Low-Temperature Methane Partial Oxidation to Syngas with Modular

View Sections. ACS2GO © 2018. ← → → ←. loading. To add this web app to the home screen open the browser option menu and tap on Add to hom...
2 downloads 0 Views 2MB Size
Subscriber access provided by Georgetown University | Lauinger and Blommer Libraries

Article

Low-Temperature Methane Partial Oxidation to Syngas with Modular Nanocrystal Catalysts Emmett D. Goodman, Allegra A Latimer, An-Chih Yang, Liheng Wu, Nadia Tahsini, Frank Abild-Pedersen, and Matteo Cargnello ACS Appl. Nano Mater., Just Accepted Manuscript • DOI: 10.1021/acsanm.8b01256 • Publication Date (Web): 16 Aug 2018 Downloaded from http://pubs.acs.org on August 20, 2018

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 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 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.

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 39 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

ACS Applied Nano Materials

Low-Temperature Methane Partial Oxidation to Syngas with Modular Nanocrystal Catalysts Emmett D. Goodman,1 Allegra Latimer,1 An-Chih Yang,1 Liheng Wu,1,2 Nadia Tahsini,1 Frank Abild-Pedersen,1,3 Matteo Cargnello1,* 1

Department of Chemical Engineering and SUNCAT Center for Interface Science and Catalysis,

Stanford University, CA 94305, USA. 2

Stanford Synchrotron Radiation Lightsource, SLAC National Accelerator Laboratory, 2575

Sand Hill Road, Menlo Park, California 94025, USA. 3

SLAC National Accelerator Laboratory, SUNCAT Center for Interface Science and Catalysis,

2575 Sand Hill Road, Menlo Park, California 94025, USA.

*Correspondence to: [email protected]

Keywords: methane partial oxidation; ruthenium; nanocrystal; catalyst; syngas

ACS Paragon Plus Environment

ACS Applied Nano Materials 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

ABSTRACT The low-temperature conversion of methane into value-added products is an appealing goal due to the abundance of methane in the form of natural gas. Industrially, methane is used to produce synthesis gas (syngas), a precursor mixture used heavily in the production of ammonia, methanol, and synthetic fuels. In practice, this mixture is produced via the energy-intensive methane steam reforming reaction at temperatures between 750-1450 oC. The exothermic methane partial oxidation reaction stands as an alternative for syngas formation at lower temperatures, especially for gas to liquid fuels applications, yet awaits large-scale implementation due to dangerous operating conditions and temperatures. Using colloidallysynthesized Ru catalysts, we identify two unifying rules that govern the low-temperature production of synthesis gas: depletion of oxygen within the catalyst bed, and facile RuO2 → Ru reduction kinetics, which is a strong function of supporting material and Ru nanostructure. Using these design rules, we demonstrate the enhanced low-temperature activity of a bifunctional Ru/Pd catalyst which produces synthesis gas at ~400 oC, with nearly complete CH4 conversion and CO selectivity at 670 oC.

ACS Paragon Plus Environment

Page 2 of 39

Page 3 of 39 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

ACS Applied Nano Materials

INTRODUCTION Due to the development of fracking technologies, natural gas is an abundant resource with projected US reserves of approximately 60 trillion cubic meters1. Of these reserves, 600 billion cubic meters are consumed annually in the United States, much of which is used as a feedstock for the production of synthesis gas (syngas), an important industrial mixture used for the synthesis of chemicals and fuels such as ammonia, methanol, and synthetic petroleum2,3. Industrially, syngas formation is most commonly achieved via the energy-intensive, hightemperature process of methane steam reforming (MSR, CH4 + H2O  CO + 3H2)4. While this reaction is very convenient for ammonia synthesis because it provides a large H2/CO ratio of ~3, it is strongly endothermic (∆H=206 kJ mol-1), and requires high temperatures to achieve a favorable equilibrium conversion. Therefore, much research has been focused on the more economical conversion of methane, the main component of natural gas, into syngas at milder conditions. Methane partial oxidation (MPO, CH4 + ½O2  CO + 2H2) is an appealing alternative to MSR for the formation of syngas. This reaction has the energetic benefit of being mildly exothermic (∆H=-36 kJ mol-1), and produces a H2/CO ratio of ~2, which is more favorable for the synthesis of hydrocarbon fuels5,6. Due to this desirable syngas ratio, non-catalytic MPO processes have been developed and in some cases industrially implemented, although reactions are performed under harsh conditions, at temperatures as high as 1500 oC and at pressures up to 150 bar7. However, in the catalytic oxidation of methane, the potentially ignitable mixture of CH4 and O2 can lead to dangerous conditions before or within the catalyst bed3,8. Specifically, the concomitant methane complete combustion reaction (CH4 + 2O2  CO2 + 2H2O) may lead to reactor hot spots. A related challenge is the lack of understanding regarding which factors govern

ACS Paragon Plus Environment

ACS Applied Nano Materials 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

Page 4 of 39

low-temperature selectivity to syngas versus complete combustion products. For these reasons, industrial adoption of methane partial oxidation for syngas production remains problematic. To mitigate such setbacks and to improve MPO viability, catalysts and conditions that favor selective syngas formation at low temperatures need to be developed. Due to their high activity, precious metal catalysts are often studied as active phases for the partial oxidation of methane to syngas9,10. Among these, Rh and Ru are generally accepted as two of the best MPO catalysts, and their high costs have led to a large body of research focusing on their more economical use11. Cheaper metals such as Ni have been extensively studied, but suffer from sintering, coking, and lower activity12–14. Given that Ru is one of the cheapest among noble metals (and an order of magnitude less expensive than Rh12) and possesses higher activity and coke-resistance than base metals15, it is an excellent compromise between cost and activity. Ruthenium catalysts have been studied in the past as effective methane activation catalysts, especially for reactions involving the conversion of methane to syngas16. For MPO, activity has

been

linked

to

oxidation

state17–20, support

properties17,19,21

and

Ru

morphology/dispersion18. In particular, bulk characterization techniques such as near edge x-ray absorption fine structure (XANES)20, x-ray photoelectron spectroscopy (XPS)17 and x-ray diffraction (XRD)18, provide strong evidence correlating high selectivity towards complete combustion products with the oxidized RuO2 phase, and high selectivity towards syngas with the reduced Ru metal phase. For example, Verykios et al. have explored the support dynamics of Ru/TiO2 materials, demonstrating that by stabilizing a reduced Ru phase, lower temperature syngas production is achievable17,21. Despite these efforts, past work only partially contributed to increase the knowledge on the mechanism of Ru catalysts for partial oxidation that would enable

ACS Paragon Plus Environment

Page 5 of 39 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

ACS Applied Nano Materials

the rational design of these materials. It is the aim of our work to provide these design rules through structure-property relationship studies. Many studies of MPO emphasize a crucial ‘ignition temperature’5,18,22–24. This temperature is the critical value at which the reaction products change from combustion (CO2, H2O) to partial oxidation (CO, H2), and is often accompanied by a large increase in methane conversion. Relatively recently, using a high-speed x-ray camera, Kimmerle et al. characterized this ignition period over a Rh-Pt catalyst, and found dramatic changes in material oxidation state occuring over a period of just tens of seconds, which was associated with a drastic change in catalytic performance

23

. To our knowledge, however, no general set of

guidelines has been developed to predict and tune this ignition temperature and onset of syngas products. Literature lacks in-depth catalytic characterization on the rapid time-scale of MPO ignition. Additionally, although many studies reporting bulk correlations between products and Ru oxidation state have emerged17–20, single-particle nanoscopic analysis remains absent. In catalyst optimization, an understanding of the relationship between catalytic activity and material nanostructure is critical. However, in traditional synthetic techniques such as incipient wetness impregnation (IWI), or in techniques involving strong thermal treatments, the initial state of the active phase is closely tied to the properties of the catalytic support. Therefore, Ru catalysts on different supports might possess drastically different nanostructures. The amount of exposed Ru, the relative composition of Ru, RuO2, and RuO4 oxidation states, and the precise Ru morphology are all variable. This fact precludes a fair comparison of the important role the catalyst support plays in Ru-catalyzed MPO. Furthermore, such prepared catalysts are often inhomogeneous, and thus local characterization techniques such as electron microscopy in any specific region might not be reflective of the bulk properties of the catalyst. These issues lead to

ACS Paragon Plus Environment

ACS Applied Nano Materials 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

difficulties in monitoring catalytic evolution, drawing clear structure-property relationships, and developing design rules. In this work, we investigate the methane partial oxidation reaction on well-defined Ru nanocrystals supported on thermally-stabilized Al2O3, SiO2, Ce0.8Zr0.2O2, (CZ80) and MgO supports and systematically explore the effect of reaction conditions. We employ preformed nanocrystals synthesized via colloidal techniques as active supported phases, which allows us to ensure that, across supports, the active phase starts with the same dispersion, oxidation state, and morphology. In this way, catalytic activity and structural evolution are solely a function of support character. Catalysis on these well-defined systems coupled with quick-sampling mass spectrometry allows characterization of this quickly ‘igniting’ reaction with excellent time resolution. Additionally, with uniform starting materials, non-bulk characterization techniques such as transmission electron microscopy (TEM) are representative of properties occurring throughout the entire system, and can help understand catalytic evolution at the nanoscopic, single-particle level. These investigations produce two unifying rules for igniting lowtemperature syngas production. Across supports, we identify oxygen depletion in the reactor bed as a prerequisite for methane partial oxidation. We find facile RuO2 → Ru kinetics as a function of Ru nanostructure and support to be crucial in the low-temperature onset of syngas production. Taking advantage of these design rules, we suggest the development of highly active bifunctional MPO materials, which take advantage of a palladium phase to combust methane at lower temperatures, and a highly reducible ruthenium phase to form low-temperature partial oxidation products once oxygen is depleted within the reactor. These materials show thermodynamicallylimited production of syngas products at temperatures starting at temperatures as low as 405 °C. Despite previous works studied the effect of support and reaction conditions on the performance

ACS Paragon Plus Environment

Page 6 of 39

Page 7 of 39 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

ACS Applied Nano Materials

of Ru MPO catalysts, we believe our materials finally prove how these parameters affect the reactivity of Ru, how conditions provide the production of syngas, and how to tune these parameters to improve the activity (by adding a methane combustion catalyst such as Pd, for example).

EXPERIMENTAL SECTION Methods Nanocrystal Synthesis All syntheses were performed using standard air-free Schlenk techniques. Ru(acac)3 (97%, Aldrich) and Ru3(CO)12 (99%, Acros Organics) were used as ruthenium metal precursors. The general synthetic methodology is as follows: Ru metal salts were added to a three-neck flask together with 9 mL 1-oleylamine (70%, Aldrich) at room temperature. For large particles, Ru(acac)3 = 20 mg, Ru3(CO)12 = 42 mg; for small particles, Ru(acac)3 = 80 mg, Ru3(CO)12 = 10 mg The mixture was evacuated at 30 oC for 30 min at a residual pressure 100 nanoparticle diameters by hand using ImageJ software. For ex-situ TEM measurements, in order to preserve working catalyst state, the catalyst was immediately flushed with 45 mL min-1 pure Ar, and rapidly cooled down to room temperature at ~300 oC min-1. Small angle X-ray scattering (SAXS) measurements were performed at Beamline 1-5 at the Stanford Synchrotron Radiation Lightsource (SSRL) of the SLAC National Accelerator Laboratory using a Rayonix 165 SX CCD area detector. Scattering patterns were analyzed by

ACS Paragon Plus Environment

ACS Applied Nano Materials 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

Page 10 of 39

fitting to a quantitative model using the IRENA package (J. Ilavsky and P. R. Jemian, Irena: Tool suite

for

modeling

and

analysis

of

small-angle

scattering,

available

at

usaxs.xray.aps.anl.gov/staff/ ilavsky/irena.html from the APS)29 to determine the size and size distribution of the as-synthesized nanocrystals.

Catalytic Characterization In general, catalytic testing was performed under conditions such that each reactor bed had the same mass of Ru, and consequently, the same initially exposed Ru nanoparticle surface area. In certain experiments, GHSV was altered by changing the amount of catalyst and dilution ratio, keeping the total bed volume, and total flow rate, constant. Due to the high temperatures required for this reaction, prior to nanocrystal impregnation, each of our catalyst supports were calcined to atleast 700 oC, to ensure minimal support change during reaction conditions. To minimize thermal gradients and bed hot-spots, low wt % (0.5% Ru/Support) catalysts were synthesized, and mixed with large volumes of diluent. In general, each catalyst, post-ligand removal, was sieved below 180 µm grain size and mixed with Al2O3 diluent (1:20, catalyst:alumina), which was found to be sufficient to eliminate thermal effects by repeated tests. 200 mg of this diluted mixture was loaded into the reactor to give a bed length of about 1.0 cm; this bed rested between two layers of granular quartz which were used for preventing displacement of the catalyst powder and for preheating the reactant gases. The reactor was heated by a square furnace (Micromeritics) and the temperature of the catalyst was measured with a K-type thermocouple inserted inside the reactor, touching the catalytic bed. All experiments were conducted at a total pressure of one atmosphere. Ignition/extinction curves

ACS Paragon Plus Environment

Page 11 of 39 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

ACS Applied Nano Materials

were performed in a U-shaped quartz microreactor with an internal diameter of 10 mm, and a temperature ramp rate of 10 oC min-1. The reactant mixture composition was controlled by varying the flow rates of CH4(5%)/Ar, O2(5%)/Ar, H2(5%)/Ar and Ar (all certified mixtures with purity >99.999% from Airgas). The following procedure represents standard reaction conditions. In all catalyst tests, the catalyst was oxidatively cleaned in-situ of residual organics and carbonates within the reactor bed under O2(5%)/Ar flow at 45 mL min-1 at 275 oC for 30 min, then pure Ar at 45 mL min-1 at 275 oC for 10 min, and subsequently reduced under H2(5%)/Ar flow at 45 mL min-1 at 275 oC for 30 min. The reactor was then cooled to 200 oC in pure Ar at 45 mL min-1, at which point the reactant mixture was introduced, and the reactor was ramped to 670 oC at 10 oC/min. A GHSV of 100k indicates 17.5 mL min-1 of a 3.05% CH4, 1.5% O2, remainder Ar mixture. For CH4 temperature-programmed reduction (CH4-TPR) experiments, the catalysts were preoxidized at 275 oC for 30 min and rapidly cooled down to room temperature under 45 mL/min flowing Ar. At this point, a mixture of 2.3% CH4, balance Ar was introduced, to mimic reaction conditions observed within the bed after O2 is depleted. The reactor bed was slowly ramped at 3 oC/min to 700 oC. Reactor effluent was measured using an online mass spectrometer (Hiden HPR-20). By comparing to calibrations of standard mixtures, carbon balances were always better than 90%, and generally better than 95%. Comparing data between a mass-spectrometer detector and a gas chromatograph, CH4 conversion was comparable to within 2%, and CO selectivity was slightly underestimated using the mass-spectrometer (Figure S2).

Computational procedures

ACS Paragon Plus Environment

ACS Applied Nano Materials 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 plane-wave QuantumESPRESSO code30 and bayseian error estimation functional with van der Waals corrections (BEEF-vdW) functional31 were used for the density functional theory (DFT)32,33 calculations. The plane-wave and density cut-offs were 550 and 5500 eV respectively. Forces on all atoms were minimized to 0.05 eV Å-1. A (6,6,1) k-point sampling was used on a 2x1 expansion of the (110) rutile surface unit cell for RuO2, a 3x3 expansion of the surface unit cell of Ru(0001), and a 3x4 expansion of the surface unit cell of Ru(101ത0). Slabs were four stoichiometric layers separated by 15 Å vacuum. The lowest two layers were fixed to simulate the bulk. Fermi-Dirac smearing with a sigma of 0.1 was used. Vanderbilt pseudopotentials were used for ruthenium34. Climbing-image nudged elastic band (CI-NEB)35–38 calculations were performed to find the transition states. The accuracy of these transition states was verified by identifying an imaginary mode corresponding to the transition state reaction coordinate. Occasionally, one or two other small imaginary vibrational modes were also present, but they were verified to correspond to low-energy rotations and were approximated to be 7 meV. Microkinetic modeling was performed with CatMAP39, a software package employing a self-consistent mean field model. Free energies were obtained by summing electronic energies, zero point energies, and entropic contributions determined from a harmonic adsorbate model. Gas phase species were treated ideally. To correct the known errors in DFT-predicted formation energy of water, either a -0.4 eV correction to H2O(g)40 or a 0.8 eV correction to O2(g)41 can be applied. Both correction schemes were tried and found no difference in the results of the microkinetic model.

The software package Cantera42 was used to estimate the gas-phase

thermodynamic equilibrium.

RESULTS AND DISCUSSION

ACS Paragon Plus Environment

Page 12 of 39

Page 13 of 39 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

ACS Applied Nano Materials

Preparation and Characterization of Supported Ruthenium Nanocrystals Catalysts The separate synthesis and impregnation of uniform ruthenium active phases was employed in order to study the support effect of Al2O3, SiO2, CZ80, and MgO on Ru-catalyzed methane partial oxidation. In the synthesis of Ru nanocrystals, different ratios of Ru(acac)3 and Ru3(CO)12 precursors were utilized to tune the size of the Ru nanocrystals to either 1.7 nm or 5.2 nm (Figures 1a, S1). Transmission electron microscopy images show that the obtained nanocrystals are crystalline and uniform in size, with an average size of 5.2±0.4 nm. Small angle X-ray scattering (SAXS) measurements corroborated this observation, although fitting of the scattering pattern produces an average particle size slightly smaller than TEM at 4.5±0.7 nm (Figure 1b).

By depositing Ru nanocrystals from the same synthesis batch onto different

thermally-stabilized support materials with same weight loading (0.5 wt. %), we prepared catalyst materials with identical active phases in order to study the effect of the catalyst support without interference from changes occurring to the supported phase. After nanocrystal impregnation, a rapid thermal annealing procedure (700

o

C, 30s) was used to remove

nanoparticle ligands and activate the catalysts, while maintaining the uniformity of the nanocrystal active phases28. TEM characterization and particle size analysis (Figure 1c-j) reveals that very little changes occurred in average size and size distribution after deposition and ligand removal.

ACS Paragon Plus Environment

ACS Applied Nano Materials 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

Figure 1 | Ru nanocrystals and supported Ru nanocrystal catalysts utilized in this work. Ru nanocrystals supported on (a) carbon square mesh TEM grid; insert nanoparticle size distribution. (b) Small angle X-ray scattering (SAX) of colloidal nanocrystals, and SAXS fit. Nanocrystals supported on (c) Al2O3, (d) SiO2, (e) Ce0.8Zr0.2O2, and (f) MgO. Supported catalysts are shown after ligand removal at 700 oC for 30 s. Below (g-j) are corresponding particle size distributions, where more than 100 particles were counted.

Low-Temperature Catalytic Activity for Syngas Production via Methane Partial Oxidation Due to the high reaction temperatures, thermal hot-spots, and inherently large thermal and concentration gradients in the methane partial oxidation reaction, light-off curves (ignition/extinction, temperature ramps) and steady-state measurements were utilized to derive design rules for low-temperature catalytic activity using heavy dilution of the catalytic powders

ACS Paragon Plus Environment

Page 14 of 39

Page 15 of 39 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

ACS Applied Nano Materials

to mitigate these problems. Meaningful TOFs were unable to be extracted due to the peculiar change in catalytic behavior before and after the ignition temperature (see below). As methane partial oxidation is known to be a quickly igniting reaction, high-frequency sampling (0.083 s-1) mass spectroscopy was chosen as the analytical technique to best characterize the rapid onset of MPO products. Comparion of the same catalyst using a mass-spectrometer versus a gaschromatograph can be found in Figure S2, highlighting the importance of using a quick-sampling detector. Clear differences in activity, stability, and reducibility of Ru catalysts were observed as a function of support (Figure 2), with each material showing a unique temperature-dependent activity and partial oxidation selectivity profile (Figure 2a). Interestingly, at temperatures below 500 oC, SiO2-, Al2O3-, and CZ80-supported materials show near-identical CH4 conversion profiles. At this point, all materials produce only complete combustion products CO2 and H2O. At ~500 oC, the samples begin to differ: Ru/SiO2 starts producing CO and H2 at the lowest temperature of 517 oC, as evidenced by the drastic increase in CO selectivity at this temperature. Ru/Al2O3 follows with a switch from complete to partial oxidation at 544 oC, and finally Ru/CZ80 at 566 oC. Ru/MgO fails to produce CO or H2 products even at temperatures approaching 700 oC. Notably, after the ignition temperature of each material, conversion and selectivity profiles quickly increase towards equilibrium values. Similar results were observed on supported 1.7 nm Ru particles (Figure S3). The H2/CO ratio is initially very high right after the ignition temperature because the WGSR is equilibrated and any small CO amount is converted into CO2.

ACS Paragon Plus Environment

ACS Applied Nano Materials 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

Figure 2 | Support-dependent MPO behavior of Ru nanocrystal catalysts. (a) CH4 conversion (top), CO selectivity (middle), and H2/CO ratio (bottom) of supported Ru-nanocrystal catalysts in methane partial oxidation reaction. Feed: 3.05 mol % CH4, 1.5 mol % O2, remainder Ar. GHSV: 100k mL g-1 h-1 (standard reaction conditions). Black lines show corresponding calculated equilibrium values. (b) Onset temperature of CO production (ignition temperature) as a function of cycle number. (c) CH4 temperature-programmed reduction curves of pre-oxidized supported Ru catalysts.

Ru-catalyzed MPO selectivity has previously been correlated with dynamics of the RuO2  Ru + O2 phase transition17–20. Under MPO conditions, this phase transition is accelerated by

ACS Paragon Plus Environment

Page 16 of 39

Page 17 of 39 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

ACS Applied Nano Materials

reaction with CH4. To investigate this aspect, we performed CH4 temperature programmed reduction (TPR) experiments on each pre-oxidized catalyst. In each experiment, we started with supported RuO2 nanocrystals, and heated these materials in the presence of flowing CH4. By monitoring the signal for CH4 uptake, we can identify the temperature at which methane reacts with and converts RuO2 into a metallic Ru phase. Remarkably, with the exception of Ru/MgO, the ignition experiments (Figure 2a) and TPR experiments (Figure 2c) order in the same way. Ru/SiO2 shows the lowest temperature and smallest reduction feature, followed by Ru/Al2O3, and finally by Ru/CZ80. However, the Ru/CZ80 sample uptakes a much larger volume of CH4 during this reduction process, and at higher temperatures, suggesting that the reduction of the ceria phase in contact with the metal occurs as well43. This phenomenon is suggested to be reason for the plateaued activity of the Ru/CZ80 material between 500 oC and 580 oC, in which RuO2 is not fully reduced to the metallic MPO active phase until the CeO2 phase is. Although Ru/MgO does show a reduction feature, other processes occur to hinder catalytic activity towards MPO (see below). Qualitatively, we can understand why Ru/SiO2 and Ru/Al2O3 produce syngas at the lowest temperatures, followed by Ru/CZ80: RuO2 is more reducible when supported on the prior than the latter due to the oxygen donation abilities of ceria that maintain Ru in a more oxidized, and unreactive, MPO state. Thus, across supports, we can order activity as SiO2