NREL is a national laboratory of the U. S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

Jesse E. Hensley†, Joshua A. Schaidle†, Daniel A. Ruddy‡, Singfoong Cheah†, Conversion of Dimethyl Ether to Branched § § Susan E. Habas†, Ming Pan†, Guanghui Zhang , and Jeffrey T. Miller Hydrocarbons Over Cu/BEA: the Roles of Lewis Acidic and Metallic Sites in H Incorporation †National Bioenergy Center, ‡Chemical and Materials Science Center, National Renewable Energy Laboratory 2 §Chemical Sciences and Engineering Division, Argonne National Laboratory Introduction Catalytic Testing Catalyst Characterization Dimethyl ether (DME) can be produced selectively from synthesis gas Hydrocarbon Productivity Structural and Chemical Characterization (CO and H2), which can be generated from renewable biomass feedstocks via gasification. Previous studies have reported that DME can Transmission Electron Microscopy • Similar induction profiles be converted to a mixture of C4–C7 hydrocarbons with high

) ∙s observed methanol/aromatic-free selectivity (~95%) over H-BEA zeolite catalysts at site 1-3 low temperatures (~200°C). The primary C7 hydrocarbon produced is • Cu addition increased rate 2,2,3-trimethylbutane (triptane), which is a valuable fuel additive having a per acid site by 1.5-2 times

research octane number of 112. The C5+ product mixture is also of molC/ mol when H2 present μ ( Cu and H produce effect interest as a high-octane fuel itself; further, C4+ olefin products hold 2 • Independent of whether Cu potential for coupling to distillate-range fuels. There are two proposed Production Hydrocarbon on zeolite or separated catalytic cycles for DME homologation, olefin and aromatic, which govern the selectivity for this reaction.4,5 Time on Stream (h) Hydrocarbon Selectivity 50nm 10nm • Carbon number distribution X-ray Absorption Spectroscopy was similar for all

experiments a.u.) • Addition of Cu caused a free Carbon free significant increase in the Cu+ Selectivity x chi(k)] x

paraffin/olefin ratios for C4- 2 C7 but only when on BEA Cu-O Methanol- FT [kFT 6h TOS and H2 present • Paraffin/Olefin decreases Normalized AbsorptionNormalized ( Carbon Number when Cu present and H2

absent or when Cu Energy (eV) R (Å)

 6h TOS separate from BEA BEA

Triptane - Treatment Components Ratio (%) • Cu/BEA shifted preference toward the olefin catalytic Cu(II)-zeolite 55 cycle Air, 500°C CuO 45 • Cu may remove ethylene as ethane, preventing it Cu(0) 86 from entering the Ratio Comparedto H Ratio

% % Paraffin/Olefin Difference H , 300°C Cu(II)-zeolite 10 aromatization pathway 2 Cu(I)-zeolite 4 Carbon Number (1) Ahn, J. H.; Temel, B.; Iglesia, E. Angew. Chem. Int. Ed. 2009, 48, 3814. (2) Simonetti, D. A.; Ahn, J. H.; Iglesia, E. J. Catal. 2011, 277, 173. r r C Treatment Scatter CN R (Å) DWF (Δσ2) ΔE (eV) (3) Hazari, N.; Iglesia, E.; Labinger, J. A.; Simonetti, D. A. Acc. Chem. Res. 2012, 45, 653. Formation rate ratio C C₂ HMB 0 (4) Ilias, S.; Bhan, A. J. Catal. 2012, 290, 186. r r C (5) Ilias, S.; Bhan, A. ACS Catalysis 2012, 3, 18. 2MBu₂ 2MBu HC Air, 500°C Cu-O 3.6 1.93 0.001 -1.70 Challenge #1: DME homologation to form alkanes is a hydrogen- Experiment 6h TOS XDME≈11% Total deficient process, resulting in the formation of alkylated aromatic H-BEA 0.23 0.16 0.12 Cu-Cu 9.5 2.54 0.001 0.92 residues such as hexamethylbenzene (HMB). These byproducts reduce H2, 300°C H-BEA + H2 0.21 0.15 0.07 Cu-O 1.3 1.92 0.001 0.32 yield, may cause catalyst deactivation, and limit catalyst lifetime.1-3 ox-Cu/BEA + H2 0.08 0.07 0.03 Consider the stoichiometric homologation of DME to triptane, which red-Cu/BEA 0.24 0.24 0.14 forces a 36% loss in C yield to HMB: Surface Acidity red-Cu/BEA + H2 0.08 0.08 0.04 Pyridine-DRIFTS Ammonia TPD + + H-BEA-Cu/SiO2 + H2 0.08 0.09 0.03 440°C

66C 42C 24C o 𝟑𝟑 𝟑𝟑 𝟕𝟕 𝟏𝟏𝟏𝟏 𝟐𝟐 𝟔𝟔 𝟑𝟑 𝟔𝟔 Isobutane Dehydrogenation at 300 C a.u.) 𝟑𝟑𝟑𝟑𝟑𝟑𝑯𝑯 𝑶𝑶𝑶𝑶𝑯𝑯 → 𝟔𝟔𝑪𝑪 𝑯𝑯 𝟑𝟑𝟑𝟑𝑯𝑯 𝑶𝑶 𝟐𝟐𝑪𝑪 𝑪𝑪𝑪𝑪 a.u.) Catalytic dehydrogenation of cracked products like Challenge #2: • 1% isobutane in He feed isobutane is not observed over HBEA at reaction conditions (180-220 °C) and alkanes are generally inactive in the catalytic cycle at these • H-BEA inactive for TCD Signal( temperatures, leading to significant production of C terminal products. recombinative H2 Absorbance( 4 abstraction β = 30°/min • Cu-containing catalysts Wavenumber (cm-1) Temperature (°C) Goals and Hypotheses produced H2 The goals of this work are to: • Cu loading and chemical Total Acid Brønsted Lewis Acid 1. Investigate H-atom management during DME homologation state impact activity Brønsted/ Catalyst Sites Acid Sites Sites Lewis Ratio 2. Develop a multi-functional metal-modified BEA zeolite catalyst (μmol/g) (μmol/g) (μmol/g) capable of: • Activating H 2 H-BEA 1642 1550 92 16.8 • Incorporating hydrogen into the desired products Isotopic Experiments ox-Cu/BEA 2058 772 1286 0.6 • Reducing deactivation H-D Exchange red-Cu/BEA 1918 1337 581 2.3

• Increasing selectivity to C7 products Our hypotheses are: • Red-Cu/BEA and ox-Cu/BEA a.u.) red-Cu/BEA possessed both metallic and cationic Cu sites 1. The addition of a Lewis-acidic metal center will lower the energetic activated D2 and evolved HD at • a significantly lower barrier for H2 activation and/or hydrogen abstraction from alkanes • Cationic Cu species introduced significant Lewis acidity and modified the temperature than H-BEA distribution of weak and strong acid sites 2. The addition of a metal hydrogenation center will facilitate Total amount of HD evolved • Metallic Cu was present on the outside of the BEA while cationic Cu was dissociation of H2 to Hads making it locally “available” for reaction • was nearly identical for all present within the pores catalysts suggesting exchange MS Intensity, m/z = 3=( m/z MSIntensity, Experimental Details of only –O-D for –O-H at Brønsted sites Catalyst Synthesis by Incipient Wetness Impregnation Temperature (°C) Summary and Conclusions

BEA (Beta) Zeolite Deuterium Incorporation • Cu/BEA achieved a two-fold increase in HC productivity when H2 cofed to Propylene Isobutane Triptane reactor Cu(NO ) •2.5H O 3 2 2 57 H-BEA + H2 H-BEA + H2 H-BEA + H2 Cu/BEA shifted preference toward products formed by the olefin catalytic H2O, dry at 50°C 41 • + n+ NH4 Cu 43 cycle over the aromatic catalytic cycle -NH NO 39 4 3 42 43 • The addition of Cu served to incorporate H2 into reaction products without 41 85 significantly impacting paraffin + olefin carbon selectivity

27 27 29 SiO2/Al2O3 = 27 Target = 5wt% Cu 58

(Tosoh) • Metallic Cu on the H-BEA surface H-BEA + D 41 H-BEA + D H-BEA + D 57 2 2 43 2 Cu functions hydrogenated olefins, increasing a.u.) 39 paraffin selectivity Catalytic Testing 42 43 41 • H2 activated on metallic Cu • Activation: 85 replenished protons • H-BEA: 500°C in dry air 27 27 • Lewis acidic cationic Cu within the 58 29 • Cu/BEA: 500°C in dry air (ox-Cu/BEA) followed by 300°C in H2 (red- MSIntensity ( zeolite pores promoted hydrogen Cu/BEA) Cu/BEA 44 Cu/BEA 45 Cu/BEA 60 abstraction from alkane products, + D2 + D2 + D2 D- creating a pathway for • H-BEA mixed with Cu/SiO : oxidized in air and reduced in H as above 41 43 2 2 reincorporation into the chain D- • Fixed-bed, 0.6gcat, 200°C, 1atm, 24h TOS, XDME < 15% growth pool 46 43 47 46 • Reaction feed: 39 40 90 28 28 57 61 29 63 • No H2: 7.1sccm DME, 4.9sccm Ar More information: Schaidle, J.A.; Ruddy, D.A.; Habas, S.E.; Pan, M.; Zhang, 20 30 40 50 20 30 40 50 60 70 20 30 40 50 60 70 80 90 100 • H2: 7.1sccm DME, 7.1sccm H2, 1sccm Ar m/z G.; Miller, J.T.; Hensley, J.E. ACS Catal 2015, 5, 1794

Information contained in this poster is subject to a government license. 24th NAM of the Catalysis Society Pittsburgh, PA June 14-19, 2015 NREL/PO-5100-64751