Part A: Operando FT-IR Studies of heterogeneous catalytic reactions: pitfalls and benefits. Fred Meunier fcm@ircelyon.univ-lyon1.fr Institut de Recherche sur la Catalyse et l Environnement de Lyon Villeurbanne, France EUROKIN meeting, Milan 18/2/2015 1
Operando FT-IR Studies of heterogeneous catalytic reactions: 1. Pitfalls 2. Benefits. Pitfalls: - Temperature control - flow control in IR reaction cells 2
Catalytic reactor and spectroscopic IR cells U-shaped quartz reactor in a tubular furnace Transmission IR cell (ca. 20 mg wafer) Heat loss through IR windows Diffuse reflectance IR cells 3 Harrick (powder, up to 80 mg) Modified Spectra-Tech (powder, up to 60 mg)
4 Measure-thermocouple positioning
CO methanation as a bulk temperature probe IR cells bed cooler than expected 5 5
DRIFTS: bed surface (top) temperature Cooler HOT Cooler HOT Typical analysis depth: < few 100 μm Bed depth: 2-5 mm 6
Surface temperature by optical pyrometry DRIFTS cells bed surface much cooler than expected at High T 7
Supported CoOx for CO hydrogenation in custom-made DRIFTS cell: reduction at 450 C? Cooler Cooler Zone probed: CoOx HOT HOT Zone active catalytically: Co metal 8
Spectra-tech DRIFTS cell hydrodynamics: ill-defined reactor t = 0 s 3 s 6 s 9 s 12 s 15 s Meunier et al., Appl. Catal. A 340 (2008) 196 9
Artistic impression of a post-experiment wafer (NOx storage-reduction) Aabspec Inlet Carbonaceous deposits? Outlet 16 mm 10
Alcohol condensation at 300 C using the same catalyst Product formation rate (micromol/g/s) 4 3 Quartz tubular PFR 2 1 0 SpectraTech DRIFTS Aabspec CX Transmission 0 30 60 90 120 150 Time /min Temperature (mis)control and bad hydrodynamics? 11
This image cannot currently be displayed. This image cannot currently be displayed. Modified DRIFTS cell from Spectra-Tech. CO + O 2 over 2% Pt/CeO 2. 100 CO conversion /% 80 60 40 20 Modified cell Non-modified cell Metallic dome KBr window Catalyst SiC 0 Quartz wool 0 373 473 573 673 Temperature / K Ceramic crucible Teflon tape Metal base Meunier et al., Appl. Catal. A 340 (2008) 196 Gas inlet Gas outlet
Validation of modified DRIFTS cell Meunier et al., Appl. Catal. A 340 (2008) 196 Au-Ce-La-O catalyst Water-gas shift: 2 % CO + 7 % H 2 O J. Catal. 247 (2007) 277 10 3 /T (1/K) Log (rate CO 2 formation, mol/g/s) -5.5-6 -6.5 2.00 2.20 2.40 2.60 (200 C) Quartz plug flow reactor (Medford) Modified DRIFT cell (Belfast) (130 C) Ensure kinetic relevance of the IR cell-based data 13
Validation of modified DRIFTS cell CO hydrogenation (30% CO + 60 % H 2 at 1 bar) on 14 wt.% Co (8 nm) /Al 2 O 3 : TOF = 14 x 10-3 s -1 DeJong et al.., J. AM. CHEM. SOC. 2006, 128, 3956 14
Operando FT-IR Studies of heterogeneous catalytic reactions: pitfalls and benefits. Benefits: - transport in zeolites - CO heat of adsorption - catalyst surface poisoning - adsorbate reactivity 15
Isooctane transport in large and small H-ZSM-5 17 µm 0.25 µm 1 mg powder is enough! H-ZSM-5: pore diameter ca. 0.55 nm kinetic diameter = 0.62 nm 16
Isooctane in H-ZSM-5: effect of mesoporosity Relative ic8 DRIFTS band intensity 1.0 0.8 0.6 0.4 0.2 0.0 17 µm 0.25 µm 0 20 Desorption 40 time of ic8 60( s) ca. 4-fold reduction in the characteristic diffusion path length Meunier et al., Micropor. Mesopor. Mater. 148 (2012) 115 17
CO adsorption on Pt-Sn/Al 2 O 3 pre-reduced at 400 C - Feed: 2%CO/H 2 - T decreased from 325-50 C 2054 175 C 200 C Log (1/R) 0.01 Sn-free Pt 2077 225 C 250 C 275 C 300 C 325 C 325 C : CO coverage = ~0 2045 2120 2100 2080 2060 2040 2020 2000 1980 1960 IR determines Wavenumber the nature (cm -1 ) and coverage of sites 18
CO heat of adsorption on Pt/Al 2 O 3 and Pt-Sn/Al 2 O 3 Moscu et al, Chem. Commun., 2014, 50, 8590-8592. 1 Carbonyl band relative intensity 0.8 0.6 0.4 0.2 0 Pt Pt-Sn 50 150 250 350 Temperature / C Catalyst -Δ ads H (kj/mol) ----------------------------------------------------------------- Pt/Al 2 O 3 E0 =180 E1 =85 Pt-Sn/Al 2 O 3 E0 = 95 E1 =75 1. CO bonding much weaker on Pt-Sn / Pt 2. Effect of surface coverage on ΔH not significant for Pt-Sn 19
Introduction of O2 at 225 C (Red spectrum is under CO/H2/Ar only) Moscu et al, Catal Today, 2015, in press. 2079 2050 2067 71 min 10 min 4 min H2 Atmosphere CO:O2 G.A. Somorjai et al., J. Catalysis, 2014, 312, 17-25 Log (1/R) 1 min Before O 2 addition 0.04 2100 2050 2000 1950 Wavenumbers /cm -1
IR is site and coordination-specific!
Toluene hydrogenation: Effect of CO 2 100% 0.17 % toluene + 17 % H 2 + 1.7 % CO 2 75 C Scalbert et al. PCCP, 14, 2159 (2012). Toluene conversion 80% 60% 40% 20% Pt/Al 2 O 3 KBr Rh/Al 2 O 3 0% 0 200 400 600 800 Time on CO 2 -containing stream /min In presence of CO 2 : Pt: fast and total deactivation Rh: slow and limited deactivation: why? 22
Rh/Al 2 O 3 : difference DRIFTS spectra following CO 2 introduction 0.17 % toluene + 17 % H 2 + 1.7 % CO 2 1637 1598 1370 Log 1/R 0.2 2886 760 min 183 min 23 min 4 min 1 min O C Rh 1956 Rh O C Rh 1785 1394 1432 1496 1328 1229 2741 3749 2926 1654 3500 3000 2500 Wavenumbers /cm -1 2000 1500 Various Rh-CO are formed, but no CO(g) 23
Correlation between toluene conversion and Pt-CO Toluene conversion /% 10 9 8 7 6 5 4 [2 % CO 2 ] [40 % CO 2 ] 0 2000 4000 6000 8000 10000 Pt-carbonyl band area /a.u. Conversion correlates Pt-CO signal whatever [CO 2 ] 24
Rh/Al 2 O 3 (D= 55%): linear carbonyls assignment Taimoor et al., J. Catal. 2011, 278, 153. Ferri et al., PCCP, 2002, 4, 2667. High-wavenumber carbonyls, on dense plans, displaced by toluene. Low-wavenumber carbonyls, on low coordination sites, more stable. Low dispersion samples should be more resistant to deactivation.
Rh/Al 2 O 3 (D=55 %) vs Rh/SiO 2 (D=19 %) Feed: 0.8 % toluene + 57 % H 2 Carbonates, Hydrogenocarbonates, Formates Log (1/R) O C 0.2 O C Rh Rh Rh Rh/Al 2 O 3 Rh/SiO 2 2200 2000 1800 1600 Wavenumbers /cm -1 1400 No carbonyl nor carbonate formation on Rh/SiO 2 when CO 2 is added. Scalbert et al., J. Catal 318 (2014) 61 26
DRIFTS + MS + SSITKA (SSITKA: Steady-State Isotopic Transient Kinetic Analysis) Shannon and Goodwin, Chem. Rev., 90 (1995) 667. Goguet et al., J Phys. Chem. B 108 (2004) 20240 12 CO + H 2 O Mass spectrometer 12 CO 2 13 CO 2 Vent reactor time 13 CO + H 2 O
DRIFTS: Formate exchange over Pt-CeO 2 2% 12 CO + 7% H 2 O 2% 13 CO + 7% H 2 O Relative 12 C-Formates IR signal Time / min 1.0 0 5 10 15 20 1.00 0.8 160 C 160 C 0.6 0.4 0.10 180 C 0.2 180 C 220 C 220 C 0.0 0 5 10 15 20 0.01 Time / min Slope = k
Formate DRIFT signal calibration Na-formate deposition (by IWI) over the CeO 2 support Spectra recorded at 100 C under Ar 0.66 wt.% 0.33 wt.% WGS at 220 C Abs Calibration standards Formate DRIFT signal /a.u. 6 4 2 0 160 C 180 C 220 C 3000 2900 2800 Wavenumber /cm -1 2700 0 0.2 0.4 0.6 0.8 wt.% formate [formates] can be determined accurately
Pt/CeO 2 : rate of CO 2 formation vs. rate of formate decomposition Meunier et al. J. Catal. 252 (2007) 18 Rate (10-6 mol s -1 g -1 ) 12 9 CO 2 formation rate GC analysis of reactor exhaust Formate decomposition rate to CO 2 = k [formate]. 100% 6 3 0 160 C 180 C 220 C Reaction temperature 30
Formate decomposition during CO hydrogenation on 14% Co/Al 2 O 3 30% CO + 60 % H 2 at 1 bar, 220 C Parades-Nunez et al., Catal. Today, 242 (2015) 178 Methanol Two-type of formates: fast and slow
Conclusions Operando FT-IR Studies of heterogeneous catalytic reactions: pitfalls and benefits. - Necessity to compare activity in IR cell and standard reactor - Understand the origin of the differences, if any. (impurities, bed by-pass, temperature gradients) - Improve cell design. - Many relevant information for kinetic modelling can be obtained (using differential conditions): transport, nature(s) and coverage of sites, heat and mode of adsorption, poisoning, adsorbate reactivity. 32
Daniele Tibiletti Julien Scalbert Mickael Rivallan Haoguang Li Alina Moscu Anaelle Paredes-Nunez Davide Lorito 33