Bifunctional catalysts based on colloidal Cu/Zn nanoparticles for the direct conversion of synthesis gas to dimethyl ether and hydrocarbons

Article abstract of DOI:10.1016/j.apcata.2018.03.008

Hybrid catalysts were prepared using well-defined, colloidal Cu/Zn-based nanoparticles as building units. The nanoparticles were immobilized on acidic supports (i.e., γ-Al₂O₃, HZSM-5, and HY) to yield a series of bifunctional catalysts with a close proximity of active sites for both methanol synthesis and its further conversion to dimethyl ether (DME) or hydrocarbons (HCs). By this model kit principle, a high comparability of the bifunctional catalysts was ensured. The catalysts were characterized in depth regarding their structure and catalytic performance in the conversion of CO-rich synthesis gas. In situ XAS studies demonstrated the formation of the active phase under reducing conditions. The present study revealed important material parameters to control activity and selectivity of the bifunctional catalysts either towards DME or liquefied petroleum gas (LPG) products in the direct conversion of simulated biomass-derived synthesis gas. In particular, Cu loading, pore structure and Si:Al ratio were investigated.

Full text of DOI:10.1016/j.apcata.2018.03.008

Accepted Manuscript
Title: Bifunctional catalysts based on colloidal Cu/Zn
nanoparticles for the direct conversion of synthesis gas to
dimethyl ether and hydrocarbons
Authors: M. Gentzen, D.E. Doronkin, T.L. Sheppard, J.-D.
Grunwaldt, J. Sauer, S. Behrens
PII:
S0926-860X(18)30118-2
DOI:
Reference:
https://doi.org/10.1016/j.apcata.2018.03.008
APCATA 16582
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
21-12-2017
6-3-2018
13-3-2018
Please cite this article as: Gentzen M, Doronkin DE, Sheppard TL, Grunwaldt J-D,
Sauer J, Behrens S, Bifunctional catalysts based on colloidal Cu/Zn nanoparticles for
the direct conversion of synthesis gas to dimethyl ether and hydrocarbons, Applied
Catalysis A, General (2010), https://doi.org/10.1016/j.apcata.2018.03.008
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.

Bifunctional catalysts based on colloidal Cu/Zn nanoparticles for the direct conversion
of synthesis gas to dimethyl ether and hydrocarbons
a
a,b
T. L. Sheppard,a,b J.-D. Grunwaldt,a,b J. Sauer, and S.
a
M. Gentzen, D. E. Doronkin,
Behrens*a
a
Institute of Catalysis Research and Technology, Karlsruhe Institute of Technology (KIT),
Herrmann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany; E-mail:
silke.behrens@kit.edu
b
Institute for Chemical Technology and Polymer Chemistry, Karlsruhe Institute of
Technology (KIT), Engesserstr. 20, 76131 Karlsruhe, Germany
*
Corresponding author: silke.behrens@kit.edu
Graphical abstract
Highlights

Using a model kit principle, well-defined Cu/Zn-based nanoparticle building
units are employed for the preparation of specific bifunctional catalysts for
syngas conversion to either dimethyl ether or hydrocarbons.
By this approach, the effects of preparation history are reduced and high
comparability of the bifunctional catalysts is enabled.
Bifunctional catalysts affording the close proximity of two catalytic functions
are obtained by subsequently depositing the nanoparticles on different
acidic catalysts.




The formation of the active phase during in situ activation is monitored by in
situ X-ray absorption spectroscopy.
The present study reveals the importance of Cu loading, Cu to acidic sites
ratio and the accessibility of acid sites on the bifunctional catalysts to
control activity and selectivity either towards DME or hydrocarbons in the
direct conversion of simulated biomass-derived synthesis gas.
Abstract
Hybrid catalysts were prepared using well-defined, colloidal Cu/Zn-based nanoparticles as
building units. The nanoparticles were immobilized on acidic supports (i.e., γ-Al , HZSM-
2
O
3

5
, and HY) to yield a series of bifunctional catalysts with a close proximity of active sites for
both methanol synthesis and its further conversion to dimethyl ether (DME) or hydrocarbons
HCs). By this model kit principle, a high comparability of the bifunctional catalysts was
(
ensured. The catalysts were characterized in depth regarding their structure and catalytic
performance in the conversion of CO-rich synthesis gas. In situ XAS studies demonstrated the
formation of the active phase under reducing conditions. The present study revealed important
material parameters to control activity and selectivity of the bifunctional catalysts either
towards DME or liquefied petroleum gas (LPG) products in the direct conversion of simulated
biomass-derived synthesis gas. In particular, Cu loading, pore structure and Si:Al ratio were
investigated.
Keywords: Single-step dimethyl ether synthesis, nanoparticles, zeolites, bifunctional catalysts
1
. Introduction
A promising route to second generation biofuels comprises the conversion of lignocellulosic
biomass to synthesis gas via gasification, and subsequent upgrading to valuable products.1,2 In
this context, the syntheses of DME or hydrocarbons via the methanol route are particularly
interesting. Apart from its direct use for domestic applications or as a clean diesel fuel, DME
is applied as an intermediate, e.g., in the production of further base chemicals and gasoline
range hydrocarbons. Traditionally, DME is produced from syngas in a two-step process,
where methanol is generated over Cu-based catalysts in the first stage, followed by
subsequent dehydration over inorganic solid acids (e.g., γ-Al
2
3
O or zeolites) in the second
stage. Alternatively, DME can be obtained in a single-step conversion from syngas which
involves the following reactions:
CO + 2 H ⇌ CH OH
(1)
2
3
2
CH OH ⇌ CH OCH + H O
(2)
3
3
3
2
CO + H O ⇌ CO + H
(3)
2
2
2
In addition, DME production from syngas is thermodynamically favored over methanol and
enables higher CO conversions. The overall conversion of syngas-to-DME (STD) is described
by Equation 4.
3
CO + 3 H ⇌ CH OCH + CO
2
(4)
2
3
3
Gasification of biomass-derived feedstocks provides syngas with a H
0
2
:CO ratio in the range of
.7 to 1.0 which is below the optimum ratio for methanol synthesis (Eq. 1).2,3 In the present
study, CO-rich syngas (CO:H
2
ratio 1:1) was employed to simulate biomass-derived
feedstocks for the STD process (Eq. 4).
The STD process has many technical and economic advantages, provided that a suitable
4
,5
catalyst exists. Typically, the two catalysts are combined by simple physical mixing.
Cu/ZnO (Al ) is industrially applied as a catalyst in methanol synthesis and has also been
2
O
3

used as the methanol active component in the STD process for several years.6,7,8 The ideal
solid acid catalyst for methanol dehydration should exhibit appropriate acid sites, high
stability, hydrophobic surface, low cost, high activity, and good selectivity for the desired
products. In particular, alumina and zeolites have been employed for methanol dehydration to
DME.9
,10,11,12,13
γ-Al
2
3
O is known as a very efficient dehydration catalyst with high DME
selectivity, low cost, excellent lifetime, and high mechanical resistance. The surface of γ-
Al
2
O
3
remains with an excess of positive charge which is compensated by hydroxyl anions
-
(OH ). The hydroxyl anions form weakly acidic sites but desorption at high temperatures
creates coordinatively unsaturated metal cations and oxygen anions that can act as Lewis acid
and base sites, respectively. Although the cubic, defect spinel-type γ-Al reveals a
2
O
3
remarkable selectivity to DME, zeolites have been reported to exhibit a better catalytic
activity and stability. In this context, H-ZSM-5 is of particular interest since it combines
strong acidity, high density of acid sites and medium size pores which prevent coking without
significantly hindering the diffusion of molecules involved in the DME synthesis. In zeolites,
the number of acid sites for methanol dehydration can be further adjusted via the Si/Al ratio.
In contrast to γ-Al
2
3
O , molecular sieves (e.g., SAPO-34, HY, H-ZSM-5) can be further
employed in hybrid catalysts for the production of hydrocarbons from syngas where, due to a
so-called “drain-off” mechanism, no thermodynamic constraints are imposed on the overall
syngas feed conversion.
The design of catalysts remains a crucial issue for enhancing catalytic efficiency. In the STD
process, the large distance between the two catalytically active sites, a rather low activity, and
low long-term stability have been described as the major drawbacks of physically mixed
1
4
catalytic systems. Diffusion of intermediates (e.g. methanol) through the interface between
methanol synthesis catalyst and acid functions are an essential part of the reaction scheme and
synergistic effects were observed for single bifunctional entities, if the two catalytically active
components were finely dispersed and maintained in close contact.15,16 A conventional
bifunctional catalyst can be prepared by mixed metal impregnation, co-precipitation
impregnation, or co-precipitation sedimentation of the dehydration catalyst.15,16,17,18 However,
the features and activity of the bifunctional catalyst strongly depend on the preparation
history. They often exhibit complex structures with broad size and shape distributions of the
active particles impeding fundamental studies on the influence of the various structural
parameters on the catalytic performance.19,20 In this context, catalysts derived from well-
defined nanoparticle building units may help to reduce this complexity and to contribute to a
2
0,21
more fundamental understanding.
Colloidal nanoparticles have been employed as quasi-
homogeneous catalysts for liquid-phase methanol and DME synthesis by dispersing the
nanoparticles in the reaction medium.22,23,24,25,26 Here, long-chain surfactants or ligands are
typically adsorbed on the nanoparticle surface to prevent agglomeration, which influences the
catalytic properties of the particles. Following the precursor concept, the nanoparticles are
immobilized on the support and subsequently converted into heterogeneous catalysts with a
nanoscale proximity of the particles and the support.27,28,29
In general, catalyst studies can often suffer from limited comparability due to differences in
their preparation history. Preparation of catalysts typically requires individually optimized
synthetic recipes, and the resulting differences in homogeneity, dispersion etc. may

additionally complicate the comparison of the catalyst performance. In our approach, the
bifunctional catalysts were prepared based on well-defined, colloidal nanoparticle building
units, thus providing a flexible toolkit to minimize the influence of differences in preparation
histories. Uniform nanoparticles obtained via a colloidal, organometallic approach were
employed as a precursor for the methanol active component and subsequently supported on a
solid acid, namely -Al
2
3
O and zeolites of different acidity (i.e., HZSM-5, HY with varying
Al/Si ratio). This versatile approach to produce bifunctional model catalysts allowed a
comparative study of the respective dehydration catalysts in the conversion of syngas. In
particular, the effects of Cu loading, pore structure and Si:Al ratio were investigated. The
catalysts were characterized by scanning electron microscopy (SEM) with energy dispersive
X-ray spectroscopy (EDX), ammonia desorption (NH
3
-TPD), N
2
-physisorption, temperature-
programmed reduction (H -TPR), X-ray diffraction (XRD), and in situ and operando X-ray
2
absorption spectroscopy (XAS). The catalytic performance was evaluated in a continuously
operated laboratory scale plant using simulated biomass-derived synthesis gas between 250
and 300 °C and correlated with structural parameters of the bifunctional catalysts.
2
2
. Experimental
.1 Materials
All chemicals were used without further purification. Unless specified otherwise, all
nanoparticle synthesis steps were carried out under argon atmosphere and water-free
conditions. Copper(II) acetylacetonate (Cu(acac)
2
, Sigma Aldrich, ≥ 99.9% trace metal basis),
diethylzinc ((C Zn, Sigma Aldrich, ≥ 52 wt.-% Zn), toluene (Sigma Aldrich, 99.8%),
2
H )
5 2
ethanol (VWR, absolute) were used for nanoparticle synthesis. Zeolites (ZSM-5: CBV3024E,
CBV 5524G, CBV 28014 (nominal cation form ammonium); HY: CBV 720), AEROSIL®
1
50, and γ-Al
2
O
3
were obtained from Zeolyst International, Evonik, and Alfa Aeser,
6.0, Ar 6.0, N 6.0) were obtained from Air Liquide.
respectively. Gases (CO 3.7, H
2
2
2
.2 Synthesis of Cu/Zn-based nanoparticles
Cu/Zn-based nanoparticles were synthesized according to a procedure previously described
by us.29 Briefly, (C
H5) Zn (12 mL, 1.2 M in toluene) was slowly added to Cu(acac)
10.49 g, 40 mmol, suspended in 800 mL toluene) over a period of 90 min at 40 °C and stirred
2
2
2
(
at 40 °C for 24 h. Eventually, all volatile components were removed, and the product was
dried in vacuum. Cu/Zn-based nanoparticles (mean particle diameter as determined by TEM
analysis: 6.2 ± 2.0 nm) were obtained as a black powder and further used for the preparation
of the bifunctional catalysts.
2
.3 Preparation of bifunctional catalysts
The following solid acids were used as both support for the immobilization of the Cu/Zn
nanoparticles and methanol dehydration catalyst: γ-Al , ZSM-5 zeolites with increasing
2
O
3
Si/Al ratio (i.e., 15 (CBV3024E), 25 (CBV 5524G), 140 (CBV 28014), and zeolite Y (Si/Al
ratio 15 (CBV 720)). As a reference, the Cu/Zn nanoparticles were also supported on pure
silica (AEROSIL® 150) and physically mixed with the methanol dehydration catalyst (CBV
5
524G). Before use, all solid acids were calcined. During the calcination procedure, they were

initially heated to 120 °C (heating rate 2 °C/min), which was maintained for 3 h. The
temperature was then raised to 550 °C (heating rate 3 °C/min) and maintained for 4 h. After
calcination, the powders were evacuated for at least 12 h and stored under argon. The
bifunctional catalysts were prepared by suspending the dried Cu/Zn-based nanoparticles with
the calcined dehydration catalysts in weight ratios between 1:1 and 4:1 in toluene (50 mL)
while stirring for 15 min. The solvent was subsequently removed in vacuum, and the resulting
bifunctional catalysts were again calcined at 350 °C for 4 h (heating rate 110 °C/h) to remove
any organic residues from nanoparticle synthesis. Finally, the catalysts were pelletized,
crushed, and sieved to the fraction of 80 to 160 µm. All catalysts are denoted as follows:
CZ (X)-Y (Z) where CZ is a bifunctional catalyst derived from Cu/Zn-based nanoparticles, X
the Cu loading [wt.-%], Y the type of solid acid with A (γ-Al
2
3
O ), HZSM-5, or HY, and Z the
Si/Al ratio of the zeolite. “CZ (7)-HZSM5 (15)”, for example, corresponds to a bifunctional
catalyst based on Cu/Zn nanoparticles, with 7 wt.-% Cu loading on HZSM-5 with Si/Al ratio
of 15.
2
.4 Characterization
SEM-EDX investigations were carried out on the bifunctional bulk catalysts with a DSM 982
Gemini SEM, equipped with a Schottky-type thermal field emission cathode (Zeiss corp.,
Germany) and an EDX spectrometer. The total specific surface area of the catalysts was
determined with N
using a Quantachrome Nova 2000e instrument. The samples were initially dried at 230 °C for
2 h. Powder XRD patterns were recorded on a PANalytical X’Pert Pro X-ray diffractometer
2
-physisorption experiments and the Brunauer Emmet Teller (BET) method
1
employing a Bragg-Brentano geometry with Cu Kα radiation and a Ni filter. The range
between 5° and 120° (2θ) was measured within 1 h. The diffraction peaks were compared to
reference compounds reported in the Joint Committee of Powder Diffraction Standards
(JCPDS) data base. The crystallite size d was estimated based on the reflections at 43° and
3
9° (2θ) for the Cu and the CuO phase, respectively, according to the Scherrer equation
0
.9 λ
d =
(5)
β cos(θ)
where 0.9 is the shape factor, β the full width at half maximum of the diffraction peak, and θ
the Bragg angle. LaB (NIST) was used as a standard in order to account for instrumental line
6
broadening. To evaluate the acidic properties of the activated, reduced bifunctional catalyst,
TPR and TPD measurements were conducted consecutively on the same catalyst. For TPR
measurements, the calcined catalyst (at least 150 mg) was initially dried in an Ar stream (30
mL/min, heating rate 5 °C/min, 300 °C, 120 min), and the reduction behavior subsequently
analyzed from 50 °C to 250 °C in a reducing gas stream (2 vol.-% H
C/min) with a Micromeritics AutoChem HP 2950. Following these TPR experiments,
programmed desorption TPD measurements were performed on the same, reduced catalyst
sample. Therefore, the catalysts were cooled to 100 °C and NH (1.221 ± 0.024 vol.-% NH in
He, 30 mL/min) was subsequently loaded for 60 min. Desorption of NH was recorded with a
mass spectrometer (MS, MKS Cyrrus) from 120 °C to 750 °C (heating rate 4 °C/min).
2
in Ar, heating rate 2
°
3
3
3
In situ XAS spectra were measured at the CAT-ACT beamline of the KIT synchrotron
radiation source (KIT, Karlsruhe) in fluorescence mode using a PIPS diode at Cu K edge and

transmission mode at Zn K edge.30 The incident energy was selected by a double crystal
monochromator with a pair of Si (111) crystals to scan energies around Cu K and Zn K edges.
Higher harmonics were rejected by a pair of Si coated mirrors. The beam size at the sample
2
was set to 1 (horizontal) × 0.5 (vertical) mm . The catalyst (6 mg, diluted 4 times (by weight)
with γ-Al
2
3
O ) was placed in an in situ microreactor (quartz capillary, 1.5 mm diameter, 20 μm
wall thickness, catalyst bed length approx. 5 mm) between two quartz wool plugs and
3
1
mounted horizontally over a hot air blower (FMB Oxford GSB-1300). The temperature was
calibrated using an external thermocouple at the top and the bottom of the capillary
microreactor without X-ray beam. He or 5 vol.% H /He were dosed using individual mass
2
flow controllers (Bronckhorst) at a flow rate of 50 mL/min, and the gas was analyzed at the
outlet using a Pfeiffer Vacuum OmniStar GSD320 mass spectrometer. X-ray absorption near-
edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) spectra were
initially recorded at 25 °C in a He flow. The gas mixture was then switched to 5 % H
XANES spectra were recorded continuously during heating to 250 °C (heating rate 1 °C/min).
At 250 °C and 5 % H /He flow, several XANES and EXAFS spectra were recorded over a
period of 60 min. Next, synthesis gas was introduced to the in situ microreactor (15 vol.-%
, 15 vol.-% CO in He, 250 °C, 18 mL/min, 3 bar) and maintained for one hour. During the
2
/He, and
2
H
2
test under synthesis gas, XANES spectra were continuously recorded and an additional set of
EXAFS spectra at the Cu and Zn K edges was recorded after 1 h time on stream. The spectra
were normalized and background subtracted to extract EXAFS data using the ATHENA
3
2
program from the IFFEFIT software package. The average oxidation state of Cu and Zn was
determined by a Linear Combination Analysis (LCA) using reference spectra of the
1
2
3
corresponding metal foils and metal oxides. The k -, k -, and k -weighted EXAFS functions
-1
were Fourier transformed in the k range of 3.0 to 11 Å and multiplied by a Hanning window
-1
with sill size of 1 Å . The structural models were based on bulk Cu metal (Inorganic Crystal
Structure Database, ICSD collection code 43493) and wurtzite ZnO (ICSD collection code
3
2
2
6170). Structure refinement was performed using ARTEMIS software (IFFEFIT). For this
3
3
purpose, the theoretical backscattering amplitudes and phases were calculated by FEFF 6.0,
then adjusted to the experimental spectra by a least square method in R-space between 1.5 and
-1
-1
2
.9 Å (Cu K) and between 1.0 and 4.0 Å (Zn K). First, the amplitude reduction factors
were calculated using the Cu foil and ZnO reference spectra, and then, the coordination
numbers, interatomic distances, energy shifts (δE
interatomic distances (σ ) were refined. The absolute misfit between theory and experiment
was expressed by ρ.
0
), and mean square deviations of
2
2
.5 Catalytic testing
The catalytic properties were evaluated in a continuously operated laboratory scale plant
equipped with a plug flow reactor. Gases (CO, H , Ar, N ) were dosed using mass flow
2
2
controllers (Bronkhorst), and effluent gases were analyzed with a gas chromatograph (Hewlett
Packard 6890). Two different columns (RESTEK RT®-U-Bond, RESTEK RT-Msieve) as
well as two detectors (thermal conductivity detector, flame ionization detector) were
employed to analyze the gas mixtures. Before the catalytic tests, mass and heat transfer
limitations were excluded for each reaction temperature using the EUROKIN webtool.
Properties of the gas mixture (e.g. viscosity) were calculated using ASPEN plus. Before

catalytic testing, the calcined catalyst (2 g) was activated in situ using the following reduction
procedure: (1) heating (rate 17 °C/h) in 2 vol.-% H in Ar (100 mLNTP/min (normal
temperature and pressure (NTP): 0 °C, 1 atm)) to 200 °C, (2) 1 h at 200°C (2 vol.-% H
Ar), (3) heating (rate 17 °C/h) to 240 °C (2 vol.-% H in Ar), (4) 1 h at 240°C (2 vol.-% H
Ar), (5) heating (rate 10 °C/h) in pure H
2
2
in
in
2
2
2
(50 mLNTP/min) to 250 °C, and (6) 1 h at 250°C
). Subsequently, the reactor was pressurized to 50 bar, and the educt gases were introduced
total flow 50 mLNTP/min, Ar:N :H :CO = 5:2:1.5:1.5). The composition of the reactant feed
(
(
H
2
2
2
stream was adjusted by independent mass flow controllers placed on the flow lines of each
reactant gas (Ar, N , H , CO). The temperature was increased every 6 h by 10 °C (heating rate
0 °C/h) until the final temperature of 300 °C was reached. The catalytic performance was
evaluated using the following equations:
2
2
1
̇
̇
N
- N
CO,0
CO
XCO
=
∙ 100
(6)
̇
N
CO,0
̇
̇
where N
is the molar flow rate of carbon monoxide at reactor inlet, NCO the molar flow
CO,0
rate of carbon monoxide at reactor outlet and XCO the CO-conversion in %;
̇
̇
ξ (N − Ni,0)
i
i
Si =
∙ 100
(7)
̇
̇
N
- N
CO,0
CO
̇
where S
i
is the selectivity of component i, N the molar flow rate of component i at reactor
i
̇
outlet, N the molar flow rate of component i at reactor inlet and ξ
i
the number of carbon
i,0
atoms in a molecule of component i. Selectivities of higher hydrocarbons (≥C ) could not be
5
determined by Equation 7 and were summarized as SC5+ calculating the residual carbon
selectivity.
The DME yield (Y) was calculated according to Equation 8.
Yi = 푋퐶푂 ∙ 푆푖
(8)
. Results and discussion
.1 Catalyst preparation and characterization
3
3
Well-defined Cu/Zn-based nanoparticles were employed as precursors for the methanol active
component and successively supported on a solid acid to yield a series of bifunctional
catalysts exhibiting close proximity of the two catalytically active sites.
By this model kit principle, a high comparability of the respective bifunctional catalysts was
ensured (Figure 1). The reaction of (C
2
H
5
)
2
Zn with Cu(acac) yielded well defined, uniform
2
Cu/Zn-based nanoparticles with a narrow size distribution. The synthesis and the structure of
2
9
the Cu/Zn-based nanoparticles were previously described in detail. This procedure was not
only highly reproducible but could be easily scaled up in a single batch experiment to yield
the quantity of nanoparticles necessary to carry out several catalyst tests. Based on TEM
analysis, the Cu/Zn-based nanoparticles were 6.2 (±2.0) nm in size. The nanoparticle building
units were subsequently immobilized on the solid acid (i.e., -Al
2
3
O , HZSM-5, HY) to yield

the dual site catalysts. A series of bifunctional systems was prepared on HZSM-5 increasing
the Si/Al ratio from 15, 25 to 140 and the Cu loading from 7 to 15 wt.-%. For all catalysts, the
molar Cu/Zn ratio was approximately 1:3. As a reference, the Cu/Zn-based particles were
supported on pure silica (i.e., AEROSIL® 150) and physically mixed with HZSM-5 (Si/Al
ratio 25). Table 1 summarizes the composition and the specific total surface areas (A) of the
bifunctional catalysts. The specific surface areas of the catalysts decreased with increasing
metal loading and, thus, simultaneously decreasing support fraction. Due to the small pore
3
4
size of the zeolites (HZSM-5 0.51 to 0.56 nm, HY 0.74 nm), the 6 nm-sized Cu/Zn particles
are deposited on the outer surface of the support which may result in a partial blockage of the
micropores. In order to account for the different CuO/ZnO/support fractions and to compare
the effect of particle deposition in the different catalysts, the BET surface area (ABET) of the
individual components were weighted by their fraction in the final catalyst according to
Equation (8).
Acalculated = wsupport ∙ Asupport + wMeOH cat ∙ AMeOH cat
(8)
where wsupport is the zeolite content (wt.-%), Asupport the specific surface area of the support
2
(
according to manufacturer, m /g), wMeOH cat the content of methanol active component (wt.-
%), and AMeOH cat the specific surface area of the calcined Cu/Zn-powder without support
(
31 m²/g as determined by N -physisorption). In general, the experimental BET surface areas
2
were in good agreement with the calculated values, indicating that deviations in BET surface
areas resulted from different weight fractions of the two catalytically active components. For
the catalysts with a high Cu loading (13 wt.-%), however, the theoretical and experimental
surface areas differed significantly (> 10% relative), suggesting partial micropore blockage of
the supports by Cu/Zn nanoparticles.
The catalysts were investigated by powder XRD before and after catalytic testing. XRD
patterns of the calcined bifunctional catalysts are shown in Figure 2.
The XRD patterns showed the reflections of both the hexagonal ZnO phase (JCPDS 01-089-
0
510) and the monoclinic CuO phase (JCPDS 01-089-5898), whereas no indication for the
formation of any mixed oxide species or alloyed Cu Zn phases were observed. Depending on
the acidic support, the bifunctional catalysts additionally revealed the reflections of the cubic
-Al phase (JCPDS 00-010-0425), the orthorhombic HZSM-5 phase (JCPDS 00-044-
003), or the cubic HY phase (JCPDS 01-077-1551). The diffraction patterns displayed broad
reflections indicating small crystallite sizes, which were estimated using the Scherrer equation
Equation (5), Figure S1). All catalysts revealed CuO crystallite sizes in the range of 11 nm
x
y

0
2
O
3
(
and 19 nm indicating a growth of the initial, 6 nm-sized Cu/Zn-based nanoparticles during the
calcination step. After catalytic testing, the STD catalysts were isolated under inert conditions
and characterized using XRD. The corresponding patterns (Figure S2) showed again
reflections of the hexagonal ZnO phase (JCPDS 01-089-0510) and the respective dehydration
catalyst. The relative intensity of the wurtzite ZnO reflections decreased after the catalyst
3
5
tests, suggesting the formation of some amorphous ZnO species. All catalysts revealed the
reflections of the metallic Cu phase (JCPDS 01-089-2838). No reflections of a CuO phase,

alloyed Cu
x
Zn phases or a reduced Zn phase were observed. According to the Scherrer
y
equation, the Cu crystallite sizes ranged from 28 nm to 45 nm and were in a similar size range
for all bifunctional catalysts (Figure S1). Copper has a low Hüttig temperature which is
reflected by a relatively low melting point (i.e., bulk Cu 1083 °C), thus Cu catalysts have to
be operated at relatively low temperatures. Usually methanol synthesis is conducted at
pressures of 50 to 100 bar and temperatures of 240 to 260 °C, accordingly. Higher
temperatures up to 300 °C may already cause severe sintering of the Cu/Zn nanoparticles.
H
2
-TPR measurements were carried out to investigate the reduction behavior of the calcined
catalysts. The H -TPR profiles of the catalysts showed a main reduction peak with shoulders
below 250°C (Figure S3). Multiple peaks were assigned to the consecutive reduction of CuO
2
3
6
over Cu
near-surface Cu species were reduced at lower temperatures than bulk Cu.37,38 Furthermore,
the formation of spinel-type metal oxides (e.g., CuAl ) or the strong interaction of ZnO and
In this study, both the Cu loading and the Si/Al ratio of the
2
O to Cu, or to the reduction of different Cu species, where the finely dispersed or
2
O
4
3
9,40
CuO have been discussed.
supports influenced the reduction temperature. In general, a higher Cu loading led to an
increase in the reduction temperature. The main reduction peaks of the HZSM-5 (140)-based
bifunctional catalysts, e.g., were shifted from 185 °C (7 wt.-% Cu loading) to 203 °C (13 wt.-
%
Cu loading). The effect of the metal loading on the reduction temperature was also
4
1
described by others. A minimum in reduction temperature of approximately 180 °C was
observed for 6 wt.-% Cu on γ-Al , whereas higher and lower Cu loadings resulted in a shift
2
O
3
of the reduction peak to higher temperatures, where shoulders appeared at >250 °C. The
interaction of the colloidal nanoparticles with the support may have further influenced the
distribution of the particles on the support during the immobilization step. Hence, the
formation of nanoparticle multilayers on the supports may additionally result in a shift of the
reduction temperatures.
The formation of the active phase was monitored for CZ (12)-A using in situ XAS
experiments. The XANES spectra at the Cu K and Zn K edge recorded during the activation
procedure are shown in Figure 3. During in situ activation, CuO underwent a fast reduction to
Cu at approximately 135 °C without any significant Cu(I) formation. The sharp reduction
behavior indicated a narrow size distribution of the Cu/Zn particles. The XANES spectra at
the Cu K edge confirmed the complete reduction of CuO to Cu nanoparticles at temperatures
below 250 °C. This was further supported by the EXAFS spectrum of the activated catalyst at
the Cu K edge revealing a Cu-Cu distance of 2.53 ± 0.01 Å in good agreement with the bulk
2
+
Cu reference (2.543 ± 0.003 Å) (Figure S4, Table S1). Zn did not change its oxidation state,
0
excluding the formation of reduced Zn species or a Cu/Zn alloy, accordingly (Figures 3b and
S4).
The formation of alloyed Cu/Zn catalysts has been previously described by others.42 In
agreement with our XAS experiments, the instability of CuZn species ((111) surface) under
typical conditions of methanol synthesis and the formation of ZnO species was reported by
4
3
Kattel et al. The EXAFS spectra at Zn K edge (Figure S4, Table S2) further didn’t show any
reduced Zn species or Cu Zn alloy formation. Only Zn-O and Zn-Zn distances which were
x
y
consistent with the bulk ZnO reference were observed. However, the backscattering intensity
corresponding to the second shell in both calcined and reduced CZ (12)-A was significantly

lower than in the hexagonal (wurtzite) ZnO reference (i.e., 6 Zn neighbors compared to 12 Zn
4
4
in bulk ZnO). This corresponds to the theoretical structure of graphitic ZnO. Recently, the
formation of metastable “graphite-like” ZnO layers during reductive activation was reported
3
5
for industrial Cu/ZnO/Al
2
O
3
catalysts induced by strong metal–support interactions.
Previous studies have also suggested the formation of oxygen vacancy sites in ZnO,
4
5
depending on the reactive conditions over the catalyst. However, taking into account the
unchanged first shell (4 O atoms) interatomic distances and the full set of XRD reflections
corresponding to the bulk ZnO, the change in the EXAFS spectra at Zn K edge rather
suggested the formation of hexagonal ZnO with a high amount of Zn vacancies. The number
of Zn vacancies significantly increased following the reductive pretreatment (Table S2). This
was in good agreement with the decrease in the relative intensities of the ZnO reflections
observed by XRD analysis after the catalyst tests.
After 1 h under synthesis gas (15 vol.-% H , 15 vol.-% CO in He, 250 °C) at 3 bar no
2
differences in the XANES spectra at Cu K edge were observed, implying no change in the
oxidation state or structure of Cu. A slight increase in the backscattering amplitude of the first
shell was observed in the Cu K EXAFS spectrum (Figure S4), which was attributed to slightly
higher average Cu coordination number than in the activated sample (Table S1). This
difference may point to some increase in the size of Cu nanoparticles under synthesis gas.
Small changes were observed also in the XANES (slight decrease of the white line, not
shown) and EXAFS spectra (Figure S4, Table S2) at Zn K edge. Considering that no peak at
2
.4 Å (corresponding to metallic Zn, uncorrected for the phase shift) appeared, exposure of
CZ (12)-A to the synthesis gas feed led to further amorphization of ZnO. This conclusion is
supported by significantly higher disorder in the coordination environment around Zn atoms
(higher Debye-Waller factors and error bars).
The acidity of zeolites has been identified as an important parameter to determine the catalytic
1
0
performance of the STD reaction. In zeolites, the acidic properties depend on the amount of
framework aluminium. Hence, it can be tuned by varying the Si/Al ratio which affects the
catalytic activity and hydrothermal stability. The acidic properties of the bifunctional catalysts
were investigated by NH
3
-TPD using the reduced catalyst samples after the H -TPR
2
experiments (Figure S5). Both the total specific amount of desorbed ammonia and the fraction
of strong acid sites (desorption temperature above 500 °C) were determined for the
bifunctional catalysts (Table 2). For all bifunctional catalysts, the specific amount of desorbed
NH
(
(
3
decreased with respect to the acidic supports. As expected, the pure silica support
AEROSIL® 150) did not reveal any acid sites. The Cu/Zn nanoparticles supported on silica
CZ (9)-AEROSIL), however, featured some desorbed NH
3
(81.9 µmolNH3/gcat). This
demonstrates that, for all catalysts, a fraction of the desorbed NH
an interaction of NH with the Cu nanoparticles. This is not surprising since Cu is known to
coordinate NH to form a variety of amine complexes.
3
has in fact to be assigned to
3
3
In previous studies, the unexpected increase of desorbed NH for bifunctional STD catalysts
3
was ascribed to interactions between the two catalytically active components. Flores et al.
observed both an increase in the number of acid sites due to the formation of new acid sites by
4
6
Al species and a decrease of this amount due to the blockage of pores. ZnO is known to
2
-
display basic O sites which may be naturalizing some acid sites. Moreover, migration and

2
+
ion exchange of Cu species with the hydroxyl groups was shown to lead to a decrease in the
concentration of zeolite Brönsted acid sites and has been considered as a potential mechanism
47
for the deactivation of STD catalysts during reaction. Selective neutralization of Brönsted
acid sites on the outer zeolite surface (e.g., using TEOS), however, was shown to slow down
copper sintering and migration and resulted in the better catalyst stability and slightly higher
selectivity to DME. Based on the experimental TPD results of the pure dehydration catalysts
and CZ (9)-AEROSIL, the theoretically expected amount of desorbed NH
according to Equation 9.
3
was calculated
N
= xsupport ∙ N
+ xCu ∙ N
NH , Cu
3
(9)
NH ,calculated
NH , support
3
3
with NNH3, Cu = 880.6 μmol/gCu (calculated from NNH3, CZ (9)-AEROSIL ∙xCu, CZ (9)-AEROSIL-1).
Within the experimental error, the theoretical number of acid sites was in good agreement
with the experimental data. For catalysts with 13 wt.-% Cu loading, however, the expected
amount of desorbed NH
3
was significantly higher than the experimentally observed values,
most probably indicating a reduced accessibility of acid sites via the partial blockage of
micropores by Cu/Zn nanoparticles. This is also in good agreement with N
experiments.
2
-physisorption
3
.2 Catalytic performance
The catalysts were activated in situ in the reactor as described above and tested in a
continuously operated, laboratory scale plant for the single-step conversion of simulated,
biomass-derived synthesis gas (H :CO = 1:1) to dimethyl ether and hydrocarbons. In all
2
experiments, pressure (50 bar) and inert gas dilution (70% inert gas dilution) were kept
constant, and the temperature was varied between 250°C and 300°C. Nitrogen was used as an
internal standard. In particular, the influence of reaction temperature, Cu loading, type of solid
acid, pore structure and Si:Al ratio on CO conversion and DME selectivity was investigated.
3
.2.1 Influence of the reaction temperature
The bifunctional catalysts were tested at reaction temperatures ranging from 250 °C to
00 °C. When γ-Al was used as support, maximum CO conversions which were consistent
3
2
O
3
with the thermodynamic equilibrium (i.e., according to Equation 4) were obtained (Fig. S6).
In this case, DME was the main reaction product with selectivities between 65 % (250 °C)
and 55 % (300 °C). At higher reaction temperatures, the zeolite-based bifunctional catalysts
yielded higher CO conversions (i.e., as compared to thermodynamic equilibrium) since
methanol or DME were continuously withdrawn by the formation of hydrocarbons (Fig. S6).
Here, short chain paraffins up to C4, methylated alkanes (up to C
were obtained. In contrast to the conventional methanol-to-gasoline (MtG) process, no
alkenes were formed due to the high H partial pressure resulting in a complete hydrogenation
of any alkene intermediates over the methanol active component, which has also been
7
) and benzenes (up to C12)
2
4
8, 49
observed by others.

3
.2.2 Influence of the metal loading
A series of bifunctional systems was prepared with Cu loadings from 7 to 13 wt.-%. The
influence of the metal loading is shown exemplarily for the HZSM-5 dehydration catalyst
with a Si/Al ratio of 140:1 in Figure 4.
In general, a higher metal loading (i.e., a higher fraction of the methanol catalyst) led to both a
higher CO conversion and a higher DME yield. At 250 °C, DME was the main reaction
product for HZSM-5 (140)-based catalysts with selectivities between 62 % and 66 %. This
agrees well with the DME selectivity expected for the STD process (Equation 4). Moreover,
at 250°C, the CO conversion increased linearly with the Cu loading, further demonstrating the
benefit of our uniform Cu/Zn-based nanoparticles as precursors for the methanol active
component and the reduced influence of the preparation history.
The increase in reaction temperature, however, led to a sharp decline in DME yield due to the
formation of hydrocarbons via the MtG process. For lower Cu loadings and therefore higher
amount of zeolite and acidic sites in the bifunctional catalysts, hydrocarbon formation was
shifted towards lower temperatures (Figure 4). A similar trend was observed for the Cu/Zn
nanoparticles supported on HZSM-5 with Si/Al ratios of 15 and 25 (Figure S7). The
formation of higher hydrocarbons from N /methanol mixtures has been described on HZSM-5
2
catalysts upon repeated flow through the catalyst bed by Choudry et al.50 A higher
hydrocarbon selectivity was also reported by Chang et al. in the MtG process with increasing
5
1
residence time. Moreover, catalysts with lower Cu loading showed lower CO conversions,
and therefore, a higher CO partial pressure. The formation of the first C-C bond from
methanol under He and CO flow was investigated by Liu et al. While there was no olefin-
5
2
formation under helium, the authors observed the formation of olefins under CO flow.
Overall, a well-balanced mixing ratio of the two catalytically active components represents a
crucial issue for the catalytic performance of these bifunctional systems.
3
.2.3 Influence of the dehydration catalyst
To determine the influence of the dehydration component, the Cu/Zn nanoparticles were
further immobilized on γ-Al , HY and HZSM-5 (Si/Al ratio of 15) to yield a series of
bifunctional catalysts (i.e., 12 wt.-% Cu loading). In Figure 5, the DME yield is shown as a
function of the reaction temperature. The γ-Al -based catalyst revealed a DME yield of
5 % (280 °C) and thus, the highest DME yield among these dehydration catalysts (Figure 5,
2
O
3
2
O
3
2
left).
Due to the weakly acidic sites of γ-Al
2
3
O , only methane was formed as a by-product at
temperatures above 290 °C. The methane selectivity was 5.5 % and 8.8 % at 290 °C and
00 °C, respectively. For HZSM-5 and HY, the acidity depends not only on the number of
framework aluminum species but also on the structure of the crystalline microporous zeolite
framework. As shown by NH -TPD experiments (Table 2), the HZSM-5-based bifunctional
catalysts had not only a higher total number of acid sites but also more strong acid sites (NH
3
3
3
desorption > 500 °C) than the HY-based catalyst. Despite the lower total number of acid sites
and less strong acid sites, hydrocarbon formation occurred already at temperatures as low as
2
50 °C for CZ (12)-HY (15) as indicated by a lower DME yield. In addition to the total

number and strength of the acidic sites, however, the different pore sizes (HY: 0.74 nm,
HZSM-5 0.51-0.56 nm) and specific surface areas of the two acid catalysts (HY: 780 m²/g,
HZSM-5: 405 m²/g) need to be considered, affecting the accessibility of acid sites and thus,
the catalytic properties of the bifunctional systems.
In another experiment, the number of framework aluminum species was increased using
HZSM-5-based zeolites with Si/Al ratios of 140, 25, and 15. The Cu/Zn-based nanoparticles
were successively immobilized on these HZSM-5 zeolites to yield a series of bifunctional
catalysts with a Cu loading of 13 wt.-%. (Figure 5, right). HZSM-5 (140) showed both the
highest DME yield (i.e., YDME 25 % at 280 °C) and the highest CO conversion (i.e., XCO 40 %
at 280 °C) for temperatures of up to 290 °C, which suggests a balanced composition of the
individual active components in the bifunctional catalyst with respect to the STD reaction. No
significant influence on the CO conversion was reported for HZSM-5 zeolites with varying
Si/Al-ratios by Kim et al., as long as methanol synthesis was the rate determining step in the
5
3
STD reaction. In this study, the methanol selectivities were below 2.6 %, and, thus,
methanol synthesis was considered the rate determining step. Here, the highest CO conversion
was observed for the HZSM-5-based catalyst with the lowest amount of Al species (Si/Al
ratio 140). At 300 °C, CO conversion (i.e., XCO 47 %) was the highest among all bifunctional
catalysts. The selectivity to C
products), whereas CZ (13)-HZSM-5 (25) and CZ (13)-HZSM-5 (140) displayed 19.8 %
37.5 % of hydrocarbon products) and 18.2 % (39.2 % of hydrocarbon products), respectively.
1
-C
4
hydrocarbons was 28.9 % for (56.3 % of hydrocarbon
(
Interestingly, CZ (13)-HZSM-5 (25) revealed the lowest temperature for hydrocarbon
formation, although CZ (13)-HZSM-5 (15) had a higher total number and more strong acid
sites. SEM images of the pristine supports are displayed in Figure S8. As compared to the
other dehydration catalysts, it is obvious from these images that this zeolite (i.e., HZSM-5
(25)) exhibited the smallest particle size and only few agglomerates. Thus, the number of acid
sites on the external zeolite surface is increased for HZSM-5 (25), leading to an improved
accessibility of the acid sites and a lower temperature for hydrocarbon formation via the MtG
process, accordingly. Higher reaction rates in the STD reaction have been described by Cai et
al. for physically admixed STD catalysts containing zeolites with small particles sizes,
5
4
however, no significant formation of hydrocarbons has been reported in this case. While the
number and strength of acid sites is important, the accessibility of these sites is also a crucial
aspect in view of the catalytic properties of bifunctional catalysts.
3
.2.4 Correlation of activity
A linear dependency of the catalyst activity on the Cu surface area has been reported for
55,56
bifunctional STD catalysts.
In this study, well-defined, uniform nanoparticles were used
as catalyst precursors in order to avoid any effects of structure-sensitivity which has been
proposed for methanol synthesis.57,58 Catalyst series based on the same acidic component
thereby showed a linear dependency of CO conversion on the Cu loading (e.g., HZSM-5
(140)-based samples at 250 °C), as long as DME was the main reaction product, clearly
demonstrating the benefit of our uniform nanoparticle toolbox (Figure 1). Taking into account

all Cu/Zn/zeolite catalysts, a single materials descriptor, however, seems not to be sufficient
to describe the catalyst behavior. According to our results, the ratio of the two catalytically
active components as well as the accessibility (e.g., micropore area) of the acid sites are
crucial issues regarding catalyst activity. Therefore, the CO turnover rate was correlated with
these parameters at 250 and 260 °C for all bifunctional, zeolite-based catalysts irrespective of
the Si/Al ratio and the type of zeolite (Figure 6). For temperatures below 270 °C, an increase
in the Cu to acid site ratio as well as a low micropore area resulted in an increase in turnover
rate. For temperature above 270 °C, catalyst deactivation became more important.
Furthermore, an influence of the mixing ratio of the two catalytically active functions on the
product spectrum of the hydrocarbons was observed. Figure 7 shows the fraction of C
1
-C
4
hydrocarbons in those experiments, where DME was already consumed in favor of
hydrocarbon formation. At 270 °C, CZ (9)-HZSM-5 (25) displayed the highest fraction of
short chain hydrocarbons (78 %) and was the only bifunctional system showing a remaining
DME selectivity of 17 %. If the number of acid sites was increased with respect to the amount
of Cu in the bifunctional Cu/Zn/zeolite catalysts, the product spectrum was shifted towards
lower hydrocarbons and the amount of C
on the desired product, LPG type compounds (in C
type applications can be produced in a single step.
1
-C
4
components increased to 60-70 %. Depending
-C products) or C5+ compounds for fuel
1
4
For granular hybrid catalysts, the increased formation of methylated aromatics has been
previously ascribed by Fujimoto et al. to the continuous formation of methanol over the entire
4
8
catalyst bed, leading to the methylation of intermediates. On the other hand, the product
spectrum was shifted to lower alkanes with respect to the conventional MtG process for
/HZSM-5 systems, which is in good agreement with our results.4 The
9
Cu/ZnO/γ-Al
2
O
3
immediate hydrogenation of unsaturated intermediates was suggested to be responsible for
this phenomenon as a consequence of the intimate contact between the two active
components, whereby the formation of higher hydrocarbons and aromatics was reduced.
4
. Conclusions
Bifunctional model catalysts were developed for the conversion of syngas to DME and
hydrocarbons using the precursor concept. Well-defined and uniform Cu/Zn nanoparticles
were prepared and successively immobilized on different dehydration catalysts to study the
influence of the acidic properties on the catalytic performance. In particular, the Cu loading,
the type of acidic sites (γ-Al
2
3
O vs. zeolites), the type of acid catalyst (HY vs. HZSM-5) and
different Si/Al ratios were investigated.
Additionally, the formation of the active phase was monitored by in situ and operando XAS
measurements. The Cu particles were completely reduced at 250 °C, also under reaction
2
+
conditions at 3 bar, whereas the Zn prevailed as Zn . In agreement with ex situ XRD
investigations of the catalysts after both calcination and catalytic testing, an amorphization of
the ZnO phase was observed. The CO conversion was correlated with the ratio of the two
catalytically active components and the micropore area of the bifunctional systems. With
decreasing micropore area and increasing ratio of Cu to acidic sites, the CO conversion

increased. The overall DME synthesis rate was controlled by both the relative ratio of
methanol synthesis to methanol dehydration components and the reaction conditions. The
selectivity to LPG products was also controlled by the ratio of the two catalytically active
components. Consequently, well-defined nanoparticle building units open an encouraging
avenue for the preparation of bifunctional catalysts for conversion of synthesis gas to either
DME or hydrocarbons, enabling further systematic studies of synthesis–structure–function
relationships. Furthermore, by choosing different building blocks and systematic balancing of
the two catalytic entities, future control of the product selectivities can be explored (e.g.,
towards olefins, LPG or gasoline range products).
Acknowledgements
M.G. kindly acknowledges financial support of the Helmholtz Research School Energy-
Related Catalysis. Further acknowledgments are given to the synchrotron radiation source at
Karlsruhe Institute of Technology for providing beamtime at CAT-ACT beamline and Dr.
Anna Zimina and Dr. Tim Prüßmann for their support during the measurements. Doreen
Neumann-Walter and Dr. Thomas Otto are acknowledged for their support with the BET
measurements.

References
1
2
N. Dahmen, E. Dinjus., Chemie Ingenieur Technik, 2010, 82, 1147-1152.
E. Dinjus, U. Arnold, N. Dahmen, R. Höfer, W. Wach, RCS Green Chemistry No. 4,
The Royal Society of Chemistry, London 2009, 125-163.
3
P. Lv, Z. Yuan, L. Ma, C. Wu, Y. Chen, J. Zhu, Renewbale Energy, 2007, 32, 2173-
2185.
4
5
D. Mao, J. Xia, Q. Chen, G. Lu, Catal. Commun., 2009, 10, 620-624.
F. S. Ramos, A. M. Duarte de Farias, L. E. P. Borges, J. L. Monteiro, M. A. Fraga, E.
F. Sousa-Aguiar, L. G. Appel, Catalysis Today, 2005, 101, 39–44.
A. Mittasch, M. Pier, US 1569775 A, 1926
J.-D. Grunwaldt, A. M. Molenbroek, N.-Y. Topsoe, H. Topsoe, B. S. Clausen, Journal
of catalysis, 2000, 194, 452-460.
6
7
8
9
1
M. Behrens, R. Schlögl, J. Anorg. Allg. Chem., 2013, 639, 2683-2695.
S. Hassanpour, M. Taghizadeh, Ind. Eng. Chem. Res., 2010, 49, 4063-4069.
0M. Stiefel, R. Ahmad, U. Arnold, M. Döring, Fuel Process Technology, 2011, 92
466-1474.
1J. Abu-Dahrieh, D. Rooney, A. Gouget, Y. Saih, Chem. Eng. Journal, 2012, 203, 201-
11.
2J.-H. Kim, M. J. Park, S. J. Kim, O.-S. Joo, K.-D. Jung, Appl. Catal. A: General, 2004,
64, 37-41.
3T. Takeguchi, K.-I. Yanagisawa, T. Inui, M. Inoue, Applied Catalysis A: General,
000, 192, 201-209.
4J. Sun, G. Yang, Y. Yoneyama, N. Tsubaki, ACS Catalysis, 2014,
,
1
1
1
1
2
2
2
1
1
1
1
4, 3346-3356.
5Q. Ge, Y. Huang, F. Qiu, S. Li, Applied Catalysis A: General, 1998, 167, 23-30.
6J.-L. Li, X.-G. Zhang, T. Inui, Appl. Catal. A, 1996, 147, 23–33.
7C. Zeng, J. Sun, G. Yang, I. Ooki, K. Hayashi, Y. Yoneyama, A. Taguchi, T. Abe, N.
Tsubaki, Fuel, 2013, 112, 140-144.
18R. Ahmad, D. Schrempp, S. Behrens, J. Sauer, M. Döring, U. Arnold, Fuel Cell Proc.
Technol. 2014, 121, 38-46.
19G. Ertl., D. Prigge, J. Catal., 1983, 79, 359-377.
20H.-J. Freund, M. Bäumer, J. Libuda, T. Risse, G. Rupprechter, S. Shaikhutdinov,
Journal of Catalysis, 2003, 216, 223–235.
21M. Boudart, Topics in Catalysis, 2000, 13, 147–149.
22S. Vukojevic, O. Trapp, J.-D. Grunwaldt, C. Kiener, F. Schüth, Angew. Chem. Int.
Ed., 2005, 44, 7978 –7981.
2
2
3M. Simon, A. Mortreux, F. Petit, J. Chem. Soc., Chem. Commun., 1988, 1445-1446.
4M. K. Schröter, L. Khodeir, M. W. E. van den Berg, T. Hikov, M. Cokoja, S. Miao,
W. Grünert, M. Muhler, R. A. Fischer, Chem. Commun., 2006, 2498-2500.
5S. Schimpf, A. Rittermaier, X. Zhang, Z.-A. Li, M. Spasova, M. W. E. vand den Berg,
2
2
M. Farle, Y. Wang, R. A. Fischer, M. Muhler, ChemCatChem 2010, 2, 214 – 222.
6N. J. Brown, A. García-Trenco, J. Weiner, E. R. White, M. Allinson, Y. Chen, P. P.
Wells, E. K. Gibson, K. Hellgardt, M. S. P. Shaffer, C. K. Williams, ACS Catal., 2015,
5
, 2895–2902.
27H. Bönnemann, R. M. Richards, Eur. J. Inorg. Chem. 2001, 2455-2480.
2
8H. Bönnemann, G. Braun, W. Brijoux, R. Brinkmann, A. Schulze Tilling, K.
Seevogel, K. Siepen, Journal of Organometallic Chemistry, 1996, 520, 143-162
9M. Gentzen, W. Habicht, D. E. Doronkin, J.-D. Grunwaldt, J. Sauer, S. Behrens, Catal.
2
3
Sci. and Technol., 2016, 6, 1054-1063.
0A. Zimina, K. Dardenne, M.A. Denecke, D. E. Doronkin, E. Huttel, H. Lichtenberg, S.
Mangold, T. Pruessmann, J. Rothe, Th. Spangenberg, R. Steininger, T.Vitova, H.
Geckeis, J.-D. Grunwaldt, Rev. Sci. Instrum., 2017, 88, 113113.

31J.-D. Grunwaldt, M. Caravati, S. Hannemann, A. Baiker, Phys Chem Chem Phys.,
2
004, , 3037–3047.
6
32B. Ravel, M. Newville, J. Synchrotron Radiat., 2005, 12, 537–541.
33J.J. Rehr, R.C. Albers, Rev. Mod. Phys., 2000, 72, 621–654.
34C. Baerlocher, L. B. McCusker, D. H. Olson, Atlas of zeolite framework types, 6th
revised version, Elsevier B. V., 2007, Amsterdam.
3
3
3
3
3
5T. Lunkebein, J. Schumann, M. Behrens, R. Schlögl, M. G. Willinger, Angew. Chem.
Int. Ed., 2015, 54, 4544-4548.
6M. Turco, G. Bagnasco, U. Costantino, F. Marmottini, T. Montanari, G. Ramis, G.
Busca, Journal of Catalysis, 2004, 228, 43–55.
7Y. Guo, W. Meyer-Zaika, M. Muhler, S. Vukojevic, M. Epple, Eur. J. Inorg. Chem.
2
006, 4774–4781.
8X.-J. Tang, J.-H. Fei, Z.-Y. Hou, X.-M. Zheng, H. Lou, Energy & Fuels, 2008, 22
877-2884.
9J. Agrell, K. Hasselbo, K. Jansson, S. G. Järas, M. Boutonnet., Appl. Catal. A 2001,
21, 239–250
,
2
2
4
4
0H. Wilmer, T. Genger, O. Hinrichsen, Journal of Catalysis, 2003, 215, 188–198.
1B. Lindström, L. J. Pettersson, P. G. Menon, Applied Catalysis A: General, 2002, 234
,
1
11–125.
2J. Sun, G. Yang, Q. Ma, I. Ooki, A. Taguchi, T. Abe, Q. Xie, Y. Yoneyama, N.
Tsubaki, J. Mater. Chem. A, 2014, , 8637-8643.
4
2
4
4
3S. Kattel, P. J. Ramírez, J. G. Chen, J. A. Rodriguez, P. Liu, Science, 355, 1296-1299.
4Z. Zhu, A. Chutia, R. Sahnoun, M. Koyama, H. Tsuboi, N. Hatakeyama, A. Endou, H.
Takaba, M. Kubo, C. A. Del Carpio, A. Miyamoto, Japanese Journal of Applied
Physics, 2008, 47, 2999-3006.
4
4
4
4
4
5
5P. Vesborg, I. Chorkendorff, I. Knudsen, O. Balmes, J. Nerlov, A. M. Molenbroek, B.
Clausen, S. Helvegb, J. Catal., 2009, 262, 65-72.
6J. H. Flores, M. I. P. da Silva, Colloids and Surfaces A: Physicochem. Eng. Aspects,
2008, 322, 113–123.
7V.V. Ordomsky, M. Cai, V. Sushkevich, S. Moldovan, O. Ersen, C. Lancelot, V.
Valtchev, A.Y. Khodakov, Applied Catalysis A: General, 2014, 486, 266-275.
8K. Fujimoto, K. Asami, H. Saima, T. Shikada, H. Tominaga, Ind. Eng. Chem. Prod.
Res. Dev., 1986,
9E. Lorenz, P. Wehling, M. Schlereth, B. Kraushaar-Czarnetzki, Catalysis Today, 2016,
75, 183-190.
0R. Choudry, T. Mandal, Fuel Science and Technology International, 1990,
036.
1C. D. Chang, Catalysis Reviews, 1983, 25, 1-118.
2, 262-267.
2
8
, 1021-
1
5
5
2Y. Liu, S. Müller, D. Berger, J. Jelic, K. Reuter, M. Tonigold, M. Sanchez-Sanchez, J.
A. Lercher, Angew. Chem. Int. Ed., 2016, 55, 5723–5726.
5
5
5
3J.-H. Kim, M. J. Park, S. J. Kim, O.-S. Joo, K.-D. Jung, Appl. Catal. A: General, 2004,
2
64, 37-41.
4M. Cai, A. Palčić, V. Subramanian, S. Moldovan, O. Ersen, V. Valtchev, V.V.
Ordomsky, A.Y. Khodakov, J. Catal., 2016, 338, 227-238.
5M. Kurtz, H. Wilmer, T. Genger, O. Hinrichsen, M. Muhler, Catal. Lett., 2003, 86
7–80
,
7
5
5
5
6J. Robbins, E. Iglesia, C.P. Kelkar, B. DeRites, Catal. Lett., 1991, 10, 1-10.
7J. Yoshihara, C. T. Campbell, Journal of Catalysis, 1996, 161, 776-782.
8H. Wilmer, M. Kurtz, K. V. Klementiev, O. P. Tkachenko, W. Grünert, O. Hinrichsen,
A. Birkner, S. Rabe, K. Merz, M. Driess, C. Wöll, M. Muhler, Phys. Chem. Chem.
Phys., 2003, 5, 4736-4742.

Figure 1. Preparation of bifunctional catalysts for syngas conversion using nanoparticle
building units.

HZSM-5
ZnO
CuO
Si:Al = 140
CZ(7)-HZSM-5 (140)
CZ(9)-HZSM-5 (140)
CZ(11)-HZSM-5 (140)
CZ(13)-HZSM-5 (140)
AEROSIL
CZ(9)-AEROSIL
HZSM-5
Si:Al = 25
CZ(7)-HZSM-5 (25)
CZ(9)-HZSM-5 (25)
CZ(13)-HZSM-5 (25)
HZSM-5
Si:Al = 15
CZ(7)-HZSM-5 (15)
CZ(12)-HZSM-5 (15)
CZ(13)-HZSM-5 (15)
CZ(12)-HY (15)
HY
Si:Al = 15
CZ(15)-HY (15)
CZ(12)-A

Al O3
2
20
40
60
80
Angle 2
Figure 2. XRD patterns of the bifunctional catalysts after calcination.

(a)
Cu K edge
(b)
Zn K edge
1
7
9
1
1
1
2
9 °C
7 °C
9 °C
14 °C
36 °C
69 °C
50 °C + 55 min
19 °C
126 °C
152 °C
181 °C
209 °C
235 °C
250 °C + 40 min
8
980
9000
9020
9040
9060
9660
9680
9700
9720
9740
Energy (eV)
Energy (eV)
Figure 3. In situ XANES spectra at the Cu K (a) and the Zn K edge (b) of CZ (12)-A between
5 and 250 °C in 5 % H in He.
2
2

30
25
20
15
10
5
0
Cu/Zn (7)-HZSM-5 (140)
Cu/Zn (11)-HZSM-5 (140)
Cu/Zn (9)-HZSM-5 (140)
Cu/Zn (13)-HZSM-5 (140)
250
260
270
280
290
300
Temperature (°C)
Figure 4. Effect of Cu loading on the DME yield for HZSM-5 (140)-based catalysts.

3
2
2
1
1
0
5
0
5
0
5
0
30
25
20
15
CZ (12)-HZSM-5 (15)
CZ (12)-A
CZ (12)-HY (15)
CZ (13)-HZSM-5 (140)
CZ (13)-HZSM-5 (15)
CZ (13)-HZSM-5 (25)
1
0
5
0
2
50
260
270
280
290
300
250
260
270
280
290
300
Temperature (°C)
Temperature (°C)
Figure 5. DME yields of bifunctional catalysts with (left) 12 wt.-% Cu loading and (right)
3 wt.-% Cu-loading.
1

CO turnover rate (mmolCO/g
)
CO turnover rate (mmolCO/gcat)
cat
-4
-4
-4
-4
-4
-4
-4
-4
-4
-3
-4
-4
-4
-4
-4
-4
-4
-4
9
.0x10
1.0x10
9.4x10
8.3x10
7.2x10
6.1x10
4.9x10
3.8x10
2.7x10
1.6x10
2
50 °C
260 °C
0
0
0
.15
.10
.05
8.0x10
0.15
0.10
0.05
7
6
5
4
.0x10
.1x10
.1x10
.1x10
3.2x10
2
1
.2x10
.2x10
60
80 100 120 140 160 180 200 220 240
60 80 100 120 140 160 180 200 220 240
2
2
Micropore area (m /gcat)
Micropore area (m /gcat)
Figure 6. Specific CO turnover rates and ratio of the two catalytically active components in
the bifunctional catalyst as a function of the micropore area (left) at 250 °C and (right) at 260
°
C.

1
0
0
0
0
0
0
0
0
0
0
.0
.9
.8
.7
.6
.5
.4
.3
.2
.1
.0
2
2
50 °C
90 °C
260 °C
300 °C
270 °C
280 °C
2
0
20
40
60
80
100
120
140
160
Mixing Ratio (µmolNH / g )
Cu
3
Figure 7. Influence of the ratio of the two catalytically active components in the bifunctional
catalysts on the formation of light hydrocarbons (i.e., fraction of C -C hydrocarbons in
organic product) in the direct conversion of synthesis gas for a temperature range of 250 °C to
00 °C.
1
4
3

Table 1. Compositions and specific surface areas of the bifunctional model catalysts
A
d
calculate
w
Cu
w
Zn
A
BET
Catalyst
(%) (%) (m²/g)
(
m²/g)
CZ (7)-HZSM-5 (140)
CZ (9)-HZSM-5 (140)
7.1
9.6
17.6 269.6
24.5 243.8
263.1
227.0
CZ (11)-HZSM-5 (140)
CZ (13)-HZSM-5 (140)
CZ (9)-AEROSIL
CZ (7)-HZSM-5 (25)
CZ (9)-HZSM-5 (25)
CZ (13)-HZSM-5 (25)
CZ (7)-HZSM-5 (15)
CZ (12)-HZSM-5 (15)
CZ (13)-HZSM-5 (15)
CZ (12)-H-Y (15)
11.2 35.3 190.1
13.2 34.7 150.4
172.8
170.5
-
9.3
7.5
9.7
21.9 84.1
21.0 253.7
27.5 207.7
249.0
225.7
176.5
264.5
173.7
176.5
311.0
219.9
12.9 33.5 154.0
7.7 19.9 243.8
12.2 34.2 161.5
13.5 38.2 144.0
11.9 33.1 305.2
15.6 39.8 214.6
CZ (16)-H-Y (15)

3
Table 2. Results of the NH -TPD experiments.
Amount of strong Calculated NH -
3
NH desorbed
3
Catalyst
acid
sites amount
(µmol/gcat)
(> 500 °C) (%)
(µmol/gcat
)
HZSM-5 (140)
HZSM-5 (25)
HZSM-5 (15)
HY
99.1
3.0
9.1
10.2
8.1
-
-
-
-
-
-
-
521.8
890.9
411.4
182.1
0
γ-Al
2
O
3
AEROSIL
-
CZ (7)-HZSM-5 (140)
CZ (9)-HZSM-5 (140)
CZ (11)-HZSM-5 (140)
CZ (13)-HZSM-5 (140)
CZ (9)-AEROSIL
CZ (7)-HZSM-5 (25)
CZ (9)-HZSM-5 (25)
CZ (13)-HZSM-5 (25)
CZ (7)-HZSM-5 (15)
CZ (12)-HZSM-5 (15)
CZ (13)-HZSM-5 (15)
CZ (12)-H-Y (15)
144.8
146.6
139.9
94.4
11.7
10.1
9.3
10.3
3.8
21.5
17.4
14.8
14.7
11.3
13.7
4.4
5.7
-
125.5
137.8
137.3
154.3
-
376.4
345.3
308.3
627.7
450.9
402.5
259.4
242.3
181.6
81.9
432.6
363.8
274.0
625.9
425.5
345.7
339.6
245.6
163.0
CZ (16)-H-Y (15)
CZ (12)-A
5
6
9
0