ZnO during ... - ACS Publications

Apr 27, 2016 - Ernst Ruska-Centrum und Peter Grünberg Institut, Forschungszentrum Jülich, 52425 Jülich, Germany. ‡. Institute of Physical Chemist...
0 downloads 0 Views 2MB Size
Subscriber access provided by University of Sussex Library

Article

Formation of ZnO Patches on ZnPd/ZnO During Methanol Steam Reforming – A Strong-Metal-Support-Interaction Effect? Marc Heggen, Simon Penner, Matthias Friedrich, Rafal E. Dunin-Borkowski, and Marc Armbrüster J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.6b02562 • Publication Date (Web): 27 Apr 2016 Downloaded from http://pubs.acs.org on May 6, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 17

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Formation Of ZnO Patches On ZnPd/ZnO During Methanol Steam Reforming – A Strong-Metal-Support-Interaction Effect?

Marc Heggen1*, Simon Penner2, Matthias Friedrich3, Rafal E. Dunin-Borkowski1, Marc Armbrüster4

1

Ernst Ruska-Centrum und Peter Grünberg Institut, Forschungszentrum Jülich, 52425 Jülich, Germany

2

Institute of Physical Chemistry, University of Innsbruck, Innrain 52A, 6020 Innsbruck, Austria

3

Fritz-Haber-Institute of Max-Planck-Society, Department of Inorganic Chemistry, Faradayweg 4-6, 14195

Berlin, Germany 4

Faculty of Natural Sciences, Institute of Chemistry, Materials for Innovative Energy Concepts, Technische

Universität Chemnitz, 09107 Chemnitz, Germany

ABSTRACT: The high CO2 selectivity of ZnPd/ZnO, an outstanding catalyst in methanol steam reforming, was recently attributed to the interaction between small ZnO patches and ZnPd particles. Yet, the detailed microstructure of this catalytic system and the formation mechanism of the ZnO patches inducing the high catalytic selectivity are unknown. In this work, we uncover the formation mechanism of ZnO patches using aberration-corrected electron microscopy, electron energy loss spectroscopy, X-ray spectroscopy and in situ heating experiments. We show that Zn-rich regions in chemically inhomogeneous ZnPd particles, penetrating the particle surface, are capped by ZnO. It is demonstrated that the ZnO patches form by direct oxidation of the particles rather than by transport from the ZnO substrate, thus ruling out a classical strong metal-support interaction (SMSI).

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1. INTRODUCTION Hydrogen generation by methanol steam reforming (MSR) is a promising route to provide clean hydrogen for fuel cells in electric vehicles and other future energy applications. Due to their outstanding performance and high CO2 selectivity, ZnO-supported ZnPd particles have shown to be highly promising MSR catalysts.1,2,3 Recently, the origin of the high CO2 selectivity of ZnPd/ZnO was revealed by linking their catalytic properties with aberrationcorrected high-resolution transmission electron microscopy (HRTEM) imaging of the catalyst at different stages of the MSR reaction4 and comparison to catalytic data obtained on unsupported ZnPd model catalysts.5,6 While as-prepared ZnPd/ZnO shows only low selectivity towards CO2, the material exhibits very high CO2-selectivity (>95%) after a “selectivation period” in the methanol steam reforming reaction mixture. 4 It was shown, that the high CO2-selectivity is intimately connected with the formation of small ZnO patches on ZnPd nanoparticles, subsequently causing a large amount of beneficial interface between ZnPd and ZnO. The formation of the large interface in turn enables efficient water activation, being a prerequisite for a high CO2-selectivity in MSR. However, whereas the presence of the ZnO patches was unambiguously revealed by aberration-corrected electron microscopy, the exact formation mechanism of the ZnO patches, and hence of the supposedly active and selective interface itself, still remained unclear. To answer this question, we present a detailed micro-structural study of ZnO-supported ZnPd nanoparticles, which focuses on the structure and composition of the ZnPd nanoparticles leading to the formation of ZnO patches. Of crucial importance with respect to the presented results is the question, whether the system Pd/ZnO is an example of a system being susceptible to true strong metal-support interaction (SMSI). The term describes the drastic change in the chemisorption properties of supported noble metals and is usually only observed using easily reduced oxide substrates (the archetypical representatives being TiO2 or CeO2), manifesting itself in structural and/or 2 ACS Paragon Plus Environment

Page 2 of 17

Page 3 of 17

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

electronic changes of the catalyst structure upon reduction in hydrogen at elevated temperatures. Structural manifestations are typically the coverage of the metal particles by sub-stoichiometric oxidic support species. However, to be classified as a true strong metalsupport interaction, several conditions have to be fulfilled: at first, the system must show conventional behavior upon reduction at low temperatures. Secondly, changes in the chemisorptive and/or catalytic behavior are observed at high reduction temperatures. Last, but not least, the changes introduced upon reduction must be fully reversible upon re-oxidation and mild re-reduction temperatures.7 Connecting to the presented results, the central questions hence refers to the exact formation mechanism of the ZnO patches, i.e. if they arise from a reduced ZnO precursor (i.e. are formed by diffusion from the ZnO substrate) or if they are formed by Zn segregation and oxidation of the chemically inhomogeneous ZnPd particles directly.

2. EXPERIMENTAL PROCEDURE A Pd/ZnO (9.2 wt.%) powder catalyst was prepared following a standard incipient-wetness technique. Reductive pretreatment to induce the formation of ZnPd was performed at 773 K in H2 (5 vol.%). Catalytic MSR experiments were conducted in a CH3OH/H2O (1:1) mixture up to reaction temperatures of 300°C. For further details concerning synthesis and catalytic testing, we refer to ref. (4). The present electron microscopy study is based upon samples after catalytic MSR testing in the reactor up to 300°C, with only short contact to air for transfer to the electron microscope. HRTEM investigations were performed using an FEI-Titan 80-300 electron microscope equipped with a spherical-aberration (Cs) corrector for the objective lens. The microscope was operated at a voltage of 300 kV using the negative-Cs imaging technique (with Cs set at around ~13 µm and defocus around +6 nm).8,9 Scanning transmission electron microscopy 3 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(STEM) studies were performed in two FEI Titan scanning electron microscopes operated at 200 and 80 kV. Both microscopes are equipped with a Cs-probe corrector and a high-angle annular dark field (HAADF) detector. ‘Z-Contrast’ conditions were achieved using a probe semi-angle of 25 mrad and an inner collection angle of the detector of 70 mrad. Electron energy loss (EEL) spectra were recorded with a Gatan image filter Tridiem ERS and Quantum ERS system analyzing the O-K, Pd-M4,5 and Zn-L2,3 edges for line profiles and spectral images. Compositional maps were furthermore obtained with energy-dispersive X-ray spectroscopy (EDX) using four large-solid-angle symmetrical Si drift detectors on an FEI ChemiSTEM. For EDX elemental mapping, Pd-L, Zn-K and O-K peaks were used. The error of the EDX composition measurement for individual particles after a typical investigation of about 10 minutes is about +/- 5 at.%. In situ heating experiments were conducted using a single-tilt MEMS-chip heating holder (DENSsolutions, Delft, The Netherlands) in an FEITitan 80-300 electron microscope in HAADF-STEM mode at 300 kV.

3. RESULTS AND DISCUSSION Figure 1 shows HRTEM micrographs of ZnPd nanoparticles of various sizes (about 10, 15, and 25 nm in diameter) after a MSR reaction cycle up to 300°C. In all cases, the surface is decorated by ZnO patches, which can be identified by their (002) and (010) lattice spacings of 0.26 and 0.28 nm, respectively. In contrast to the observation of ZnPd particles with ZnO patches, particles before contact with the MSR reaction mixture do not show the presence of ZnO (cf. Figure 2 in ref. (4)). The ZnO patches have a typical maximum thickness of about 2 nm extending from the particle surface. Although electron microscopy images represent projected images of the patches on the surface which may overlap, individual patches are clearly separated with associated free intermetallic ZnPd surface in between.

4 ACS Paragon Plus Environment

Page 4 of 17

Page 5 of 17

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 1: HRTEM micrographs of ZnPd nanoparticles of various sizes. On the surface of the particles several Wurtzite-type ZnO patches are identified by their (002) and (010) lattice spacings. In Figure 1C strain within the ZnO particles is revealed by bending of the (002) planes by 8 and 6°, respectively.

Sometimes bending of the ZnO lattice planes is observed, indicating a strained state of the ZnO particles. Figure 1C shows bending of (002) planes within a patch of 8 and 6°, respectively, which gives rise to obvious strong lattice strain within the patches. Several studies reported a transformation from Wurtzite to a graphite-like ZnO in thin ZnO

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

layers.10,11,12 The observed transformation is accompanied by a significant lateral expansion and straining. Such transformation would change the electronic structure and the chemical properties of the ZnO layer and subsequently would lead to altered catalytic properties. Lunkenbein and coworkers found clear evidence for the overgrowth of Cu nanoparticles with a layered graphite-like ZnO during reductive activation of Cu/ZnO/Al2O3.13 In our present investigation, however, it is unclear, if the observed straining effect in the ZnO patches is related to a partial structural transformation to a graphite-like ZnO. Our observation of Wurtzite-type (002) lattice spacings in Figure 1 contradicts a transformation to a graphite-like ZnO structure – at least in the ZnO patches on the surface of the ZnPd nanoparticles. Figure 2A highlights an HAADF-STEM micrograph of a ZnPd particle. The image offers Zcontrast, i.e. the image intensity is roughly proportional to Z1.6 - Z1.9 where Z is the atomic number of the present element.14,15 Very unexpected, the ZnPd particle clearly shows a nonuniform contrast with dark regions across the particle – similar to GaPd2/MgO/MgGa2O4 after being used as semi-hydrogenation catalyst.16 While earlier HRTEM investigations on ZnPd/ZnO showed that each ZnPd particle is built from several crystals, no indications for a chemical inhomogeneity within the particles were obtained. To reveal the elemental nature of the variations, an EEL spectrum profile was recorded across the particle along the white arrow in Figure 2a. The individual line profiles represent the elemental distribution of Pd, Zn, and O across the particle. The intensity represents the amount of the respective element along the direction of view. The sketch in Figure 3 illustrates the basic geometry and elemental distribution of an exemplified nanoparticle. with respect to the direction of view. A comparison of the HAADF micrograph and the EELS data shows that the dark areas (local dip in the HAADF curve in Figure 2B) correspond to a local depletion of Pd, which can be noted as local minima in the Pd distribution at about 20 and 33 nm (small arrows). The amount of Zn shows an increase from the outside of the particles, which is consistent with the 6 ACS Paragon Plus Environment

Page 6 of 17

Page 7 of 17

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

geometrical increase of thickness of the ZnPd particle towards its center. At about 20 and 33 nm, however, two distinct additional bumps are visible. From these observations, it is obvious, that the local Pd-depletions in the dark areas are accompanied by an enrichment of Zn. The O EELS signal is almost zero along most of the scan; only at about 20 and 33 nm two distinct peaks are noted. The presence of oxygen points towards the presence of ZnO and is consistent with the HRTEM observations. Quantification of the composition was approached by i) quantitative EELS analysis as well as ii) EDX spectroscopy. Following the EELS results, the dark areas at 20 and 33 nm in Fig. 2 correspond to Zn45Pd15O40 and Zn45Pd25O30, respectively (see also Supporting Information, Figure

S1).

In

contrast,

the

bright

areas in-between

have

a

composition

of

Zn57.5±2.5Pd40O2.5±2.5.

Figure 2: A) HAADF-STEM micrograph of a ZnPd particle showing non-uniform contrast. An EELS line scan across the particle was performed (white arrow). B) EEL spectrum profiles for Pd, Zn, and O and HAADF intensity profile. Dark spots (black arrows) in the

7 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

HAADF image represent regions which are Pd-depleted and enriched in Zn. O is only present at the dark spots.

The EELS analysis is supplemented by an EDX compositional analysis of two representative ZnPd particles. Figure 3 displays the respective HAADF-STEM image and the EDX maps showing the distribution of Pd, Zn, and O. Table 1 specifies the results of local EDX composition analyses taken from the highlighted areas in Figure 3.

Figure 3: HAADF-STEM image of two ZnPd nanoparticles and respective EDX maps representing Pd (red), Zn (blue), and O (green). The EDX map representing all elements is a quantitative map (O is amplified with respect to Pd and Zn for a better visualization). Table 1 shows the results of local EDX composition analyses taken from the boxed areas in the HAADF-STEM image. The sketch illustrates the ZnPd and ZnO distribution in cross section of a particle with bright and dark areas.

Again, it is clearly visible that the bright regions in the HAADF image in Figure 3 (regions 3 and 4) are Pd-rich and generally have an O content close to zero. In contrast, the dark areas are Pd-depleted and show the presence of about 15-20% Pd. Although slight compositional 8 ACS Paragon Plus Environment

Page 8 of 17

Page 9 of 17

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

variations are found between the EELS and EDX investigations (Figures 2 and 3) due to variation between the investigated particles and/or potential systematic errors of the EELS and EDX measurement, several conclusions can be drawn: The bright regions in the HAADF images have an average composition of Zn60 Pd40 which is slightly richer in Zn than the compositional limits of the tetragonal bulk ZnPd phase between about 39 and 50% Zn.6 HRTEM and previously published X-ray diffraction investigations, however, (Figure 1, Figure S3 and investigations in Ref. (4)) demonstrate solely the presence of the ZnPd and ZnO phase. Therefore, the bright regions are clearly assigned to the ZnPd phase. Most remarkable, the EDX and EELS investigations show the absence of oxygen in the bright regions (see also Figure S2). Hence, we can conclude that ZnPd phase regions exist which are not covered by ZnO.

Position 1 2 3 4 5 6

Composition Zn62Pd18O20 Zn60Pd27O13 Zn57Pd42O1 Zn60Pd40O0 Zn51Pd38O11 Zn62Pd23O15

Table 1: Local EDX composition analyses in different areas indicated in Figure 3.

In contrast, the dark areas are assigned to a mixture of the ZnPd and ZnO phase, and show, depending on the ratio of both phases, a stronger compositional variation than the bright areas. Most remarkable is the high amount of oxygen found at some of the dark spots. These results indicate the presence of large amounts of ZnO even within the nanoparticles (see sketch in Figure 3). Taking an average composition of Zn60Pd20O20 in the dark regions into account, 40% of matter would be assigned to the ZnO phase (20% Zn, 20% O). After 9 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

subtraction of the amount of ZnO, material with composition Zn67Pd33 would remain. Again, as no other phases than ZnPd and ZnO were detected by HRTEM, this remaining phase can only be assigned to ZnPd. The sketch in Figure 3 illustrates the distribution of ZnPd and ZnO in cross section of a particle with above mentioned compositions of dark and bright areas. Depending on the error of measurement of the EELS and EDX measurement and the detection limit of minor amounts of additional phases in the HRTEM analysis, however, the presence of small amounts of additional Zn-rich phases like Zn2Pd cannot be ruled out. Furthermore, the range of existence for ZnPd is most likely different in the nanoparticulate state than for the bulk, allowing more zinc-rich compositions. To study the microstructural evolution of the particles during a temperature treatment typical for MSR, in situ heating experiments at 320°C were conducted in the electron microscope in accordance to a temperature range of 250 – 400°C reported previously.4 In accordance with earlier work, as-prepared ZnPd particles reveal strong internal HAADF contrast variation and the absence of ZnO phase before being exposed to MSR conditions. During heating, a coarsening of the Pd-rich and Pd-depleted regions is observed in situ (see for instance the change of shape of the Pd-depleted dark area indicated by arrows in Figure 4), demonstrating the presence of dynamic structural changes and of compositional variation due to atomic diffusion.

10 ACS Paragon Plus Environment

Page 10 of 17

Page 11 of 17

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 4: In situ HAADF-STEM images of ZnPd/ZnO before (RT) and at different time intervals during heating at 320°C.

Such dynamic coarsening of Pd-rich and Pd-depleted regions during thermal treatment may assist the structural changes taking place during MSR which lead to the extensive oxidation of Pd-depleted regions and to the formation of large ZnO regions, not only confined to the surface but extending into the nanoparticles. This conclusion is supplemented by the observation of particles like the one in Figure 5, showing a “channel-like” distribution of Pddepleted regions. The dynamic nature of structural changes at elevated temperature, as well as the channel-like structure may allow the oxidation of the Pd-depleted zones and the formation of ZnO even in subsurface regions of the nanoparticles. During formation of the ZnO phase, the surplus Pd in the Zn-rich regions would have to be adsorbed by the surrounding ZnPd phase during oxidation. Due to the wide compositional existence range of the ZnPd phase and the dynamic nature of the process at elevated temperature, a substantial local enrichment of

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Pd is not expected, which is in accordance with the findings from our microstructural investigation.

Figure 5: HAADF-STEM image of a large ZnPd particle with dark Pd-depleted regions. The Pd-depleted regions penetrate the surface where Zn can be oxidized (arrows).

In summary, the ZnPd particles show an inhomogeneous elemental distribution after initial reduction of the material and even before being exposed to MSR conditions. Pd-depleted Znrich regions are found within the particles, having a dark contrast in the HAADF-STEM images. These extend towards the surface, where they become capped by ZnO in the oxidative environment of the MSR reaction mixture. Although the possibility that the inhomogeneity is directly caused by the oxidation of the Zn-rich areas cannot be ruled out, this appears unlikely. Coarsening of the Zn-rich areas inside the particles due to heating during the MSR reaction may assist the formation of localized ZnO surface plumes, which may extend deeply into the particles due to their channel-like appearance. All these findings indicate that the ZnO patches are formed by oxidation of the particles directly rather than by transport from the ZnO substrate to the particle surface. The latter 12 ACS Paragon Plus Environment

Page 12 of 17

Page 13 of 17

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

pathway is highly unlikely, since diffusion of oxidized Zn species from the support must include partial reduction of ZnO and the associated formation of substoichiometric Zn oxide species,17 subsequently partially encapsulating the ZnPd particles. Furthermore, if the ZnO patches would be the result of a true SMSI effect, complete coverage of the ZnPd particles would be expected, not being limited just to the Zn-rich areas. Such a complete encapsulation by graphitic ZnO on Cu/ZnO catalysts as a result of a SMSI effect was recently demonstrated by Lunkenbein and coworkers.13 Our presented results provide direct evidence for a dynamical state of the catalyst in MSR conditions, adapting itself to the specific catalytic conditions, subsequently giving rise to lowered activation barriers for the associated catalytic reaction to CO2. In the case of ZnPd/ZnO, the simple presence of the intermetallic compound ZnPd is definitely insufficient to explain the high CO2 selectivities. Rather, the catalyst dynamically “selectivates” during the induction period by surface oxidation of the Zn-rich areas, thereby creating a highly synergistic ZnPd-ZnO interface, enabling efficient water activation and, in turn, high CO2selectivity. Note that the only partial coverage of the ZnPd particles by ZnO is also in line with the catalytic results. Full encapsulation of the ZnPd particles would inevitably diminish the overall activity of the catalyst, as only the catalytic activity of ZnO alone would be measured, which was recently established as being rather low.18

4. CONCLUSION Using aberration-corrected electron microscopy, electron energy loss spectroscopy and in situ heating experiments, a likely formation mechanism of ZnO patches at ZnPd/ZnO, which account for the high selectivity of this catalyst in methanol steam reforming, is suggested. We demonstrate that during contact with the MSR reaction mixture at elevated temperature, Znrich regions penetrating the surface of the ZnPd particles oxidize and become capped by ZnO. 13 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

EELS lines scans and EDX spectrum images demonstrate, that the formation of ZnO is localized at the Zn-rich areas. This effect is supported by the coarsening of the Zn-rich areas, which takes place at elevated temperatures during the MSR reaction. Furthermore, it is demonstrated that the ZnO patches form by oxidation of the particles directly rather than by transport from the ZnO substrate, thus ruling out a classical SMSI in ZnPd/ZnO.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]. Tel.: 49-2461-619479.

ACKNOWLEDGEMENTS S.P. thanks the FWF for financial support under project F45-N16 and M.H. and M.A. acknowledge financial support by Deutsche Forschungsgemeinschaft (DFG) under grant HE 7192/1-1 and AR 617/3-1, respectively. We thank M. Behrens and R. Schlögl for insightful discussion of the “plume-effect”.

ASSOCIATED CONTENT Supporting Information Additional microstructural analyses: Quantitative EELS analysis, EELS spectrum images, and a comparison of high-resolution TEM and corresponding HAADF-STEM images of the same ZnPd particles capped with ZnO patches.

14 ACS Paragon Plus Environment

Page 14 of 17

Page 15 of 17

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

REFERENCES (1) Iwasa, N.; Takezawa, N. New Supported Pd and Pt Alloy Catalysts for Steam Reforming and Dehydrogenation of Methanol. Top. Catal. 2003, 22, 215-224. (2) Behrens, M.; Armbrüster, M. Catalysis for Alternative Energy Generation; Springer: New York, 2011. (3) Föttinger, K.; van Bokhoven, J.A.; Nachtegaal, M.; Rupprechter, G. Dynamic Structure of a Working Methanol Steam Reforming Catalyst: In Situ Quick-EXAFS on Pd/ZnO Nanoparticles. J. Phys. Chem. Lett. 2011, 2, 428-433. (4) Friedrich, M.; Penner, S.; Heggen, M.; Armbrüster, M. High CO2 Selectivity in Methanol Steam Reforming through ZnPd/ZnO Teamwork. Angewandte Chemie Int. Ed. 2013, 52, 4389-4392. (5) Friedrich, M.; Teschner, D.; Knop-Gericke, A.; Armbrüster, M. Influence of Bulk Composition of the Intermetallic Compound ZnPd on Surface Composition and Methanol Steam Reforming Properties. J. Catal. 2012, 285, 41-47. (6) Armbrüster, M.; Behrens, M.; Föttinger, K.; Friedrich, M.; Gaudry, É.; Matam, S. K.; Sharma, H. R. The Intermetallic Compound ZnPd and Its Role in Methanol Steam Reforming. Catalysis Reviews: Science and Engineering 2013, 55, 289-367. (7) Tauster, S.J.; Fung, S.C.; Garten, R.L. Strong Metal-Support Interactions. Group 8 Noble Metals Supported on Titanium Dioxide. J. Am. Chem. Soc. 1978, 100, 170-175. (8) Jia, C. L.; Lentzen, M.; Urban, K. Atomic-Resolution Imaging of Oxygen in Perovskite Ceramics. Science 2003, 299, 870-873. (9) Lentzen, M. Progress in Aberration-Corrected High-Resolution Transmission Electron Microscopy Using Hardware Aberration Correction. Microscopy and Microanalysis 2006, 12, 191-205. (10) Claeyssens, F.; Freeman, C. L.; Allan, N. L.; Sun, Y.; Ashfold, M. N. R.; Harding, J. H.; Growth of ZnO Thin Films - Experiment and Theory. J. Mater. Chem. 2005, 15, 139-148. (11) Freeman, C. L.; Claeyssens, F.; Allan, N. L.; Harding, J. H. Graphitic Nanofilms as Precursors to Wurtzite Films: Theory. Phys. Rev. Lett. 2006, 96, 066102.

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(12) Tusche, C.; Meyerheim, H. L.; Kirchner, J. Observation of Depolarized ZnO(0001) Monolayers: Formation of Unreconstructed Planar Sheets. Phys. Rev. Lett. 2007, 99, 026102. (13) Lunkenbein, T.; Schumann, J.; Behrens, M.; Schlögl, R.; Willinger, M. G. Formation of a ZnO Overlayer in Industrial Cu/ZnO/Al2O3 Catalysts Induced by Strong Metal-Support Interactions. Angew. Chem. Int. Ed. 2015, 127, 4627-4631. (14) Hartel, P.; Rose, H.; Dinges, C. Conditions and Reasons for Incoherent Imaging in STEM. Ultramicroscopy 1996, 63, 93-114. (15) Nellist, P.D.; Pennycook, S.J. Incoherent Imaging Using Dynamically Scattered Coherent Electrons. Ultramicroscopy 1999, 78, 111-124. (16) Ota, A.; Kröhnert, J.; Weinberg, G.; Kasatkin, I.; Kunkes, E.L.; Ferri, D.; Girgsdies, F.; Hamilton, N.; Armbrüster, M.; Schlögl, R.; et al. Dynamic Surface Processes of Nanostructured Pd2Ga Catalysts Derived from Hydrotalcite-Like Precursors. ACS Catal. 2014, 4, 2048-2059. (17) Penner, S.; Jenewein, B.; Gabasch, H.; Klötzer, B.; Wang, D.; Knop-Gericke, A.; Schlögl, R.; Hayek, K. Growth and Structural Stability of Well-Ordered PdZn Alloy Nanoparticles. J. Catal. 2006, 241, 14-19. (18) Lorenz, H.; Friedrich, M.; Armbrüster, M.; Klötzer, B.; Penner, S. ZnO is a CO2Selective Steam Reforming Catalyst, J. Catal. 2013, 297, 151-154.

16 ACS Paragon Plus Environment

Page 16 of 17

Page 17 of 17

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

TOC GRAPHIC

17 ACS Paragon Plus Environment