suggestion that the acid sites of γ-Al2O3 were
responsible for DMM hydration to methanol and
formaldehyde (II), whereas Cu–Zn oxide was repson-
DMM, similarly to methanol and DME, is a read-
ily synthesized oxygenated compound of C1 chemis-
try. Since DMM is noncorrosive and nontoxic mate-
rial, it is easily stored, transported, widely used [16,
17], and considered [18–23] as a promising source of
hydrogen-rich gas for PEMFC feeding applications
(together with methanol and DME).
1
The article was translated by the authors.
577

578
PECHENKIN et al.
sible for the formed methanol (III)/formaldehyde (IV) fraction data were processed using the PowderCell 2.4
SR to hydrogen-rich gas.
program and JCPDS database.
In the present paper which is a continuation of
[23], the results on the DMM SR performance of
γ-Al2O3, 5 wt % ZnO/γ–Al2O3 and 10 wt % CuO–
5 wt % ZnO/γ-Al2O3 are presented in more detail.
Physicochemical characterization of the samples,
including identification of the nature of active sites
on the catalyst surface, was performed using the fol-
lowing techniques: BET, XRD, TPR, FTIR spectros-
copy, HAADF-STEM and EDX. Besides, the data on
the catalytic properties of 10 wt % CuO–5 wt % ZnO/γ-
Al2O3 in methanol SR and DME SR are given. Based
on these data, the analysis is performed whether this
catalyst is suitable for developing a multi-fuel proces-
sor generating hydrogen-rich gas from DMM, metha-
nol and DME for PEMFC feeding applications.
Temperature-programmed reduction (TPR) of the
samples was performed using an STA 409 PC Luxx
(NETZSCH, Germany) derivatograph equipped with
a QMS-200 mass-spectrometer. The samples (∼50 mg)
were heated from room temperature to 400°C at a rate
of 5°C/min in a 5 vol % Н2 + 95 vol % Ar mixture fed
at 140 cm3/min. During these experiments, variation
in the hydrogen concentration was registered by the
mass-spectrometer.
The analysis of the surface composition and com-
positional homogeneity of the supported particles was
performed using a high-angle annular dark field scan-
ning transmission electron microscopy (HAADF-
STEM) images and energy-dispersive X-ray (EDX)
patterns. The experiments were performed using an
electron microscope JEM2200FS (“JEOL,” Japan) fit-
ted with a HAADF detector to obtain images of high
atomic contrast in scanning mode, and with EDX-ana-
lyzer for local microanalysis and EDX-local mapping.
EXPERIMENTAL
10 wt % CuO–5 wt % ZnO/γ-Al2O3 (denoted herein-
after as CuO–ZnO/γ-Al2O3) was prepared by incipient
wetness co-impregnation of γ-Al2O3 (SBET = 200 m2/g,
DMM, methanol and DME steam reforming reac-
tions were studied in a quartz U-shaped continuous-
flow reactor (i.d. 4 mm, wall thickness 1 mm) at 150–
370°С under atmospheric pressure. Catalyst loadings
in the reactor were ∼0.3–0.4 g, granule diameter –
0.25–0.50 mm. The temperature was measured with a
chromel–alumel thermocouple placed in the center of
the catalyst bed. The catalysts were exposed to the feed
compositions (vol %): 14 DMM, 70 H2O and 16 N2;
40 CH3OH, 40 H2O and 20 N2; 20 DME, 60 H2O and
−1
Vpore = 0.7 cm3/g, granule diameter 0.25–0.5 mm)
with aqueous solutions of copper(II) and zinc(II)
nitrates taken at the desired ratio. Catalyst 5 wt %
ZnO/γ-Al2O3 (ZnO/γ-Al2O3 hereinafter) was prepared
by incipient wetness impregnation of γ-Al2O3 by aqueous
solution of zinc(II) nitrate. The samples were dried at
100°С for 2 h and calcined at 400°C for 3 h. γ-Al2O3 (JSC
Katalizator, Russia) was calcined at 600°С for 4 h
before being used as the support.
20 N supplied at GHSV of 10000 h–1 (∼16 L
h–1).
gcat
2
Actual CuO and ZnO loadings in the catalysts was
determined by inductively coupled plasma atomic
emission spectrometry using an Optima 8X00 instru-
ment (Perkin-Elmer, USA). The specific BET surface
areas (SBET) of the catalysts were calculated from the
Prior experiments, the catalysts were reduced in a flow
of 5 vol % Н2 + 95 vol % N2 mixture at 300°С for 1 h.
The inlet and outlet gas compositions were ana-
lyzed by a gas chromatograph Chromos-1000 (Russia)
nitrogen adsorption isotherms at –196°C using a TriS- fitted with TCD/FID detectors and Porapack T and
tar 3000 apparatus (Micromeritics, USA).
CaA columns. The procedure allowed detection of
methanol, DMM, DME, H2, CO, CO2 and methyl
The acidity of the catalysts was determined using
the IR-spectra of adsorbed CO at –196°C [24]. The
catalyst in a form of a disc was placed into the IR cell,
reduced in hydrogen flow, degassed in vacuum at
300–400°С, cooled to –196°С and treated with CO
doses. The IR spectra were recorded on a Shimadzu
FTIR-8300 spectrometer (Shimadzu, Japan) in the
range of 1000–6000 cm–1 with a resolution of 4 cm–1.
According to this procedure, the Brönsted acid sites
(BAS) and Lewis aid sites (LAS) exhibit absorption
bands at 2150–2175 cm–1 and 2175–2240 cm–1, respec-
tively. The strength of BAS is characterized by proton
affinity, that of LAS, by CO adsorption heat (QCO).
The XRD pattern of the samples were recorded on
a ARL X’TRA diffractometer (Thermo Fisher Scien-
tific, Switzerland) with CuKα-radiation and graphite
monochromator, in the scanning range of 2θ = 5°–75°
with a step of 0.01° and sampling time of 5 s. The dif-
formate with a sensitivity of ≤5 × 10–3 vol %. Carbon
imbalance was less than 3 relative percent. The data
were reproducible in several cycles of rising/lowering
the temperature.
The DMM conversion (XDMM, %) and hydrogen
−1
productivity (W , L
h–1) were calculated using
gcat
H2
the following equations:
СN0
СD0 MM − С
2
DMM CN
2
X DMM
=
×100,
CD0 MM
CN0
2
FCH
2
C
N2
WH
=
,
2
100m
KINETICS AND CATALYSIS
Vol. 58
No. 5
2017

STEAM REFORMING OF DIMETHOXYMETHANE
Table 1. Physicochemical characteristics of the catalysts
Content, wt %
579
SBET
,
NLAS
μmol/g
,
QCO,
Catalyst
calculated
CuO ZnO
experiment
m2/g
kJ/mol
CuO
ZnO
γ-Al2O3
ZnO/γ-Al2O3
СuO–ZnO/γ-Al2O3
–
–
10
–
5
5
–
–
9.6
–
4.4
4.5
200
180
160
600
430
240
30
29
28
where C0 and C are the inlet and outlet concentrations
(vol %), respectively, F is the total flow rate of the inlet
reaction mixture (L/h), m is the catalysts weight (g).
exist on the support surface in a form of well dispersed
species, in good agreement with the above XRD data.
Calculated hydrogen uptake at TPR of CuO–ZnO/γ-
Al2O3 proved complete reduction of the supported
CuO to metallic copper.
RESULTS AND DISCUSSION
Figure 2 presents the results of the HAADF-STEM
and EDX studies of fresh and spent CuO–ZnO/γ-
Al2O3 catalysts. The HAADF-STEM image of fresh
CuO–ZnO/γ-Al2O3 (Fig. 2a) demonstrates only large
agglomerates (white patches) of size 80–100 nm.
EDX-maps (Figs. 2b, 2c) indicate simultaneous pres-
ence of copper and zinc. According to EDX micro-
analysis, atomic ratio Cu/Zn ≈ 2, being almost equal
to that value calculated from the chemical composi-
tion data (Table 1). The HAADF-STEM images of
spent CuO–ZnO/γ-Al2O3 (Figs. 2d, 2e) show both
large (∼50–80 nm) and small (∼20 nm) species. Fig-
ure 2f demonstrates EDX-spectrum for the spot with
small species. It is seen that atomic ratio Cu/Zn ≈ 2.
Catalyst Characterization
Table 1 presents the calculated and actual CuO and
ZnO loadings, SBET, acidic properties (LAS concen-
tration (NLAS) and CO adsorption heat (QCO)) for the
studied catalysts. It is seen that the calculated CuO
and ZnO loadings are similar to the actual values. SBET
decreases from 200 m2/g (γ-Al2O3) to 180 m2/g
(ZnO/γ-Al2O3) and 160 m2/g (CuO–ZnO/γ-Al2O3).
γ-Al2O3, ZnO/γ-Al2O3 and CuO–ZnO/γ-Al2O3 are
the solid Lewis acids. The LAS concentration is
600 μmol/g for γ-Al2O3, and 430 and 240 μmol/g for
ZnO/γ-Al2O3 and CuO–ZnO/γ-Al2O3, respectively.
Judging by similar QCO values, the LAS strength in the
studied samples is the same. Note also that some BAS
(<20 μmol/g) were detected on the γ-Al2O3 surface, and
no BAS – on ZnO/γ-Al2O3 and CuO–ZnO/γ-Al2O3.
The above TPR and XRD data allow suggestion
that agglomerates in fresh catalyst consist of well-dis-
persed copper and zinc oxide species, whereas the
used catalyst contains fine-dispersed particles of
metallic copper and zinc oxide.
The XRD patterns of γ-Al2O3, fresh (i.e., prior
reduction and DMM SR experiments) and used
CuO–ZnO/γ-Al2O3 (i.e., after DMM SR experi-
ments) catalysts showed only peaks characteristic for
γ-Al2O3. No diffraction peaks corresponding to Cu,
Zn, CuO and ZnO were observed – most likely, these
species present on γ-Al2O3 surface in highly dispersed
state [25]. According to calculations, the unit cell
parameter of γ-Al2O3 was 7.918 Å, the size of coherent-
scattering region (CSR) equaled 45 Å. Compared to γ-
Al2O3, CuO–ZnO/γ-Al2O3 (fresh and used) demon-
strated somewhat higher values of the lattice parame-
ter and CSR size, which equaled 7.924 and 50 Å,
respectively, more intensive peak at 2θ ≈ 37° and
almost no peak at 2θ ≈ 39°. These facts allow sugges-
tion that copper and zinc cations may insert into alu-
mina lattice to form mixed spinel-type phases.
H2 uptake, a. u.
2
1
Figure 1 presents the H2-TPR profiles of fresh
ZnO/γ-Al2O3 and CuO–ZnO/γ-Al2O3 catalysts. It is
seen that ZnO/γ-Al2O3 uptakes no hydrogen; CuO–
ZnO/γ-Al2O3 demonstrates one hydrogen uptake peak
at ∼190°С. According to [26], such a low reduction
temperature most likely means that CuO particles
50
100 150 200 250 300 350 400
Temperature, °C
Fig. 1. H uptake at TPR of fresh ZnO/γ-Al O (1) and
СuO–ZnO/γ-Al O (2). Dotted line indicates the maxi-
mum of H absorption on СuO–ZnO/γ-Al O .
2
2 3
2
3
2
2 3
KINETICS AND CATALYSIS Vol. 58 No. 5 2017

580
PECHENKIN et al.
(а)
Cu
(b)
Zn
(c)
200 nm
100 nm
100 nm
(d)
(e)
(е)
ZnL
CuL
CuK
ZnK
200 nm
20 nm
0 1.1 2.2 3.3 4.4 5.5 6.6 7.7 8.8 9.9 11
Energy, keV
Fig. 2. HAADF-STEM micro-image (а) and EDX-mapping (distribution) Cu (b) and Zn (c) of a selected patch on fresh CuO–
ZnO/γ-Al O . HAADF-STEM micro-image (d, e) and EDX-spectrum (f) of the selected patch on CuO–ZnO/γ-Al O after
2
3
2 3
DMM SR.
Thus, the results obtained allow conclusion that nol and formaldehyde) at DMM hydration on γ-Al2O3
operating catalyst CuO–ZnO/γ-Al2O3 (i.e., reduced
and exposed to DMM SR conditions) possesses
bifunctional activity. The catalyst surface contains
active sites of two types: LAS (typical for γ-Al2O3 and,
and ZnO/γ-Al2O3. Besides, it presents the equilibrium
values of DMM conversion and methanol/formalde-
hyde concentrations calculated on the assumption that
only DMM hydration (reaction (II)) proceeds in the
most likely, responsible for DMM hydration to meth- system. Obviously, both catalysts demonstrate similar
anol and formaldehyde) and a 20–80 nm agglomer- dependencies. As the temperature increases from 150
ates with atomic ratio Cu/Zn ≈ 2 (responsible, most to 250–280°С, the DMM conversion and methanol
likely, for methanol and formaldehyde SR to hydro- concentration increase and approach equilibrium val-
gen-rich gas).
ues (100% and 23.8 vol %, respectively). Note that
methanol concentration slightly under-reaches the
equilibrium value due to side-reaction of methanol
dehydration to DME (VI). With increasing tempera-
ture, formaldehyde concentration first increases, tends
to reach the equilibrium value, and then decreases. The
latter observation is most likely explained by the pres-
ence of reaction (VII) which was ignored at the calcula-
tion of the equilibrium composition.
DMM SR on γ-Al2O3, ZnO/γ-Al2O3
and CuO–ZnO/γ-Al2O3
Under DMM SR conditions, γ-Al2O3 and ZnO/γ-
Al2O3 catalyze mainly the formation of methanol and
formaldehyde (i.e., DMM hydration, reaction (II)).
Note that at temperatures exceeding 225–250°С,
small amounts (0.5–2.0 vol %) of DME and methyl
formate are formed, most likely, by the following reac- Fig. 3 are shifted towards higher temperatures. This
tions:
Compared to γ-Al2O3, the ZnO/γ-Al2O3 curves in
fact means that γ-Al2O3 is more active in DMM hydra-
tion than ZnO/γ-Al2O3. Since DMM hydration pro-
ceeds on the acid sites, this observation is quite expect-
able, because the acid sites concentration on γ-Al2O3
exceeded that value for ZnO/γ-Al2O3 (Table 1),
2CH3OH = CH3OCH3 + H2O,
2CH2O = HCOOCH3.
(VI)
(VII)
Figure 3 shows the temperature dependencies of
DMM conversion and product concentrations (metha- whereas the TOF values of both catalysts at 175°С
KINETICS AND CATALYSIS
Vol. 58
No. 5
2017

STEAM REFORMING OF DIMETHOXYMETHANE
581
DMM conversion, %
(DMM conversion ≤ 15%) appeared almost the same
Concentration, vol %
50
for (6.0 × 10–3 s–1 for γ-Al2O3 and 5.3 × 10–3 s–1 for
ZnO/γ-Al2O3).
100
80
In contrast to γ-Al2O3 and ZnO/γ-Al2O3, CuO–
ZnO/γ-Al2O3 is active for DMM SR to hydrogen-rich
gas, because this catalyst possesses bifunctional activ-
ity (see previous Section). It contains the surface acid
sites typical for γ-Al2O3, which are responsible for
DMM hydration to methanol and formaldehyde, and
copper-containing species (agglomerates of highly-
dispersed metallic copper and zinc oxide), which cat-
alyze SR of DMM hydration products to hydrogen-
rich gas.
40
30
20
10
0
CH3OH 60
40
20
0
CH2O
Figure 4 shows the temperature dependencies of
DMM conversion and product concentrations (Н2,
СО2, СО and СН3ОН) at DMM SR on СuO–
ZnO/γ-Al2O3, as well as the equilibrium values of
DMM conversion and Н2, СО2 and СО concentra-
tions. The equilibrium concentrations of СН3ОН,
150
200
250
300
Temperature, °C
Fig. 3. Temperature dependencies of methanol and formal-
dehyde concentrations, and DMM conversion at DMM
СН2О, DME and DMM were below 6 × 10–3 vol %
and are not shown in Fig. 4.
hydration on γ-Al O (h) and ZnO/γ-Al O (j). Experi-
2
3
2 3
–1
mental conditions: pressure 1 atm, GHSV 10000 h , inlet
mixture composition (vol %) DMM : H O : N (14 : 70 : 16).
2
2
It is seen that the DMM conversion increases with
increasing temperature and reaches 100% at tempera-
tures above 250°C. The curve of methanol concentra-
tion passes through maximum at 200°С and then
decreases to zero, in good agreement with the sug-
gested kinetic scheme of DMM SR assuming metha-
nol formation as an intermediate product (reactions
(II)–(IV)). Methanol is formed by the DMM hydra-
tion reaction (II) on the acid sites of γ-Al2O3 and is
consumed by SR reaction (III) on the Cu-based sites.
Note that formaldehyde (another intermediate formed
by reaction (II)) is detected only in trace amount
(≤10–2 vol %). There is no contradiction with the sug-
gested kinetic scheme, because formaldehyde is more
active for SR reactions compared to methanol [27].
Points stand for experiment, dotted lines stand for equilib-
rium values.
Concentration, vol %
100
DMM conversion, %
100
80
60
40
80
H2
60
40
Concentrations of Н2, СО2 and СО increase with
increasing temperature (Fig. 4). At 280–300°С, the
Н2 and СО2 concentrations slightly exceed the equilib-
rium values, while the CO concentration (0.9 vol %) is
below the equilibrium one. This observation is most
likely explained by that H2 and CO2 are the primary
products of reactions (III) and (IV), whereas CO is
formed by reaction (V). Obviously, if reaction (V) does
not reach equilibrium during the experiment, the H2
and CO2 concentrations should exceed, and the CO
concentration should be below the corresponding equi-
librium values.
CO2
20
CH3OH
20
CO
0
0
150
200
250
300
Temperature, °C
Fig. 4. Temperature dependencies of H , CO and СО
2
2
concentrations and DMM conversion at DMM SR on
CuO–ZnO/γ-Al O . Experimental conditions: pressure
1 atm, GHSV 10000 h , inlet mixture composition (vol %)
DMM : H O : N (14 : 70 : 16). Points stand for experi-
2
3
–1
2
2
ment, dotted lines stand for equilibrium values.
The results presented in Fig. 4 prove that complete
DMM conversion is reached in the temperature range
of 280–300°С, yielding Н2, СО2 and СО as the main
reaction products. Under these conditions, the cata-
lyst demonstrates the maximum productivity of
−1
DME SR and Methanol SR on CuO–ZnO/γ-Al2O3
H2, CO2 and CO were the main products at DME
and methanol SR. Figures 5 and 6 present the tem-
perature dependencies of DME and methanol conver-
∼16.5
LH
g
h−1,
the CO concentration in the
cat
2
hydrogen-rich gas is below 0.9 vol %.
KINETICS AND CATALYSIS Vol. 58 No. 5 2017

582
Concentration, vol %
PECHENKIN et al.
DME conversion, % Concentration, vol %
Methanol conversion, %
100
70
60
50
40
30
20
10
0
100
100
80
60
40
20
0
H2
80
60
40
20
0
30
20
10
0
H2
H2
CO2
CO
CO2
CO
350
150
200
250
300
200
250
300
Temperature, °C
Temperature, °C
Fig. 6. Temperature dependencies of H , CO and СО
Fig. 5. Temperature dependencies of H , CO and СО
2
2
2
2
concentrations and methanol conversion at methanol SR
concentrations and DME conversion at DME SR on
on CuO–ZnO/γ-Al O . Experimental conditions: pres-
CuO–ZnO/γ-Al O . Experimental conditions: pressure
2
3
2
3
–1
–1
sure 1 atm, GHSV 10000 h , inlet mixture composition
(vol %) CH OH : H O : N (40 : 40 : 20). Points stand for
1 atm, GHSV 10000 h , inlet mixture composition (vol %)
DME : H O : N (20 : 60 : 20). Points stand for experi-
3
2
2
2
2
experiment, dotted lines stand for equilibrium values.
ment, dotted lines stand for equilibrium values.
tion (III), while CO is formed by reaction (V). Clearly,
if reaction (V) does not reach equilibrium during the
experiment, the H2 and CO2 concentration will exceed,
while CO concentration remain below respective equi-
librium values. Note that this tendency is similar to that
observed in DMM SR.
sions, and the outlet H2, CO2 and CO concentrations
at DME and methanol SR on CuO–ZnO/γ-Al2O3. It
is seen that the conversions and product concentra-
tions increase with increasing temperatures in both
reactions. In DME SR experiments (Fig. 5), the DME
conversion and product concentrations considerably
under-reach the equilibrium values in the temperature
interval of 250–350°С. At higher temperatures, the
catalyst shows unstable activity: the DME conversion
and product concentrations fall with time. On the
contrary, the catalyst performance in methanol SR
(Fig. 6) was much better. As the temperature increased
from 150 to 300°С, the methanol conversion attained
the equilibrium value ∼100%. The H2 and CO2 concen-
trations slightly exceeded the equilibrium values, while
the CO concentration (≤1 vol %) remained below the
equilibrium value. This tendency was observed in other
works as well [10, 22], and is attributed most likely to the
fact that H2 and CO2 are the primary products of reac-
The data obtained allow conclusion that CuO–
ZnO/γ-Al2O3 is an efficient catalyst for methanol SR
to hydrogen-rich gas with low CO content. At 300°C,
it provides ∼100% conversion of methanol with a
−1
−1
hydrogen productivity of ∼15
well compet-
LH
g
h ,
cat
ing the most active copper-based2catalysts [12, 28, 29].
Assessment of CuO–ZnO/γ-Al2O3
as a Catalyst for Multi-Fuel Processor
The above data prove CuO–ZnO/γ-Al2O3 to be an
active and selective catalyst for DMM and methanol
SR to hydrogen-rich gas. To estimate whether CuO–
Table 2. Performance of CuO–ZnO/γ-Al2O3 in methanol and DMM SR
Concentration,
vol %
,
WH
2
Compound
Inlet mixture, vol %
T, °C
−1
LH
g
h−1
,
cat
2
H2
CO
CH3OH
DMM
СН3ОН : Н2О : N2 (40 : 40 : 20)
DMM : Н2О : N2 (14 : 70 : 16)
300
280
59.4
60.5
0.9
0.9
15
16.5
−1
Note: reaction conditions—pressure 1 atm, GHSV 10 000 h
.
KINETICS AND CATALYSIS
Vol. 58
No. 5
2017

STEAM REFORMING OF DIMETHOXYMETHANE
ZnO/γ-Al2O3 is promising as a catalyst for multi-fuel ACKNOWLEDGMENTS
583
processor (i.e., a unit capable to generate hydrogen-
rich gas from various fuels under similar conditions
using the same catalyst) applications, we compared its
performance in DMM SR and methanol SR reac-
tions. Table 2 presents the compared parameters: tem-
perature (T) at which ∼100% conversion of DMM and
methanol was observed, the outlet Н2 and СО con-
The authors are grateful to Dr. Zyuzin D.A. for
assistance in a set of experiments. This work was con-
ducted within the framework of budget project
no. 0303-2016-0011 for Boreskov Institute of Catalysis.
REFERENCES
1. Peighambardoust, S.J., Rowshanzamir, S., and
Amjadi, M., Int. J. Hydrogen Energy, 2010, vol. 35,
p. 9349.
2. Chandan, A., Hattenberger, M., El-kharouf, A., Du, S.,
Dhir, A., Self, V., Pollet, B.G., Ingram, A., and Bujal-
ski, W., J. Power Sources, 2013, vol. 231, p. 264.
centrations, hydrogen productivity (
WH
). It is seen
2
that complete conversion of DMM and methanol was
reached at 280 and 300°С, respectively, and the
obtained gas contained almost the same amounts of
hydrogen (∼60 vol %) and CO (<1 vol %). The latter
fact (low CO content) is especially attractive, because
it means essential simplification of hydrogen-rich gas
production scheme for PEM FC feeding. Indeed, the
obtained mixture can be used for high-temperature
PEM FC feeding without any special CO-cleanup
procedures [1]. For low-temperature PEM FC feed-
ing, CO removal to the level of <10 ppm is needed [2].
For this purpose, the processes of partial oxidation
[30] or CO methanation [31, 32] are used; the step of
WGSR becomes unnecessary.
3. Mehta, V. and Cooper, J.S., J. Power Sources, 2003,
vol. 114, p. 32.
4. O'Connell, M., Kolb, G., Schelhaas, K.-P., Wichert, M.,
Tiemann, D., Pennemann, H., and Zapf, R., Chem.
Eng. Res. Des., 2012, vol. 90, p. 11.
5. Idem, R. and Bakhshi, N.N., Ind. Eng. Chem. Res.,
1994, vol. 33, p. 2056.
6. Jones, S. and Hagelin-Weaver, H., Appl. Catal., B,
2009, vol. 90, p. 195.
7. Palo, D., Dagle, R., and Holladay, J., Chem. Rev.,
Hydrogen productivities of CuO–ZnO/γ-Al2O3 in
2007, vol. 107, p. 3992.
methanol and DMM SR are similar as well (15 and
8. Yong, S., Ooi, C., Chai, S., and Wu, X., Int. J. Hydro-
gen Energy, 2013, vol. 38, p. 9541.
−1
cat
16.5
respectively). That is, to provide
LH
g
h−1,
2
9. Li, D., Li, X., and Gong, J., Chem. Rev., 2016, vol. 116,
operation of a 1 kW PEMFC-based power unit fuelled
by DMM or methanol, ca. 50 g of the catalyst is
needed.
p. 11529.
10. Galvita, V., Semin, G., Belyaev, V., Yurieva, T., and
Sobyanin, V., Appl. Catal., A, 2001, vol. 216, p. 85.
Thus, the CuO–ZnO/γ-Al2O3 bifunctional catalyst
is quite efficient for DMM and methanol SR and
shows high promises for using in a multi-fuel proces-
sor generating hydrogen-rich gas for PEM FC-based
power units fueling.
11. Vicente, J., Gayubo, A., Erena, J., Aguayo, A., Olazar, M.,
and Bilbao, J., Appl. Catal., B, 2013, vols. 130–131,
p. 73.
12. Semelsberger, T., Ott, K., Borup, R., and Greene, H.,
Appl. Catal., A, 2006, vol. 309, p. 210.
13. Badmaev, S., Volkova, G., Belyaev, V., and Sobyanin, V.,
React. Kinet. Catal. Lett., 2007, vol. 90, p. 205.
CONCLUSIONS
14. Volkova, G., Badmaev, S., Belyaev, V., Plyasova, L.,
Budneva, A., Paukshtis, E., Zaikovsky, V., and
Sobyanin, V., Stud. Surf. Sci. Catal., 2007, vol. 167,
p. 445.
It is found that CuO–ZnO/γ-Al2O3 is an active and
selective catalyst for DMM and methanol SR producing
hydrogen-rich gas with low CO content (<1 vol %). It 15. Snytnikov, P., Badmaev, S., Volkova, G., Potemkin, D.,
Zyryanova, M., Belyaev, V., and Sobyanin, V., Int. J.
Hydrogen Energy, 2012, vol. 37, p. 16388.
provides complete conversion of methanol and DMM
−1
cat
LH
g
h−1,
and hydrogen productivity of 15 and 16.5
2
16. http://huayan-formaldehyde.com/methylal.html.
17. Zheng, Y., Tang, Q., Wang, T., and Wang, J., Chem.
Eng. Sci., 2015, vol. 134, p. 758.
respectively. The catalyst is promising for using in a
multi-fuel processor generating hydrogen-rich from
various fuels (methanol and DMM).
18. Sun, Q., Auroux, A., and Shen, J., J. Catal., 2006,
vol. 244, p. 1.
Physicochemical characterization and catalytic
experiments proved the presence on the CuO–
ZnO/γ-Al2O3 surface of LAS typical for γ-Al2O3 and
20–80 nm agglomerates comprised of copper and zinc
oxide with atomic ratio Cu/Zn = 2. The former are
responsible for DMM hydration to methanol and
formaldehyde, the latter are responsible for methanol
and formaldehyde SR to hydrogen-rich gas.
19. Fu, Y. and Shen, J., J. Catal., 2007, vol. 248, p. 101.
20. Shen, H., Fu, Y., Sun, Q., Zuo, S., Auroux, A., and
Shen, J., Catal. Commun., 2008, vol. 9, p. 801.
21. Badmaev, S.D., Pechenkin, A.A., Belyaev, V.D.,
Ven’yaminov, S.A., Snytnikov, P.V., Sobyanin, V.A.,
and Parmon, V.N., Dokl. Phys. Chem., 2013, vol. 452,
p. 251.
KINETICS AND CATALYSIS Vol. 58 No. 5 2017

584
PECHENKIN et al.
22. Pechenkin, A., Badmaev, S., Belyaev, V., and 27. Takezawa, N., Catal. Today, 1997, vol. 36, p. 45.
Sobyanin, V., Appl. Catal., B, 2015, vols. 166–167,
28. Oguchi, H., Nishiguchi, T., Matsumoto, T., Kanai, H.,
Utani, K., Matsumura, Y., and Imamura, S., Appl.
Catal., A, 2005, vol. 281, p. 69.
p. 535.
23. Badmaev, S., Pechenkin, A., Belyaev, V., and
Sobyanin, V., Int. J. Hydrogen Energy, 2015, vol. 40,
p. 14052.
29. Kühl, S., Friedrich, M., Armbrüster, M., and Beh-
rens, M., J. Mater. Chem., 2012, vol. 22, p. 9632.
24. Paukshtis, E.A., Infrakrasnaya spektroskopiya v getero-
gennom kislotno-osnovnom katalize (Infrared Spectros-
copy Applied to Heterogeneous Catalysis), Novosi-
birsk: Nauka, 1992.
30. Snytnikov, P.V., Potemkin, D.I., Rebrov, E.V.,
Sobyanin, V.A., Hessel, V., and Schouten, J.C., Chem.
Eng. J., 2010, vol. 160, p. 923.
31. Snytnikov, P.V., Zyryanova, M.M., and Sobyanin, V.A.,
25. Yahiro, H., Murawaki, K., Saiki, K., Yamamoto, T.,
Top. Catal., 2016, vol. 59, p. 1394.
and Yamaura, H., Catal. Today, 2007, vol. 126, p. 436.
32. Konishcheva, M.V., Potemkin, D.I., Badmaev, S.D.,
Snytnikov, P.V., Paukshtis, E.A., Sobyanin, V.A., and
Parmon, V.N., Top. Catal., 2016, vol. 59, p. 1424.
26. Dow, W.-P., Wang, Y.-P., and Huang, T.-J., J. Catal.,
1996, vol. 160, p. 155.
KINETICS AND CATALYSIS
Vol. 58
No. 5
2017