Non-oxidative methane aromatization in a catalytic membrane reactor Olivier RIVAL, Bernard GRANDJEAN, Abdelhamid SAYARI, Faïçal LARACHI Department of Chemical Engineering and CERPIC Université Laval, Ste-Foy, Québec and Christophe GUY Department of Chemical Engineering, Ecole Polytechnique de Montreal Environmentally Friendly Gas Technologies 2 nd Canadian-Korean joint WORKSHOP Feb. 28 to Mar. 2, 2000 Montreal / Boucherville / Varennes / Bells Corners Canada 1 Production of H 2 : growing interests Demand in ecofriendly fuels and processes Fuel-cell technology development F ideal energy carrier 2
Major source of H 2 Natural gas - Abundance of methane reserves - Canada is the 3 rd world largest producer the 2 nd world exporter F Economic significance of CH 4 conversion into H 2 3 H 2 production from CH 4 Main industrial processes : Gas steam reforming Catalytic methane decomposition Methane pyrolysis Disadvantages : energy intensive and costly low H 2 purity greenhouse gas emissions F non-oxidative process 4
Non-oxidative methane conversion into H 2 and hydrocarbons x CH 4 y H 2 + z C n H m C n H m = alkanes, alcenes (C 2 to C 8 ) 1- or 2- step processes : Garnier et al.,1997, Smith et al.,1995, Cheikhi et al.,1994 C n H m = aromatics (benzene, toluene,...) 1- or 2-step processes: Iglesia et al., 1999, Shu et al., 1999, Weckhuysen et al.,1998 F low yield, complexity 5 Aromatization: Equilibrium-limited conversion 6 CH 4 9 H 2 + C 6 H 6 CH4 CONVERSION % mol 35 30 25 20 15 10 5 0 373 473 573 673 773 873 973 1073 1173 TEMPERATURE (K) F Limitation: 11.3% at 973K 6
Shift of the thermodynamic equilibrium using permselective membrane k 1 6 CH 4 9 H 2 + C 6 H 6 k 2 Permselective membrane H 2 withdrawal F lower temperature, higher yield separation in situ 7 Objective of this study Investigation of methane aromatization in a catalytic membrane reactor 8
Experimental set-up Membrane reactor REB Research (with Palladium-Coated Tantalum and/or Niobium membrane http://www.rebresearch.com) GC permselective Membrane CH 4, Ar, H 2, C 6 H 6 (reaction side) CH 4, Ar (feed in) H 2 (permeation side) under vacuum (10-2 Pa) T Catalyst fixed bed Ru~Mo-HZSM5 prepared by wet impregnation Furnace 9 Test on H 2 permeation through REB- Research Membrane H 2 permeation rate (mlstp min -1 ) 100 90 Sievert s type permeation equation J = J o e (-Ep / RT) (P H2,r -P H2,p ) 80 70 60 50 40 30 20 10 524 K 599 K 686 K 870 K J o = 10-5 m.s -1.Pa - E p = 18 kj.mol -1 0 0 50 100 150 200 250 300 350 10 P reaction - P permeation (Pa )
Test on catalytic activity without hydrogen permeation Formation of molybden carbides which are the actives species for aromatization 1.75 1) Catalyst is 100% selective in benzene 2) Catalyst activity: complex pattern CH4 CONVERSION TO C6H6 [CH4]o=100% FCH4 o =12,2mL.mIn -1 T=600C) CH4 to C6H6 ( %mol) 1.5 1.25 1 0.75 0.25 0 quasi steady state reactivity 0 60 120 180 240 300 360 420 480 540 600 Time (min) methane adsorption and decomposition and catalyst reduction and slight occurrence of catalyst deactivation 11 Residence time effect on methane conversion (without hydrogen permeation) CH 4 conversion into C 6 H 6 ( % mol) 3 2,5 2 1,5 1 0,5 Experimental Conditions No permeation P reaction side = 101 kpa Temperature = 873K CH4 feed conc. = 100% GHSV = 400 h-1 GHSV = 710 h-1 GHSV = 1100 h-1 0 0 60 120 180 240 300 360 420 480 540 600 TIME (min) F still under kinetics control 12
Effect of hydrogen permeation on conversion CH 4 conversion into C 6 H 6 ( % mol) 3.0 2.8 2.5 2.3 2.0 1.8 1.5 1.3 1.0 0.8 0.3 0.0 1) hydrogen permeation increases conversion No permeation With permeation after 360min on stream 0 60 120 180 240 300 360 420 480 Time on stream (min) 2)..but it promotes coke-laydown catalyst deactivation T = 873K [CH4]o = 100% GHSV = 710h-1 13 Results on CH 4 conversion: H 2 withdrawal (permeation) Temperature K CH 4 feed dilution % mol G.H.S. V. h -1 CH 4 conversion b,c % mol Thermodynamic CH 4 conversion % mol No 773 100 350 0.20 1.8 Yes a 773 100 350 0.36 1.8 No 823 100 380 0.76 3.2 No 873 100 400 2.5 5.2 Yes a 873 100 400 5.8 5.2 No 873 100 710 1.6 5.2 Yes a 873 100 710 2.8 5.2 No 873 100 1100 1.0 5.2 No 873 43 800 2.2 7.2 No 873 24 1500 2.8 8.9 a = P TOTAL permeation = 0.2 Pa b = calculated on a benzene basis c without permeation, quasi steady state conversion is reported with permeation, initial conversion is reported 14
Permeation effect on methane conversion CH 4 conversion into C 6 H 6 ( % mol) 6,0 5,0 4,0 3,0 2,0 1,0 Experimental Conditions P permeation side = 0,2 Pa P reaction side = 101 kpa G.H.S.V. = 350-400 h -1 CH4 feed conc. = 100% x 1.8 Thermodynamic equilibrium Permeation No permeation 0,36 0,2 0,0 723 773 823 873 923 TEMPERATURE (K) 5,8 2,5 x 2.3 15 Literature comparison Best yield in C 6 H 6 at 873K: 2.6 %mol (G.H.S.V.= 50h -1, Pt-HZSM-5, Marczewski et al., 1994) Permeation effect : Conversion of propane into aromatics is increased by a factor 2 in a membrane reactor (Uemiya et al., 1990) 16
Membrane Reactor Modelling: Schematic of the membrane reactor (plug-flow) Reaction side H F i,r Fi,r + F i,r Permeation side F i,p F i,p + F i,p Membrane H F i,r F i,r + F i,r Shell Simplified reaction rate r CH4 = k 1 P CH4 R.T α - k 2 P C6 H 6 R.T β P H2,r R.T γ 17 Membrane Reactor Modelling: CH 4 balance: Permeation rate: d F H2,p = J da H 2 balance: d X CH4 r CH4 = d W F CH4 J = J o e (-Ep / RT) (P H2,r -P H2,p ) P H2,r = P t,r 3 /2 X CH4 F CH4 - F H2,p (1+ 2 / 3 ) X CH4 F CH4 + F o Ar - F H 2,p Fitting of kinetic parameters using Powell algorithm: α = 0.41 β = 0.41 γ = 0.31 k 1 = 6.10-5 k 2 =4.10-4 18
Modelling results (at 873 K) EXPERIMENTAL or equilibrium CH 4 conversion ( % mol) 0,10 0,09 0,08 0,07 0,06 0,05 0,04 0,03 0,02 0,01 Experimental Conditions: P permeation side = 0,2 and 101 kpa P reaction side = 101 kpa CH4 feed conc. = 24-100% Temperature = 873 K G.H.S.V. = 350-1500 h -1 Active reactor lentgh = 0.035 m E: Equilibrium conversion simulation with an hypothetical 1m-length reactor 0,00 0 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08 0,09 0,1 E E E PREDICTED CH 4 conversion ( % mol) 19 Conclusions Ru~Mo-HZSM5 catalyst was prepared Methane aromatization with 100% benzene selectivity was observed REB-Research membrane reactor has been tested: Hydrogen permeation: J improves the conversion rate (by a factor 2) at 600 o C, conversion of 5.8 % vs 2.5% without permeation L contributes to catalyst deactivation Membrane reactor model has been proposed and validated Future work Enhancement of catalyst performances: Æ X CH4, coking, Æ stability adding small amounts of CO, CO 2 (Ichikawa et al.,1999) 20
Boudouart: CO [C] active + CO 2 carbide species formation (promoting effect) [C] actif + x / 2 H 2 CH x «Decokefaction» and regeneration (stabilizing effect) [C] inert + CO 2 CO (coke) ex: {+1.8% CO during 100h} 2x 4% = 8% conv. of CH 4 with a selectivity in C 6 H 6 alm. cste (# 67%) (Ichikawa et al., 1999) 21