Operando XRD-DRIFTS study

Renewable energies Eco-friendly production Innovative transport Eco-efficient processes Sustainable resources Operando XRD-DRIFTS study during Fischer-Tropsch synthesis over a Co/Al 2 O 3 catalyst Innovative tool Advanced methodology New relevant insights Julien Scalbert, Christèle Legens, Isabelle Clémençon, Fabrice Diehl, Dominique Decottignies, Sylvie Maury

Summary Introduction Fischer-Tropsch (FT) synthesis Operando methodology Development of an innovative XRD-DRIFTS prototype Operando investigation during FT synthesis over a model Co/alumina catalyst under representative FT conditions Monitoring the crystalline phases of the catalyst Monitoring the surface of the catalyst Relationships between structure/surface/catalytic properties? Conclusions 2

Summary Introduction Fischer-Tropsch (FT) synthesis Operando methodology 3

Introduction Fischer-Tropsch synthesis: allows hydrocarbons production from synthesis gas n CO + 2(n+1) H 2 C n H 2n+2 + n H 2 O Syngas can be obtained from natural gas, coal,biomass. FT fuels of very high purity No sulfur, no heavy metals, etc. Cobalt-based catalysts Favor high molecular weight n-alkanes Good activity/price ratio, slow deactivation. Reaction and deactivation mechanisms still unclear 4

Introduction The power of operando Operando methodology: powerful tool allowing catalyst characterization during the reaction Monitoring the catalyst evolution with time on stream Establishing relationships between intrinsic catalyst properties and catalytic properties (activity, selectivity, stability) Better understanding reaction mechanisms and deactivation phenomena Improving/upgrading processes by adjusting reaction conditions Rational design of optimal catalysts. 5

Summary Introduction Fischer-Tropsch (FT) synthesis Operando methodology Development of an innovative XRD-DRIFTS prototype 6

XRD-DRIFTS prototype X-ray detector To X-ray detector From IR source Be To IR detector IR source IR detector X-ray source Be Inlet gas Outlet gas From X-ray source GC Catalyst Furnace Fritted disk 7

XRD-DRIFTS prototype Technical challenges Coupling XRD and DRIFTS in a unique apparatus The operando reaction cell must allow working under FT conditions: Activation temperature: 500 C (under H 2 ) Reaction temperature: 190 240 C Maximal pressure: 18 bar (15-30 bar in industrial processes) H 2 /CO ratio: around 2.0 and 2.2 Fixed-bed-like reactor with the help of simulation hydrodynamic tools (see poster: Improvement of operando DRIFTS-XRD cells by chemical engineering tools) On-line GC analysis of products Heating the lines to prevent condensation 8

Summary Introduction Fischer-Tropsch (FT) synthesis Operando methodology Development of an innovative XRD-DRIFTS prototype Operando investigation during FT synthesis over a model Co/alumina catalyst under representative FT conditions 9

FT Operando XRD-DRIFTS study Experimental Catalyst: Co/Alumina 13wt% Co Particle size: 10 nm Activation: H 2 flow, 500 C, 16 h Reaction: 220 C 6 bar H 2 /CO = 2 10

Conversion (%) Selectivity (%) FT Operando XRD-DRIFTS study Catalytic properties 14 12 S C5+ 100 90 80 10 CO Conversion 70 8 6 Period I Increasing activity Period II Fast deactivation Period III Stabilization 60 50 40 4 30 2 S CH4 20 10 11 0 0 0 1 2 3 4 5 6 7 8 Time on stream (days) TOF ~ 13 mmol CO mol Cosurf -1 s -1 Similar to classical fixed-bed reactors J.P. den Breejen et al., JACS 131 (2009) 7197.

Co (fcc) Co (fcc) CoO CoO FT Operando XRD-DRIFTS study Operando XRD patterns t 0 2 days 5 days 8 days Enhanced reduction of residuel CoO into Co (fcc) No visible sintering Appearance of crystalline defects No Co 2 C or other crystalline phase detectable γ-al 2 O 3 γ-al 2 O 3 γ-al 2 O 3 12 d (Å) 2.1 2.0 1.9 1.8 1.7 1.6 1.5 1.4

Yield (%) Reduction rate (%) FT Operando XRD-DRIFTS study Does the increase in Co 0 content induce an enhanced activity? 13 12 10 8 6 4 Yield C5+ Reduction rate Maybe NO! Both stable 20 2 Period I Period II Period III 10 Increasing activity Fast deactivation Stabilization 0 0 0 1 2 3 4 5 6 7 8 Time on stream (days) 100 90 80 70 60 50 40 30 Why it does not? new Co sites not active for FT? not enough to compensate opposite deactivation phenomena?

1645 Log(1/R) =C-H -CH 2 - O-C-H FT Operando XRD-DRIFTS study Operando DRIFT spectra 0.2 a.u. -CH 3 C=C Carbonlys 6 days 2 days 4 h 1 h 14 3500 3000 2500 2000 1500 1000 Wavenumbers (cm -1 )

2058 2045 Log(1/R) FT Operando XRD-DRIFTS study Operando DRIFT spectra 0.2 a.u. 1 h 4 h 2 days 6 days 1940 1850 15 2200 2000 1800 1600 1400 Wavenumbers (cm -1 )

lin. bridg.? lin., bridg.?? Yield (%) Area (a.u.) FT Operando XRD-DRIFTS study Might a relationship be drawn between carbonyls and activity? 12 50 10 Yield C5+ 40 Less carbonyls AND more activity! 8 6 Linear carbonyls 30 4 2 Period I Increasing activity Period II Fast deactivation Bridged carbonyls Period III Stabilization 20 10 What does it mean? 16 0 0 1 2 3 4 5 6 7 8 Time on stream (days) 0

FT Operando XRD-DRIFTS study Less carbonyls but enhanced activity! Less carbonyls less sites favoring CO adsorption Enhanced activity more sites favoring FT reaction Cobalt reconstruction: sites previously active for CO adsorption became active for CO dissociation and FT reaction. Carbonyls not intermediates in FT reaction, but spectator/poison species! FT mechanism likely involving CO dissociation! 17

Yield (%) Area (a.u.) FT Operando XRD-DRIFTS study Relationship between adsorbed C x H y O z species and activity? 12 10 8 6 Yield C5+ Period I Increasing activity Period II Fast deactivation Oxygenated species Period III Stabilization 120 100 80 60 Oxygenated and unsaturated species formation parallel to catalyst deactivation. 4 2 Unsaturated species 40 20 Likely one of the main origins of deactivation. 18 0 0 0 1 2 3 4 5 6 7 8 Time on stream (days)

Summary Introduction Fischer-Tropsch (FT) synthesis Operando methodology Development of an innovative XRD-DRIFTS prototype Operando investigation during FT synthesis over a model Co/alumina catalyst under representative FT conditions Monitoring the crystalline phases of the catalyst Monitoring the surface of the catalyst Relationships between structure/surface/catalytic properties? Conclusions 19

Conclusions Powerful XRD-DRIFTS prototype designed to investigate operando Fischer-Tropsch synthesis Over a Co/alumina catalyst: XRD: enhanced reduction, no visible sintering DRIFTS: carbonyls, ads. hydrocarbons & oxygenated species Evidence for cobalt reconstruction Support towards mechanism involving CO dissociation Oxygenated species as a main cause of deactivation Prospects: study of crystalline structure effect, cobalt particle size, promoters, reaction conditions... 20

Acknowledgments IFP Energies nouvelles Christèle Legens, Isabelle Clémençon Fabrice Diehl, Dominique Decottignies, Sylvie Maury, Antoine Fécant Thank you for your kind attention! Contact: julien.scalbert@ifpen.fr 21

Renewable energies Eco-friendly production Innovative transport Eco-efficient processes Sustainable resources www.ifpenergiesnouvelles.com