TEM Study of Rhodium Catalysts with Manganese Promoter

TEM Study of Rhodium Catalysts with Manganese Promoter Adrian Merritt NSF REU program at UIC, 7/29/2010 1

Outline I. Research Objectives and Methods II. Sample Characterization III. Particle Size Results IV. Research Conclusion V. Future Work NSF REU program at UIC, 7/29/2010 UIC Physics 2

Research Objectives �The core objective is to better understand how the manganese promoter affects the rhodium catalyst performance �Some current possibilities are: • Particle size • Oxide species • Changes to interfacial interaction • Formation of surface oxides NSF REU program at UIC, 7/29/2010 UIC Physics 3

� Due to the de Broglie wavelength, electron microscopes can have a fundamentally finer resolution than light microscopes � Electrons passing through the sample are scattered by various mechanisms � Spatial, mass/thickness and analytical information is available from the scattered electrons TEM Image from Transmission Electron Microscopy, B. Williams and C. Carter, volume IV NSF REU program at UIC, 7/29/2010 UIC Physics 4

Fischer-Tropsch (and related) Processes � Invented by Franz Fischer and Hans Tropsch � Utilizes syngas to produce hydrocarbon products (methane, ethanol, diesel and gasoline fuels) � Syngas is a mixture of CO and H 2, which can be produced from coal gasification, natural gas, or biogas, and is used as the base feedstock for the process � In all cases though, the reaction relies upon the proper catalysts for selectivity and efficiency NSF REU program at UIC, 7/29/2010 UIC Physics 5

Rhodium Catalyst, Manganese Promoter � Rhodium is a useful catalyst for the FT process as it lies at an intermediate mass level and so works to create ethanol for use as an alternative fuel source � Manganese acts as a promoter, which changes the effects of a catalyst without being a catalyst itself � Manganese improves the selectivity and overall efficiency of rhodium catalysts for the FT process � E. g. from T. Feltes: 1% Mn loading on 3% Rh on Si. O 2 support raises CO conversion ten fold and increases ethanol selectivity from 0. 0% to 9. 2% Image from The Selective Adsorption of a Manganese Promoter Over Supported CO Hydrogenation Catalysts, Theresa E. Feltes, 2010 NSF REU program at UIC, 7/29/2010 UIC Physics 6

Holey Carbon Films � Carbon film on copper support grid � d = 3 mm � Allows deposition of catalyst particles and easy viewing � Powdered samples are prepared by dry impregnation (DI) or strong electrostatic adsorption (SEA) NSF REU program at UIC, 7/29/2010 UIC Physics 7

Final Sample �Final sample has many medium-sized clusters of silica particles �Best (most useful) clusters are those overhanging an edge (reduces impact of Cfilm) NSF REU program at UIC, 7/29/2010 UIC Physics 8

Current Samples � � Rhodium on silica, 3% loading by DI with 1% manganese Calcination at 350° C for 4 hours in air Reduction (when applicable) at 300° C for 2 hours under H 2 Images from The Study of Heterogeneous Catalysts by Highflow Resolution Transmission Electron Microscopy, A. Datye & D. Smith, Catalyst Review, 1992 NSF REU program at UIC, 7/29/2010 UIC Physics 9

Imaging Samples � Typical magnification is x 300 k � Use diffraction contrast imaging to differentiate rhodium particles (crystalline) from the silica support (amorphous) NSF REU program at UIC, 7/29/2010 UIC Physics 10

Particle Sizes (Unpromoted) Averages: 3. 12 nm vs. 3. 08 nm Standard deviations: 0. 80 nm vs. 0. 83 nm The same (within experimental limits)! 25 Rh. Ox (6. 18. 2010) Rh (Reduced) (6. 21. 2010) 25 20 20 Frequency 15 10 5 5 0 0. 5 1 1. 5 2 2. 5 3 3. 5 4 4. 5 5 5. 5 >5. 5 Particle Size (nm) NSF REU program at UIC, 7/29/2010 1 1. 5 2 2. 5 3 3. 5 4 4. 5 Particle Size (nm) 5 5. 5 >5. 5 UIC Physics 11

Particle Sizes (Promoted) Averages: 2. 26 nm vs. 2. 44 nm Standard deviations: 0. 54 nm vs. 0. 67 nm Rh. Ox + Mn (7. 7. 2010) 45 40 40 35 35 30 30 Frequency 45 25 20 Rh+Mn (Reduced) (7. 7. 2010) 25 20 15 15 10 10 5 5 0 0 0. 5 1 1. 5 2 2. 5 3 3. 5 4 4. 5 Particle Size (nm) 5 5. 5 >5. 5 NSF REU program at UIC, 7/29/2010 0. 5 1 1. 5 2 2. 5 3 3. 5 4 4. 5 Particle Size (nm) 5 5. 5 >5. 5 UIC Physics 12

Particle Sizes (Promoted, in situ Heating) 35 30 30 25 25 20 20 Frequency Averages: 2. 55 nm vs. 2. 43 nm Standard deviations: 0. 91 nm vs. 0. 69 nm Heated at 300° C for 2 hours, then allowed to cool 35 Rh+Mn (Heated) (7. 19. 2010) 15 15 10 10 5 5 0 Rh+Mn (Cooled) (7. 9. 2010) 0 0. 5 1 1. 5 2 2. 5 3 3. 5 4 4. 5 Particle Size (nm) 5 5. 5 >5. 5 0. 5 NSF REU program at UIC, 7/29/2010 1 1. 5 2 2. 5 3 3. 5 4 4. 5 Particle Size (nm) 5 5. 5 >5. 5 UIC Physics 13

Summary Sample Average Particle Size (nm) Standard Deviation (nm) Rh. Ox (unreduced) 3. 12 0. 80 Rh 3. 08 0. 83 Rh+Mn Ox (unreduced) 2. 26 0. 54 Rh+Mn 2. 44 0. 67 Rh+Mn (in situ heating) 2. 55 0. 91 Rh+Mn (after cooling) 2. 43 0. 67 Averages not different enough to cause all phenomena observed in catalysts with a promoter NSF REU program at UIC, 7/29/2010 UIC Physics 14

Future Work � Catalyst particle size has been ruled out � Next step is JEOL JEM-2010 F work • Better resolution through Z-contrast imaging • EELS setup � EELS allows changes in electronic structure to be characterized � Together, structure allows better characterization of NSF REU program at UIC, 7/29/2010 UIC Physics 15

FEFF Counts (Arb. units) Experimental vs. Theoretical 530 Rh 2 O 3 O K-Edge FEFF � University of Washington ab initio program for simulation EELS spectra � Full multiple scattering simulation � Preparation for JEM-2010 F EELS work, distinguishing rhodium oxide species Rh. O 2 O K-Edge FEFF Rh 2 O 3 O K-Edge Gatan 535 540 545 550 555 560 Energy (e. V) 565 570 575 NSF REU program at UIC, 7/29/2010 580 UIC Physics 16

Acknowledgements �National Science Foundation and Department of Defense for funding, EECNSF Grant # 0755115 �Professors Takoudis and Jursich as REU organizers �Professor Robert Klie as PI �Yuan Zhao as mentor �Ke-Bin Low for TEM training and aid �The RRC for its support in TEM work NSF REU program at UIC, 7/29/2010 UIC Physics 17