Effect of Sn additives on the CuZnAl-HZSM-5 hybrid catalysts for the direct DME synthesis from syngas

Article abstract of DOI:10.1016/j.apcata.2015.06.030

Abstract Tin-promoted bifunctional Cu-Zn-Al/HZSM-5 catalysts were prepared in order to increase the selectivity of carbon monoxide hydrogenation to dimethyl ether (DME). The catalysts were characterized by H₂-TPR, in-situ XRD, XPS, in-situ XANES and FTIR. The characterization revealed that the electronic properties and surface distribution of Cu in the catalysts were significantly changed by tin addition leading to the decrease in the fraction of ionic Cu species (Cu⁺ and Cu²⁺) and positively charged metallic Cu^δ+ sites. Addition of tin suppressed the water gas shift reaction and decreased the selectivity to undesirable CO₂. As a result, the selectivity to valuable DME and methanol increased for the mildly promoted catalysts. Further increase in Sn loading led to the decrease in the overall rate of carbon monoxide conversion.

Full text of DOI:10.1016/j.apcata.2015.06.030

Accepted Manuscript
Title: Effect of Sn additives on the CuZnAl–HZSM-5 hybrid
catalysts for the direct DME synthesis from syngas
Author: M. Cai V. Subramanian V.V. Sushkevich V.V.
Ordomsky A.Y. Khodakov
PII:
S0926-860X(15)30046-6
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2015.06.030
APCATA 15442
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
24-3-2015
19-6-2015
22-6-2015
Please cite this article as: M.Cai, V.Subramanian, V.V.Sushkevich, V.V.Ordomsky,
A.Y.Khodakov, Effect of Sn additives on the CuZnAlndashHZSM-5 hybrid
catalysts for the direct DME synthesis from syngas, Applied Catalysis A, General
http://dx.doi.org/10.1016/j.apcata.2015.06.030
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.

Effect of Sn additives on the CuZnAl-HZSM-5 hybrid catalysts for
the direct DME synthesis from syngas
M. Caia, V. Subramaniana, V.V. Sushkevichb, V.V. Ordomskya*, A. Y. Khodakova*,
aUnité de catalyse et de chimie du solide (UMR 8181 CNRS), Université Lille 1-ENSCL-EC Lille,
Bat. C3, Cité Scientifique, 59655 Villeneuve d’Ascq, France
bDepartment of Chemistry, Lomonosov Moscow State University, Leninskye Gory 1/3, 119991
Moscow, Russia
Highlights
► Tin-promoted Cu-Zn-Al/HZSM-5 catalysts were studied in dimethyl ether synthesis ►
Presence of tin in reduced state leads to modification of electronic state of Cu ► Tin suppresses
the WGS reaction and consequently reduces CO2 production ► The effect is due to the
deactivation of positively charged Cu species active in WGS ► Further increase of tin loading
leads to decrease of carbon monoxide conversion
Abstract
Tin-promoted bifunctional Cu-Zn-Al/HZSM-5 catalysts were prepared in order to
increase the selectivity of carbon monoxide hydrogenation to dimethyl ether (DME). The
catalysts were characterized by H2-TPR, in-situ XRD, XPS, in-situ XANES and FTIR. The
characterization revealed that the electronic properties and surface distribution of Cu in the
catalysts were significantly changed by tin addition leading to the decrease in the fraction of ionic
Cu species (Cu+ and Cu2+) and positively charged metallic Cuδ+ sites. Addition of tin suppressed
the water gas shift reaction and decreases the selectivity to undesirable CO2. As a result, the
selectivity to valuable DME and methanol increased for the mildly promoted catalysts. Further
increase in Sn loading led to the decrease in the overall rate of carbon monoxide conversion.
Keywords: Selective poisoning; tin; CuO-ZnO-Al2O3 catalyst; DME synthesis

_________________________________________
*Corresponding authors: E-mail address: vitaly.ordomsky@univ-lille1.fr (V.V. Ordomsky);
andrei.khodakov@univ-lille1.fr (A.Y. Khodakov)
1. Introduction
Recently, research toward alternative energy has triggered considerable attention due to
the depletion of oil resources and environmental concerns [1]. To meet the future demand,
utilization of the alternative energy feedstocks such as natural gas, coal and biomass could be
extremely important. These alternative feedstocks can be converted via steam reforming, partial
oxidation or gasification into syngas (H2/CO mixture) which is a valuable intermediate for
manufacturing fuels and chemicals [2–5].
Conversion of syngas to dimethyl ether (DME) has attracted immense interest due to the
merits of DME: (i) an attractive alternative to the diesel due to its low NOx, CO emission and
high cetane number, (ii) chemical feedstock for the synthesis of dimethyl sulfate, methyl acetate
and lower olefins, (iii) LPG substitute and hydrogen source for fuel cell applications [6–8]. DME
is conventionally prepared by dehydration of methanol over acid catalysts (zeolites, γ-Al2O3) [9–
11]. In the indirect DME synthesis, methanol is first synthetized from syngas using mainly Cu-
ZnO based catalysts. The single pass conversion in methanol synthesis is low and limited by
thermodynamics. Methanol is then dehydrated to DME at the second stage.
Direct DME synthesis from syngas overcomes thermodynamic limitations and could
result in higher DME production rates. The direct DME synthesis from syngas is usually
composed of three reaction steps: hydrogenation of CO to methanol, methanol dehydration to
DME and water-gas-shift reaction (WGSR). The relevant chemical equations (1)-(3) are shown
below:

2 CO + 4 H2 = 2 CH3OH
2CH3OH = CH3OCH3 + H2O (2)
H2O + CO = H2 + CO2 (3)
(1)
The combination of reactions (1)–(3) gives the overall equation:
3CO + 3H2 = CH3OCH3 + CO2
The bifunctional catalyst for direct DME synthesis from syngas contains two major
components: a methanol synthesis catalyst and a methanol dehydration catalyst [12]. A high
degree of intimacy and synergism between these components is extremely important to get the
enhanced catalytic performance. The copper based catalyst for methanol synthesis is mature in
terms of understanding its preparation, catalytic performance and reaction conditions [13, 14].
Hence, considerable attention has been recently paid toward the research on the dehydration
component and optimization of its preparation and mixing with the methanol synthesis catalyst
[15–17]. The research conducted so far suggests that a moderate acidity is required for effective
dehydration step in order to avoid lower DME selectivity because of hydrocarbon formation.
During the direct synthesis of DME, the reactions mentioned in Eq (1) and Eq (3) are
catalyzed by the methanol synthesis catalyst. It is also known that copper based catalysts exhibit
high activity in water gas shift reaction, which could result in the ample amount of CO2. Most of
previous works have addressed either optimization of overall activity of bifunctional catalysts for
direct DME synthesis or lowering hydrocarbon production, while very few reports have focused
on reducing carbon dioxide selectivity. Thus, Sun et al. [18] studied the effect of addition of
different amounts of ZrO2 in CuO/ZnO/ZrO2/HZSM-5 instead of commonly used alumina. The
authors explained the increase in the activity and selectivity of the catalyst to DME by Cu+
species stabilized by ZrO2 in the presence of metallic Cu. Moreover, a study reported by Li et al.
[19] gave similar results about addition of La2O3 promoter with better catalytic performance

compared with Cu-ZrO2/γ-Al2O3 catalyst in CO conversion and enhanced DME selectivity due to
the higher dispersion of copper species. Manganese has been also found as effective promoter of
the catalyst activity in DME synthesis [20-21]. The authors have explained the effect by
formation of Cu-Mn spinel oxide with higher dispersion of metallic copper in the reduced
catalyst and higher rate in WGSR. Thus, these works have been more focused on the
development of active catalyst usually by increase of the copper surface area.
Note that CO2 is a major product in this reaction; its production can drastically undermine
the carbon efficiency of the whole process. One of the ways to reduce the CO2 selectivity in
direct DME synthesis could be catalyst promotion. It can be expected that the efficient promoter
would reduce the rate of WGSR, while the rate of DME synthesis could be only slightly affected.
Tin has been used as a promoter for a number of metal catalysts where it principally
affects their hydrogenation activity. There have been numerous studies during the last decade
devoted to noble metal hydrogenation catalysts (Pd, Pt, Ru) modified by tin [22-24]. Particularly,
the alloys of tin with platinum have been shown to significantly enhance the selectivity of the
hydrogenation to the α-β-unsaturated aldehydes. The reason of this catalytic effect is not well
understood and different hypotheses exist: Sn might induce electronic effects on Pt, changing its
adsorption properties or act a site-blocking agent changing the adsorption site of the molecules.
To the best of our knowledge, the effect of tin on Cu hydrogenation catalysts has not been yet
addressed in the literature.
The goal of the present work is to reduce the carbon dioxide selectivity in direct DME
synthesis from syngas by promotion of CuZnAl-HZSM-5 hybrid catalysts with tin. We suggest
that there could be different active species for methanol synthesis and WGSR. Indeed, it has been
shown earlier that activity in WGSR correlates with the amount of Cu+ sites with low redox
potential between Cu+ and Cu2+ during WGSR [25]. At the same time, the activity in methanol

synthesis correlates usually with surface area of metallic copper in the catalyst [26]. Thus, it
might be possible to selectively deactivate the most active positively charged copper sites by
electron donation from tin with increase in the selectivity to methanol and DME. A combination
of characterization techniques (in-situ XRD, TPR, XPS, in-situ XANES and FTIR) were used to
characterize the properties of the promoted catalysts. The catalytic performances in direct DME
synthesis and water gas shift reaction were evaluated in a fixed reactor.
2. Experimental
2.1 Catalyst preparation
The CuO–ZnO–Al2O3 precursor (Cu:Zn:Al = 60:30:10 atomic ratio) was prepared by co-
precipitation at constant pH (ca. 7) and constant temperature (70ºC). A solution containing metal
nitrates ([Cu2+]+[Zn2+]+[Al3+] =1.0 M) and a sodium carbonate solution (1.0 M) were
simultaneously added to a reaction vessel containing 10 mL of deionized water. The suspension
was continuously stirred and kept at the desired pH by adjustment of the relative flow rates of the
two solutions. The final suspension was aged under stirring at 70ºC for 1 h. The precipitate was
filtered off, repeatedly washed with sufficient deionized water to remove residual sodium ions,
dried at 110ºC overnight and then finally calcined at 350ºC in flowing air for 6 h yielding the Cu–
Zn–Al (CZA) catalyst.
The xSn-CZA catalysts were prepared by sequential impregnation using incipient wetness
method at different Sn loading, 0.1, 0.2, 0.5 and 1.0 wt. %, respectively. After the impregnation
with a SnCl2 solution, the promoted catalysts were dried at 90ºC for 12 h followed by calcination
in air at 360 ºC for 5 h.

The HZSM-5 zeolite was prepared by calcination of NH4-ZSM-5 (Zeolyst CBV2314,
SiO2/Al2O3 = 23) in tubular furnace under the flow of air at 550ºC for 6 h (at the heating rate of 2
ºC/min).
The hybrid catalysts were prepared by physically mixing of the powders (under 150
mesh) of the prepared CZA catalyst and zeolite (5:3 wt/wt, 1:1 wt/wt, or 1/2 wt/wt), pressing the
mixture into tablets and finally crushing them into fractions before reaction.
2.2 Characterizations
NH3-TPD profiles were obtained in an Autochem 2910 equipment (Micromeritics) coupled
with a quadrupole mass spectrometer (OmniStar from Balzers Instruments). Prior to ammonia
adsorption, ca. 100 mg of sample was pretreated at 500ºC for 60 min in He. Subsequently,
calibrated pulses of NH3 were introduced at 50ºC until saturation of the sample. The chemisorbed
ammonia was then desorbed by heating from 50ºC to 1000ºC at a heating rate of 10ºC/min. The
m/z = 16 signal in the coupled mass spectrometer was used to obtain the corresponding NH3
desorption profiles.
The temperature-programmed reduction (TPR) profiles were obtained by passing 10%
H2/Ar gas mixture through the catalyst while increasing the temperature at a linear rate. 100 mg
of the samples were loaded into a quartz reactor and pretreated by a He flow (30 ml/min STP) at
100ºC for 2 h. Then He was replaced by the reductive gas H2/He (5 mol% H2 in He) at the same
flow-rate of 30 ml/min. The temperature was increased up to 800ºC at a ramp rate of 5ºC/min.
The effluent gas was analysed by a thermal conductivity detector (TCD).
Copper surface area and dispersion were measured using a N2O decomposition method. The
catalysts (100 mg) were first reduced upon heating to 290°C in a 10% H2/Ar mixture (40
mL/min) with a ramp of 5ºC/min, and the temperature was maintained for 2 h. The amount of
hydrogen consumed is denoted as n1. Then, the reduced samples were cooled to 50°C and purged

with Ar for 30 min. The samples were exposed to 10% N2O/Ar for 1 h to oxidize surface copper
to Cu2O and then flushed with Ar to remove the N2O. Finally, TPR was performed under a 10%
H2/Ar flow to reduce Cu2O back to metallic Cu using a ramp rate of 5ºC/min to 300ºC. The
amount of H2 consumed is denoted as n2. The Cu dispersion and exposed Cu surface area of the
catalysts were calculated as:
2 n2
DCu =
×100%
;
n1
2 n2 N
1.4 1019
SCu=
(m2/g) is the exposed copper surface area per
SCu
DCu
where
(%) is the dispersion of Cu,
gram of the catalyst, N is the Avogadro's constant (6.02 × 1023 atoms/mol), and (1.4 × 1019) is the
number of copper atoms per square meter.
The crystal structure of the catalysts were revealed by X-ray diffraction (XRD) on a Bruker
D8 Advance diffractometer, using CuKα radiation (λ = 1.5406 Å) as X-ray source. A
10º＜2θ＜80º range was scanned with step of 0.02º/s and the 2 s acquisition time.
X-ray photoelectron spectra (XPS) were acquired on a Kratos Axis Ultra DLD apparatus
equipped with a hemispherical analyses and a delay line detector on calcined powder samples and
after in situ reduction (at 290ºC for 6 h for the CZA catalysts). The reduction was carried out in
the atmospheric pressure reaction cell attached to the preparation chamber of the spectrometer
with a heating rate of 2ºC/min like during reaction. The experiments were performed using an Al
monochromatized X-ray source (10 kV, 15 mA) with a pass energy of 40 eV (0.1 eV/step) for
high resolution spectra and a pass energy of 160 eV (1 eV/step) for the survey spectrum in hybrid

mode and slot lens mode respectively. The binding energies were referenced to the Al 2p core
level (74.6 eV).
The in situ time-resolved XRD (λ=0.5 Å) and XAS data were measured at the BM01B
beamline (ESRF, France). For better comparison with the ex-situ XRD results, the diffraction
angles in the in-situ XRD patterns were recalculated for the Cu Kα X-ray source ( λ = 1.5406 Å).
5–10 mg of catalyst was loaded in the quartz capillary reactor (OD=1 mm, wall thickness = 0.020
mm) and pressed from both sides with quartz wool. A Cyberstar gas blower was used to control
the reactor temperature. A gas manifold system and a capillary reactor are described elsewhere
[27-28]. All the catalyst samples were reduced in situ under flow of pure H2 (P = 1 bar) during
the temperature ramping (3ºC/min) up to 290ºC and then cooled down to 40ºC in H2 flow. In situ
XAS spectra at Cu K-edge were acquired in transmission mode. XRD patterns were acquired
using a monochromatic X-ray beam (Si 111 channel-cut monochromator) and a high-resolution
diffractometer. The Scherrer equation has been used for the calculation of sizes of CuO and Cu
crystallites by analysis of the (002) and (111) reflections in XRD profiles, respectively.
FTIR measurements are carried out on a Nicolet Protege 380 spectrometer equipped with a
DTGS detector. Before the measurements all samples where diluted with calcined silica (1:3 by
weight) to avoid the full adsorption of IR beam by the catalyst. About 20 mg of the samples were
pressed in self-supported wafer. The sample was then introduced in a glass cells specially
designed for in-situ FTIR spectroscopy at low temperature (-140ºC) and enabling preliminary
thermal treatments. The activation procedure started with a heating step (5ºC/min) under a
vacuum from ambient temperature to 290ºC with a 3 h plateau in atmosphere of hydrogen with
subsequent evacuation. When the activation procedure was completed, the sample was cooled
down to -140ºC. Carbon monoxide was then gradually introduced into the cell and IR spectra
subsequently recorded. The procedure was repeated until the evolution of adsorbed species

indicated that CO saturation of the surface was reached. Difference spectra were obtained by the
subtraction of the reference spectra from the spectra of the samples with adsorbed CO.
2.3 Catalyst reaction
The DME synthesis reaction was carried out in a fixed-bed stainless-steel tubular reactor
(dint =8 mm) operating at high pressure. The catalyst was crushed and sieved to obtain catalyst
grains of 63–200 µm in diameter. The catalyst loading was typically 0.5 g. Before reaction, the
samples were reduced in hydrogen flow with the flow rate of 30 ml/min. The temperature was
increased to 290ºC with a heating rate of 2ºC/min and then kept at this temperature for 6 h. The
thermocouple was in direct contact with the catalyst, so that the thermocouple measurements
could reflect the temperature inside the reactor. No temperature spike and temperature swings
were observed during the whole catalytic testing. After reduction, the gas flow was switched to
premix the syngas with H2/CO molar ratio of 2/1. The reaction was carried out at 260ºC under a
pressure of 20 bar maintained using a back pressure regulator, and the gas hourly space velocity
(GHSV) of 3600 ml/(g·h). All experimental data were obtained under the steady-state conditions
that were usually maintained at least for 3 h. Carbon monoxide gas contained 5% nitrogen, which
was used as an internal standard for the conversion and selectivity calculation. The products were
sampled from the high pressure side and analysed using an online gas chromatograph with TCD
(H2, CO, CO2 and CH4) and FID (MeOH, DME and other hydrocarbons) detectors. The CO
conversions and product selectivities were calculated on carbon basis.
WGSR was carried out at the temperature range of 260ºC and atmospheric pressure in a
continuous fixed-bed reactor packed with 0.5 g of the Cu-based catalyst. Prior to each
experiment, the catalyst was reduced for 6 h at 290ºC in an H2 flow of 30 ml/ min. Then He
saturated by pure water (0.01 ml/min water in 20 ml/min He) was used as a feed with the space
velocity (GHSV) of 3600 ml/(g·h). Moreover, CO with the flow rate of 10 ml/min was

introduced to the reactor for the water gas shift reaction. The performance of the catalysts was
measured after 2 h time on stream (TOS) for each experiment. The analysis of the reaction
products was carried out on-line using a gas chromatograph equipped with a TCD detector. In the
reproducibility tests, the absolute variation of CO conversion and DME selectivity were
respectively, less than 5 % and 0.5 %.
The carbon monoxide contained 5% of helium, which was used as an internal standard for
calculating carbon monoxide conversion:
in
CO
out
CO
n − n
X CO=
×100%
;
in
n
CO
The product selectivity (S) was reported on carbon basis:
kC nj
S j =
×100%
,
in
out
CO
n − n
CO
in
CO
out
CO
where
,
is the mole flow rate of CO in and out of the reactor;
is the mole flow rate
n
n
nj
of formed product;
– number of carbon atoms in the product.
kC
3. Results and discussion
3.1 Catalyst characterization
The ZSM-5 zeolite used for the preparation of hybrid catalysts has uniform particle size
distribution in the range 5-7 µm (Figure 1a). The zeolite particles consist of agglomerates of
intergrown zeolite crystallites. The acidity of materials was studied by TPD NH3 (Figure 1b).
ZSM-5 exhibits the ammonia desorption at two temperatures: a low-temperature peak at 100-
300ºC and a high-temperature peak at 400-600ºC corresponding respectively to the weaker and
stronger acid sites. The total amount of acid sites determined by TPD of NH3 is 993 µmol/g. The

aluminum content in the sample determined by ICP analysis is higher (1280 µmol/g) due to the
fact that a part of aluminum could be in extraframework positions with significantly weaker
adsorption of ammonia.
The XRD patterns of calcined CZA catalysts with and without Sn are depicted in Fig. 2. The
observed diffraction lines are assigned to CuO and ZnO. The absence of any peaks relevant to
Al2O3 can be correlated to the amorphous nature of the alumina. Also, no extra peaks
corresponding to tin oxide were seen in the XRD patterns. This can be due to low tin content (<1
wt.%) and its high dispersion in matrix CuO-ZnO. Interestingly, tin promotion does not affect
the width of CuO and ZnO peaks. Thus, the size of copper and zinc oxide nanoparticles does not
change after the tin promotion.
To evaluate the reduction behavior, all the catalysts were analyzed by TPR. The reduction
temperature lies between 264 to 280ºC (Table 1, Fig. 3). The parent catalyst, without the Sn
promotion, showed a lowest reduction temperature of 264ºC. An increase in Sn content from 0.1
to 1.0 wt.%, results in higher reduction temperature. The same effect of the promotion with tin
was earlier observed in Pt catalysts [29]. The effect has been explained by a reduction of both
metals (Pt and Sn) with subsequent alloy formation. Reduction of the parent CZA sample results
in the hydrogen consumption (Table 1) similar to the content of CuO in the sample (7.3
mmol/gcat). This indicates almost total reduction of Cu. Promotion by tin leads to lower hydrogen
consumption (Table 1, Fig. 3). This suggests interaction of tin with copper with probable
incorporation of tin species into copper oxide and/orpossible deposition on the surface of copper
oxide with decrease in the catalyst reducibility. Presence of tin in this case might suppress
adsorption and migration of hydrogen on the surface with decrease of the rate of copper
reduction. N2O decomposition (Table 1) shows that the surface area of Cu decreases from 36 to
19 m2/g with the increase in the tin content. The effect can be due to the formation of Cu-Sn

mixed species but without significant segregation of copper with decrease of the amount of active
Cu atoms.
To elucidate the change in the surface composition after Sn doping, XPS analysis was
performed on the CZA and tin containing catalysts pretreated under both oxidizing and reducing
atmospheres. As shown in the Fig. 4a, the Cu 2p3/2 binding energy was observed at 933.0 eV and
the presence of Cu2+ shake up satellite peaks at 942 eV and 962 eV confirms the existence of
mainly Cu2+ state in the catalysts after oxidation [30]. The XPS spectra of the oxidized catalysts
were almost identical. The Cu 2p XPS spectra of xSn-CZA catalysts reduced in H2 at 290ºC are
shown in Fig. 4b. The disappearance of satellite peaks and simultaneous shift of the principal Cu
2p3/2 peak towards lower binding energies upon H2-reduction are indicative of copper reduction
to Cu+ or Cu0 [31-32]. XPS indicates almost complete copper reduction. Somewhat lower
hydrogen consumption in TPR experiments can be due to autoreduction of non-reduced part of
copper oxide into metallic copper at higher temperatures over tin promoted catalysts. The Cu
2p3/2 BE values and widths of Cu+ and Cu0 species are almost identical. These species are
difficult to distinguish from the XPS spectra [33]. However, it is interesting to note that the
maximum of the peak is slightly shifting for tin treated catalysts in the direction of lower BE
(Fig. 4, Table 1). Earlier the shift of the XPS peak to the lower BE has been observed over
reduced noble metal catalysts promoted by tin [34] and was explained by charge transfer from Sn
to the noble metal due to the formation of alloy with tin. The area of Cu 2p3/2 peak decreases
with increase in the tin content (Table 1). This could be due to the analysis of only surface
species by XPS where content of Cu should be lower due to the presence of tin.
For better clarity, the Sn 3d5/2 XPS spectra have been analyzed before and after reduction
of the catalysts. The spectra recorded over fresh catalyst (Fig. 5) show a peak about 486.5 eV
which might be assigned to the oxidized tin (SnIV or SnII) [35]. A new peak appears at 485 eV

after reduction which might be assigned to metallic tin. Thus, the XPS results suggest that tin is
partially reduced to metallic state. However, the relative amount of metallic tin seems to be
significantly smaller in comparison with residual tin oxide species and does not depend on the
amount of tin in the sample (Fig. 5). The XPS results have been supported by in situ XRD and
XANES data of the parent CZA and promoted 1Sn-CZA catalysts.
The in-situ XRD patterns of calcined and reduced catalysts are shown in Fig. 6. The XRD
patterns of the calcined catalyst are indicative of the presence of CuO and ZnO oxides over both
catalysts without effect of tin on XRD patterns. The size of CuO crystallites calculated using the
Scherrer equation was about 8 nm for both samples. The in-situ XRD suggests that CuO
reduction into Cu is complete at 290ºC with formation of similar Cu crystallites over both
catalysts with sizes about 12 nm.
Fig. 7 shows the in-situ Cu K-edge XANES spectra of the non-promoted and tin-promoted
samples in comparison with reference materials before and after reduction. CuO sample shows
the pre-edge absorption at 8984 eV ascribed to a dipole forbidden transition from 1s to 3d orbital
in Cu2+ [36]. The oxidized CZA and 1.0Sn-CZA catalysts exhibit XANES spectra similar to that
of CuO. However, the pre-edge adsorption over CZA catalysts becomes less obvious in
comparison with the reference CuO catalyst. This fact might be explained by smaller sizes of
CuO crystallites in CZA catalysts with incorporation of Cu2+ into the ZnO-Al2O3 framework with
different coordination environment and electronic states in comparison with CuO leading to
distortion of octahedral coordination [37]. Indeed, TEM analysis of CZA catalyst indicates the
high dispersion of Cu in ZnO-Al2O3 matrix [38].
Fig. 7b presents the Cu K-edge spectra of in-situ reduced catalysts in comparison with Cu
foil. Cu foil shows the absorption edge at 8979 eV and two resonance features at 8993 and 9003
eV. The absorption at the edge at 8979 eV corresponds to 1s to 4p electronic transitions [39]. The

XANES shapes for reduced samples CZA and 1.0Sn-CZA are similar to reference foil indicating
the presence of mainly metallic Cu in the sample at the reaction conditions. However, the
intensity of edge absorption at 8979 eV over reference copper foil is significantly lower than over
CZA and 1.0Sn-CZA. The intensity decreases in the order CZA>1.0Sn-CZA>Cu which can be
attributed to the increase in the filling of unoccupied Cu 4p orbital [34]. In the case of CZA
sample reduced Cu should be positively charged due to the strong interaction with oxides of
ZnO-Al2O3 matrix in comparison with the reference Cu foil. The presence of Sn might lead to
partial donation of electrons to 4p orbital with decrease of the transition from 1s to 4p.
FTIR spectroscopy yields valuable information about the interaction of Cu surface sites
with adsorbed CO. Fig. 8 shows IR spectra measured during CO adsorption at low temperature
over CZA and 1.0Sn-CZA catalysts after reduction with increase in CO pressure in the case of
reduced catalysts. At low CO coverage (Fig. 8) both samples reveal the most intense band
centered at 2120 cm-1. The position of these peaks shifts to 2115 cm-1 with increase in CO
pressure. This intense band has been earlier attributed over similar catalyst to the adsorption of
CO over Cu+ cations or small clusters of metallic copper [40, 41]. This band was very intense
over both oxidized catalysts (not shown). Thus, taking also into account high stability of this
carbonyl [42] most probably this band might be assigned to Cu+-CO species. The shift of the
band with increase of CO coverage has been explained by conversion between Cu+–CO species
and Cu+(CO)2 dicarbonyls [42]. CO adsorption over Cu+ species is significantly lower over tin
promoted sample. Although, Cu+ species have not been observed by “bulk methods” like XRD
and XANES, small amounts of Cu+ were identified in the catalysts from FTIR analysis of the
bands of adsorbed of CO. Further increase in CO coverage leads to appearance of the bands at
2189, 2175 and 2156 cm-1. The first two weak bands correspond to the vibrations of CO adsorbed
over Lewis acid sites, Al3+ and Zn2+, respectively [41]. The band at 2156 cm-1 is typical for the

vibrations of hydrogen bonded CO, in particular, bonded with hydroxyl groups [40]. This is result
of CO adsorption over hydroxyl groups of the ZnO-Al2O3 matrix and silanol groups of silica
which has been used for dissolution of highly absorbing IR light Cu samples. Interestingly, the
spectrum of the parent reduced catalyst shows a shoulder at 2135 cm-1 which might be attributed
to the presence of more positively charged cationic copper species. The band at 2135 cm-1 has
been observed earlier during CO adsorption over Cu+ species in ion exchange position of zeolite
ZSM-5 [43]. Isolated Cu+ species in zeolite should be more positively charged in comparison
with Cu+ stabilized by ZnO-Al2O3 matrix. Note that this feature is absent in the FTIR spectra of
CO adsorbed over the promoted catalysts.
To conclude, promotion of the CZA catalyst by tin leads to modification of the properties of
both cationic and metallic copper species. Tin promotion leads to the decrease of ionic copper
fraction (Cu+ and Cu2+) in the reduced catalysts due to the strong interaction between copper and
tin in the catalyst. At the same time, tin inhibits reduction of copper, decreases the surface area
of metallic copper probably by formation Cu-Sn alloys and increases the electronic density on the
metal.
3.2 DME synthesis and WGSR paths in the unpromoted and tin-promoted catalysts
The catalytic activities and selectivities in DME synthesis of the composite CZA/HZSM-5
and Sn-CZA/HZSM-5 catalysts are shown in Figs. 9-12. The carbon monoxide conversions and
selectivities are strongly affected by the presence of tin in CZA. The conversion decreases as a
function of tin content in the catalysts and follows the following order (Fig. 9): CZA/HZSM-5 ≈
0.1Sn-CZA/HZSM-5 > 0.2Sn-CZA/HZSM-5 > 0.5Sn-CZA/HZSM-5 > 1.0Sn-CZA/HZSM-5.
The observed phenomena might be explained by the lower activity of the formed Cu-Sn
nanoparticles containing tin speciesin comparison with copper species in the parent CZA (Table
1). The decrease in the activity of the noble metals after doping by tin has been explained earlier

by blocking of the surface sites by ligand effect and electronic effects [35]. The effect of change
of catalytic performance during hydrogenation/dehydrogenation reactions by tin promotion has
been explained by several authors by electronic interaction between tin and metal [44-45]. In the
case of Pt the influence of tin as promoter has been assigned to donation of electrons to 5d band
leading to deactivation of the active sites [44]. In the case of copper it should result in decrease of
the participation of 3d electrons in bonding with exclusion of these sites from interaction with
CO. The effect of tin as deactivating agent with inhibition of hydrogenolysis has been found to be
similar to deactivation by coke deposition [45]. In both cases coke and tin selectively deactivated
unselective Pt sites.
It is interesting to note (Fig. 9) that small amount of added Sn (0.1 %) almost does not
affect the activity of the catalyst while the tin amount higher than 0.2 wt. % significantly reduces
the catalytic activity. It indicates that Cu sites in the parent CZA samples might be not uniform.
Indeed, FTIR spectra of adsorbed CO over in situ reduced CZA catalysts (Fig. 8) indicated the
presence of Cu+ species in addition to copper metallic phase observed by in situ XRD (Fig. 6)
and XANES (Fig. 7).
DME is the main product of the reaction with the selectivity in the range of 68-76 % (Fig.
10). The selectivity to methanol varies from 3 to 6 %. Methanol is the primary reaction product of
DME synthesis. As expected for primary products, highest methanol selectivity is observed at
low carbon monoxide conversions. Light hydrocarbons are produced with low selectivities (0.1-2
%). All the catalysts showed very significant selectivity to CO2 which increases with carbon
monoxide conversion. Carbon dioxide forms in WGSR. The ratio of DME to CO2 selectivities
over parent catalyst (2.3:1) is higher than the theoretical ratio of 2:1. This suggests that part of
water produced via methanol dehydration during DME synthesis does not participate in WGS
reaction.

Higher tin content in the CZA catalyst results in the gradual increase in the selectivity to
DME from 68 % over the parent CZA/ZSM-5 catalyst to 76 % over the tin promoted catalyst.
Taking into account methanol, the combined selectivity to DME and methanol increases from 71
to 80 %. This effect is mainly the result of lower CO2 selectivity which decreases from 28 to 18
%. The variation of the selectivity might be explained both by lower CO conversion in the tin-
promoted catalysts and by the effect of tin on WGSR.
Tin was introduced by impregnation of the CZA catalyst. We assume that tin introduction
selectively affects only the CZA catalyst. In order to further clarify the effect of tin on the DME
selectivity, the tests have been conducted over the parent CZA/ZSM-5 catalytic systems with
lower CZA content to simulate the lower activity of tin promoted Sn-CZA/ZSM-5. Fig. 10
shows conversion of CO and selectivities over CZA/ZSM-5 with the ratio CZA/ZSM-5 in the
catalyst equal to 5/3, 1/1 and 1/2. The conversion of CO decreases from 69.2 to 51.3 and 29.3 %
as a function of CZA content in the bifunctional catalyst. Lower CO conversion coincides with
the increase in DME selectivity from 68.3 to 69.6 and 70.4 %, respectively. Lower carbon
monoxide conversion results in the lower CO2 selectivity similar to the tin promoted samples and
might be explained by lower amounts of produced water in the system. Interestingly lower carbon
dioxide and higher DME selectivities were observed at the similar conversions over the tin
promoted catalysts compared to the parent catalyst (Figs. 10 and 11). Indeed, at the conversion of
about 29 % over 0.5Sn-CZA/ZSM-5 and CZA/ZSM-5=1/2 the selectivity to DME is 6 % higher
over the tin promoted catalyst. The selectivity to CO2 is lower by approximately 5 % over
promoted systems in comparison with parent catalyst at the same conversions (Fig. 11). In the
case of mild promotion (0.1 and 0.2 wt. % Sn) higher methanol selectivity is also observed which
might be explained by higher water content in the reactor with its significant adsorption over acid
sites of ZSM-5 which explains lower activity in dehydration of methanol to DME. The selectivity

to hydrocarbons has been also slightly increased with higher amount of tin. This effect is due to
the lower conversion of CO and might be observed also over CZA/ZSM-5 with higher zeolite
fraction in the catalyst (Fig. 10).
One of the possible explanations of these observations can be related to the selective
effect of tin addition on WGSR in comparison with CO hydrogenation to methanol. The most
interesting effect in terms of the catalyst selectivity is observed over mildly promoted CZA
catalysts (0.1Sn-CZA). Indeed, almost at the same conversion of CO the combined selectivity to
DME and methanol increases from 71 to 76 %.
Higher CO2 and lower DME selectivities in the parent CZA/HZSM-5 catalyst might be
explained by the presence of the sites in the CZA catalyst more active in WGS reaction in
comparison with tin-promoted catalysts. Probably these are the Cuδ+ sites more positively
charged in ZnO-Al2O3 matrix shown by FTIR (Fig. 8). Earlier investigations of the WGS reaction
over Cu catalysts doped by oxides (ZrO2, CeO2) have shown that Cu sites interacting with O
vacancy with easy redox Cu2+ to Cu+ pairs are responsible for high WGS activity [46,47]. The
presence of tin might result in the preferably reduction of these Cu+ sites with formation of Cu-Sn
alloy with increase of the selectivity to DME. This assumption is supported by the fact that the
selectivity to CO2 over promoted catalysts is always lower by 5 % in the whole conversion range
in comparison with the unpromoted system (Fig. 11). This means that deactivation of most part of
unselective sites take place already at low amount of added tin. Further increase in the amount of
tin leads to the decrease in the fraction of the active surface metallic Cu and lower reaction rate in
carbon monoxide hydrogenation (Table 1). To provide further insights into the effect of tin, the
transformation of CO in the presence of water has been studied over the promoted catalysts. Fig.
12 shows CO conversion of versus tin content in the CZA catalysts. The main reaction products
were CO2 and H2. The conversion of CO significantly decreases with the increase in Sn content.

The presence of 0.1 wt. % of Sn results already in decrease in the conversion of CO from 71 to
60 %. The conversion drops to 23 % at higher tin content. Thus, indeed, the effect of tin over the
CZA catalyst might be explained by lower rate of the WGS reaction. Small amounts of tin
introduced to the CZA catalyst can principally deactivate the sites active in WGS reaction.
Consequently lower CO2 and higher DME selectivities are observed on the tin-promoted
catalysts.
Most of previous research in the field of DME synthesis has addressed the catalyst
promotion in order to decrease hydrocarbon selectivity. For example, promotion of zeolite by
metal oxides (ZrO2, MgO) resulted in increase in the selectivity to DME mainly due to
suppression of hydrocarbons formation [48, 49]. The maximal selectivity to DME at the similar
reaction conditions over CZA/ZSM-5 catalyst promoted by MgO was 64.5 % [48]. The
neutralization of external acid sites of ZSM-5 performed in [38] significantly increased the
stability of the catalyst with only slight increase of the selectivity to 69 % due to higher
hydrophobicity of the catalyst and lower holdup of the water in the system. Application of
zeolite-CZA capsule catalyst exhibited very high selectivity to DME due to suppression of further
dehydration of DME to form alkane/alkene byproducts, although the activity of the composite
catalyst was very low [50]. It seems important to emphasize however those hydrocarbons are not
the main side reaction products of carbon monoxide hydrogenation over bifunctional copper-
zeolite catalysts with mild acidity. CO2 is a major undesirable product which significantly
reduces the carbon efficiency of the direct DME synthesis. Modification of the Cu based catalyst
with tin provides new opportunities to increase the selectivity to DME by suppression of WGS.
4. Conclusion
The structure of copper species in Cu-Zn-Al catalyst has been shown to be significantly
affected by the presence of tin as a promoter. TPR, XPS, in-situ XRD, in-situ XANES and FTIR

measurements reveal the presence of the tin in reduced state and the presence of positively
charged Cu+ copper species in the parent catalyst. The presence of these copper species were
indicated by FTIR.
The catalytic performance in direct DME synthesis from syngas of hybrid CuZnAl/ZSM-5
prepared by grinding depends on the amount of tin. An increase in the amount of the promoter
results in the decrease in the catalyst activity which is accompanied by the increase of the DME
selectivity. The presence of tin suppresses the WGS reaction and consequently reduces CO2
production. The optimum amount of tin in the catalyst has been found to be 0.1 wt.%. At this tin
content, almost no effect on the carbon monoxide hydrogenation rate was observed and the
combined selectivity to DME and methanol increased by 5 % relative to the unpormoted catalyst.
5. Acknowledgments
This research is being performed within the ANR project CATSYN-BIOFUELS (ANR-12-BS07-
0029). The authors gratefully acknowledge the support of the French National Research Agency.
The authors gratefully acknowledge technical support from SNBL. The ESRF is acknowledged
for use of synchrotron radiation. The authors would like to thank Dr. Wouter Van Beek
(SNBL/ESRF) for technical assistance
5.
References
1. A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (6) (2007) 2411-2502.
2. T. Fleisch, E. Iglesia, J. Spivey, Natural Gas Conversion VI, Vol. 136, Elsevier, 2001.
3. S. Lee, Methanol synthesis technology, CRC Press, 1989.
4. J. Rezaiyan, N. P. Cheremisinoff, Gasification technologies: a primer for engineers and
scientists, CRC press, 2005.

5. K. Liu, C. Song, V. Subramani, Hydrogen and syngas production and purification
technologies, Wiley Online Library, 2010.
6. Q. Ge, Y. Huang, F. Qiu, S. Li, Appl. Catal. A 167 (1) (1998) 23-30.
7. T. A. Semelsberger, R. L. Borup, H. L. Greene, J. Power Sources 156 (2) (2006) 497-511.
8. C. Arcoumanis, C. Bae, R. Crookes, E. Kinoshita, Fuel 87 (7) (2008) 1014-1030.
9. F. Yaripour, F. Baghaei, I. Schmidt, J. Perregaard, Catal. Commun. 6 (2) (2005) 147-152.
10. S. Jiang, J.-S. Hwang, T. Jin, T. Cai, W. Cho, Y.S. Baek, S.E. Park, Bull. Korean Chem.
Soc. 25 (2) (2004) 185-189.
11. S.-M. Kim, Y.J. Lee, J. W. Bae, H. Potdar, K.-W. Jun, Appl. Catal. A 348 (1) (2008) 113-
120.
12. K. Sun, W. Lu, F. Qiu, S. Liu, X. Xu, Appl. Catal. A 252 (2) (2003) 243-249.
13. Y. Amenomiya, Appl. Catal. A 30 (1) (1987) 57-68.
14. C. Jiang, D. Trimm, M. Wainwright, N. Cant, Appl. Catal. A 93 (2) (1993) 245-255.
15. F. Ramos, A. Farias, L. E. P. Borges, J. Monteiro, M. A. Fraga, E. F. Sousa-Aguiar, L. G.
Appel, Catal. Today 101 (1) (2005) 39-44.
16. T. Takeguchi, K.-I. Yanagisawa, T. Inui, M. Inoue, Appl. Catal. A 192 (2) (2000) 201-
209.
17. K. S. Yoo, J.-H. Kim, M.-J. Park, S.-J. Kim, O.-S. Joo, K.-D.Jung, Appl. Catal. A 330
(2007) 57-62.
18. K. Sun, W. Lu, F. Qiu, S. Liu, X. Xu, Appl. Catal. A 252 (2003) 243-249.
19. Z. Li, J. Li, Ch. Yang, J. Wu, J. Natur. Chem. 21 (2012) 360-365.
20. X.-J. Tang, J.-H. Fei, Zh.-Y. Hou, X.-M. Zheng, H. Lou, Energ. Fuel 22 (2008) 2877-
2884.
21. J.-H. Fei, X.-J. Tang, Zh.-Y. Huo, H. Lou, X.-M. Zheng, Catal. Comm. 7 (2006) 827-831.

22. M. Toba, S.-I. Tanaka, S.-I. Niwa, F. Mizukami, Z. Koppany, L. Guczi, K.-Y. Cheah, T.-
S. Tang, Appl. Catal. A 2 (189) 243-250.
23. R. Gavagnin, L. Biasetto, F. Pinna, G. Strukul, Appl. Catal. A 38 (2002) 91-99.
24. S. Galvagno, Z. Poltarzewki, A. Donato, G. Neri, R. Pietropaolo, J. Mol. Catal. 35 (1986)
365-373.
25. J.B. Ko, C.M. Bae, Y.S. Jung, D.H. Kim, Catal. Lett. 105 (2005) 157-161.
26. W.R.A.M. Robinson, J.C. Mol, Appl. Catal. 60 (1990) 73-86.
27. M. Rønning, N.E. Tsakoumis, A. Voronov, R.E. Johnsen, P. Norby, W. van Beek, Ø.
Borg, E. Rytter, A. Holmen, Catal. Today 155 (2010) 289-295.
28. H. Karaca, J. Hong, P. Fongarland, P. Roussel, A. Griboval-Constant, M. Lacroix, K.
Hortmann, O.V. Safonova, A.Y. Khodakov, Chem. Commun. 46 (2010) 788-790.
29. P.D. Zgolicz, V.I. Rodríguez, I.M.J. Vilella, S.R. de Miguel, O.A. Scelza, Appl. Catal. A
392 (2011) 208-217.
30. W.L. Dai, Q. Sun, J.F. Deng, D.Wu, Y.H. Sun, Appl. Surf. Sci. 177 (2001) 172-179.
31. P.H. Matter, D.J. Braden, U.S. Ozkan, J. Catal. 223 (2004) 340-351.
32. L. Alejo, R. Lago, M.A. Pena, J.L.G. Fierro, Appl. Catal. A 162 (1997) 281-297.
33. R. T. Figueiredo, A. Martinez-Arias, M. Lopez Granados, J. L. G. Fierro, J. Catal. 178
(1998) 146-152.
34. J.H. Kim, S.M. Choi, S.H. Nam, M.H. Seo, S.H. Choi, W.B. Kim, Appl. Catal. B 82
(2008) 89-102.
35. F. Coloma, A. Sepulveda-Escribano, J.L.G. Fierro, F. Rodrigueze-Reinoso, Appl. Catal.
A 148 (1996) 63-80.
36. E.B. Fox, S. Velu, M.H. Engelhard, Y.H. Chin, J.T. Miller, J. Kropf, C. S. Song, J. Catal.
260 (2008) 358-370.

37. D. Gamarra, G. Munuera, A.B. Hungria, M. Fernandez-Garcia, J.C. Conesa, P. A.
Midgley, X. Q. Wang , J. C. Hanson , J. A. Rodriguez , A. Martinez-Arias, J. Phys.
Chem. C 111 (2007) 11026-11038.
38. V.V. Ordomsky, M. Cai, V. Sushkevich, S. Moldovan, O. Ersen, C. Lancelot, V.
Valtchev, A.Y. Khodakov, Appl. Catal. A 486 (2014) 266-275.
39. Z. Sun, M. Meng, L. Zhang, Y. Zha, X. Zhou, Z. Jiang, S. Zhang, Y. Huang, Int. J.
Hydrogen Energy 37 (2012) 18860-18869.
40. K.I. Hadjiivaniov, G.N. Vayssilov, Adv. Catal. 47 (2002) 307-511.
41. G. Busca, U. Costantino, F. Marmottini, T. Montanari, P. Patrono, F. Pinzari, G. Ramis,
Appl. Catal. A., 310 (2006) 70-78.
42. K. Hadjiivanov, T. Tsoncheva, M. Dimitrov, C. Minchev, H. Knӧzinger, Appl. Catal. A
241 (2003) 331.
43. J. Szanyi, M.T. Paffett, J. Chem. Soc., Faraday Trans. 92(24) (1996) 5165-5175.
44. G. Cocco, S. Enzo, J. Chem. Soc., Faraday Trans 81 (1985) 321.
45. J. Vӧlter, U. Kürschner, Appl. Catal. 8 (1983) 167.
46. J.B. Ko, C.M. Bae, Y.S. Jung, D.H. Kim, Catal. Lett. 105 (2005) 157-161.
47. K. Nishida, D. Li, Y. Zhan, T. Shishido, Y. Oumi, T. Sano, K. Takehira, Appl. Clay Sci.
44 (2009) 211-217.
48. M. Dongsen, Y. Weimin, X. Jianchao, Zh. Bin, S. Qingying, Ch. Qingling, J. Catal. 230
(2005) 140-149.
49. S.-H. Kang, J.W. Bae, K.-W. Jun, H.S. Potdar, Catal. Commun. 9 (2008) 2035-2039.
50. G. Yang, N. Tsubaki, J. Shamoto, Y. Yoneyama,Y. Zhang, JACS 132 (2010) 8129-8136.

Table 1.
Characterization of CZA
Sample
TPR results
XPS analysis Cu 2p 3/2 Cu surface are and
dispersion (N2O
decomposition)
T reduction, H2 consumption
Position,
eV
Area
S,
m2/gcat
36.2
31.2
24.5
18.9
D, %
ºC
(mmol H2/gcat)
CZA
264
265
273
280
6.9
5.0
4.7
4.3
932.8
932.5
932.5
932.4
5572.3
5149.5
5004.4
4817.2
12.2
14.5
12.0
10.2
0.1Sn-CZA
0.5Sn-CZA
1.0Sn-CZA
Fig. 1. SEM micrographs of ZSM-5 (a) and TPD NH3 profiles for zeolite ZSM-5

Fig. 2. XRD patterns of the parent and Sn promoted CuO-ZnO-Al2O3 catalysts
Fig. 3. TPR patterns of the parent and Sn promoted CuO-ZnO-Al2O3 catalysts

Fig. 4. XPS spectra (Cu 2p region) of the parent and Sn promoted CuO-ZnO-Al2O3 catalysts
before (a) and after reduction (b)

Fig. 5. XPS spectra (Sn 3d5/2 region) of the Sn promoted CuO-ZnO-Al2O3 catalysts before (a)
and after reduction (b)

Fig. 6. In situ XRD patterns of CZA or 1.0Sn-CZA measured at before (a) and after (b) reduction
in hydrogen

Fig. 7. In situ XANES spectra of catalysts CZA and 1.0Sn-CZA before (a) and after reduction (b)
in hydrogen in comparison with CuO and Cu foil reference compounds