CATALYTIC PARTIAL OXIDATION OF ETHANE WITH SULPHUR IMPURITIES OVER Rh AND Pt CATALYSTS

Transcription

1 CATALYTIC PARTIAL OXIDATION OF ETHANE WITH SULPHUR IMPURITIES OVER AND CATALYSTS S. Cimino*, G. Mancino**, L. Lisi* *Istituto Ricerche sulla Combustione CNR P.le V. Tecchio 8, Napoli Italy **Dipartimento Ingegneria Chimica, dei Materiali e della Produzione Industriale, Università di Napoli Federico II P.le V. Tecchio 8, Napoli Italy Abstract The impact of sulphur impurities (up to 58 ppmv) during the catalytic partial oxidation (CPO) of ethane was investigated on - and - based honeycomb catalysts tested under self-sustained high temperature conditions. Both steady state and transient operation of the CPO reactor were investigated particularly with regards to poisoning/regeneration cycles. A detailed analysis of products distribution in the effluent and a heat balance of the CPO reactor demonstrates that sulphur reversibly adsorbed on selectively inhibits the ethane hydrogenolysis and, to a lower extent, steam reforming reaction. A further, simultaneous adverse effect of S on the kinetics of the reverse water gas shift reaction on catalyst operating at temperatures<75 C can cause an unexpected increase in the H 2 yield above its equilibrium value for low concentrations of the poison. catalyst is less active for those reactions but in turn is more S-tolerant. Introduction The production of syngas (CO and H 2 ) via the catalytic partial oxidation (CPO) of light hydrocarbons is an attractive and feasible alternative to steam reforming reaction in the utilisation of the world s abundant natural gas reserves [1]. The syngas can then be converted to clean fuels such as sulphur-free diesel or gasoline by Fischer-Tropsch synthesis [1-2]. Furthermore, CPO of various hydrocarbons has been proposed as a preliminary conversion stage in low-no x hybrid catalytic burners [2-4]. A number of catalysts have been tested for the CPO of hydrocarbons from methane up to diesel and jet-fuels and - and -based catalysts have shown the highest activity and selectivity to syngas[1,2,4,5]. The adverse impact of sulphur compounds on catalytic performance is well known [5,6], however, studies on the effect of sulphur bearing compounds naturally occurring in the fuel, or added as odorants, during the CPO over precious metal based catalysts have so far remained scarce and mainly limited to methane feed [7,8]. It is largely accepted that in the case of CH 4 CPO over and the gas-phase paths are negligible at atmospheric pressure and the catalytic route dominates [1]. However, following the progressive lower stability of the C-H bond in higher

2 alkanes, gas phase chemistry can play an increasingly important role [9-1], as in the case of CPO of ethane or propane [11,14], when large quantities of ethylene can be formed with based catalysts (but not with ) mainly through homogeneous reactions if proper conditions are met (T>75 C, C/O feed ratio 2, addition of a sacrificial fuel) [9,11-14]. Accordingly we set out to investigate the effect of S addition to the feed during the CPO of ethane for syn-gas production (feed ratio C/O=1) under self sustained high temperature short contact time conditions over γ-alumina supported and catalysts anchored to honeycomb monoliths. Apart from assessing the impact of S poisoning with respect to practical issues for industrial implementation such as catalyst activity and durability and multi-fuel operability of a CPO reactor, an attempt is made to shed light on the complex interaction of hetero-homogeneous reaction paths that become available as soon as the strength of the C-H bond in the hydrocarbon feed is lowered with respect to the case of methane. With respect to this last point, S-poisoning is often reported to be structure sensitive [6] and the resulting inhibition of specific surface active sites can be used to titrate the contribution of some heterogeneous reactions to the overall performance of the CPO reactor. Experimental Commercial honeycomb monoliths with straight and parallel channels of roughly square section (cordierite, 6cpsi by NGK) were cut in the shape of disks of 17 mm diameter and 1 mm long and washcoated with 3% La/ γ-al 2 O 3 (SCFa14-L3 Sasol) by a dip-coating procedure [8]. or (respectively 1 and 2 % by weight of the alumina, in order to obtain similar atomic concentrations of metals) were loaded via repeated incipient wetness impregnations onto washcoated monoliths using an aqueous solution of (NO 3 ) 3 or H 2 Cl 6 (Aldrich). Catalyst conditioning was performed directly under CPO reacting atmosphere. The impact of sulphur addition (2-58 ppmv SO 2 ) on the CPO of ethane to syn-gas was studied under selfsustained high temperature conditions at fixed preheating and H 6 /O 2 feed ratio = 1-2, under both transient and steady state conditions. Air or enriched air (O 2 :N 2 =1:2) was used as oxidant. Conversions and selectivities were calculated from exit dry-gas mol fractions of the main species independently measured with by a continuous analyzer and an on line gas chromatograph equipped FID and SCD detectors, calibrated to measure light hydrocarbons up to C 4 and main S species. In fact O 2 was always completely converted after catalytic light off was achieved. Results and Discussion Steady state and transient experiments with addition/removal of up to 58ppm of SO 2 to the feed demonstrated that S-poisoning is rapid, completely reversible, directly dependant on S-concentration, and leads to a new steady state

3 characterised by a higher surface temperature of the catalyst and a correlated drop in fuel conversion and yield to syn-gas. catalyst, which is more active than with S-free feed (Fig. 1), mainly due to its higher activity for steam reforming (eq. 1) [1,15] and hydrogenolysis of ethane (eq. 2) [15], is also more prone to the adverse impact of sulphur [7], which strongly and preferentially adsorbs on highly dispersed crystallites [7,9]. H 6 + 2H 2 O 2CO + 5H 2 H = +347 kj/mol (1) H 6 + H 2 2CH 4 H = - 65 kj/mol (2) CO + H 2 O CO 2 + H 2 H = - 41 kj/mol (3) A close examination of the variation in the flows of the main products from the based monolith (Fig. 1a-c) indicate that sulphur selectively inhibits ethane hydrogenolysis (eq. 2) forming methane (Fig. 1c) and, to a lower extent, steam reforming of ethane (eq. 3) and ethylene, in good agreement with the reported structure sensitivity of these reactions on supported metals [6]. A third inhibiting effect of sulphur on the kinetics of the reverse water gas shift reaction (eq. 3) was observed only for operation with at temperatures below 75 C (in air at low preheating), and prevented the equilibrium to be reached (Fig. 2), in contrast to what normally observed during CPO over [1]. On the other hand WGS kinetics and approach to equilibrium are not affected by S on -based catalyst (Fig. 2), which in turn operates at higher temperatures than its - based counterpart under identical feed conditions F i mol/h S in feed, ppm H 2 CO F i mol/h S in feed, ppm H 2 O CO 2 F i mol/h S in feed, ppm Figure 1. Effect of SO 2 addition on the steady state flows of products from the - (filled symbols) and - (open symb.) monolith catalysts during the CPO of ethane in air. Feed H 6 =2Sl/h, C/O=1, GHSV= h -1, preheating 235 C. Dashed lines represent equilibrium values (p, H=constant) excluding C s formation. H 6 CH 4 H 4

4 K wgs ppm S-free 3ppm S-free 58ppm temperature, C Figure 2. Comparison between WGS equilibrium constant and values calculated from the measured concentrations of products during ethane CPO over - and monoliths, at varying SO 2 concentration in feed (see legend), as a function of gas temperature at the exit of catalyst. CPO at C/O=1, with air and preheating at 235 C (circles), or enriched air at 35 C (squares). The combined poisoning effect of ethane hydrogenolysis and reverse WGS at low sulphur concentrations during ethane CPO in air with -based catalyst explains the apparently surprising increase in the H 2 yield above its equilibrium value recorded with a simultaneous drop in fuel conversion (Fig 1a). As shown in Figure 3, the temperature increase measured in the -based monolith upon variation of S from 3 up to 58 ppm of S is proportional to the reduction in ethane conversion, which agrees well with the data collected for the CPO of methane over the same catalyst [8]. The adverse impact of sulphur on steam reforming which was previously argued from the variation in the flows of products can be further corroborated by a simple heat balance in the gas phase (eq.(4)). Assuming adiabatic operation and thermal equilibrium between solid and gas in the second part of the reactor [7,8], it is possible to calculate an apparent heat of reaction for ethane or methane from the slope of the lines in Figure 3: T W Cp T H = r (4) F fuel x fuel where Cp is the specific heat of the product gas mixture, W is the total molar flow rate, and F fuel is the hydrocarbon feed molar flow rate. For the case of ethane, with a on La-γ-Al 2 O 3 catalysts operating at roughly 75 C with S-free feed and air as

5 oxidant, it turns out that H r 75 C = +36 kj per mole of H 6, which corresponds well to the heat of the ethane steam reforming reaction at that temperature level. The superimposed and larger poisoning effect of the exothermic ethane hydrogenolysis (eq. 2) that was found for the addition of 2-3 ppm of sulphur (Fig. 1), is the cause of the initial low temperature increase and change in the slope of the curve relevant to ethane feed in Figure 3. It is also evident from Figure 3 that the higher reforming reactivity of ethane in comparison to methane [15] limits the adverse impact of sulphur on syn-gas production and on the associated risk of catalyst overheating during CPO. Tcat, C ppm Ethane 37ppm 2ppm 9ppm 6ppm 3ppm 58ppm 2ppm 37ppm 2ppm 9ppm 3ppm 2ppm Methane S-free a) conversion Figure 3. Increase in catalyst temperature as a function of the variation of fuel conversion measured upon the addition of sulphur to the feed during the CPO of methane or ethane over the same /La-γ-Al 2 O 3 monolith. Feed: methane or ethane at C/O = 1, air as oxidant (O 2 =2 Sl/h), preheating 235 C. Finally, it was found that S-poisoning significantly increases the net production of ethylene (Fig. 1 c) particularly over catalyst, but also over, discovering the role of heterogeneous reactions consuming both the reactant (ethane) and the product (ethylene) of gas phase dehydrogenation chemistry, which occurs simultaneously, supported by the high temperatures in the CPO reactor [9-14]. Thus selective S-poisoning of catalytic sites can be used as a strategy to alter the balance between homogeneous and heterogeneous chemistry in the CPO of highly reactive feeds. Acknowledgements Financial support from MiSE-CNR Accordo di Programma per l Attività di Ricerca di Sistema under project: Biocombustibili is acknowledged. References [1] Horn R, Williams KA, Degenstein NJ, Bitsch-Larsen A, Dalle Nogare D, Tupy SA, Schmidt LD, Methane catalytic partial oxidation on autothermal

6 and foam catalysts: oxidation and reforming zones, transport effects, and approach to thermodynamic equilibrium. J Catal 249: 38 (27). [2] Lyubovsky M, Roychoudhury S, LaPierre R, Catalytic partial oxidation of methane to syngas at elevated pressures, Catal Lett 99: (25). [3] Cimino S, Russo G, Accordini C, Toniato G, Development of a hybrid catalytic gas burner, Combust Sci Tech 182: 38 (21). [4] Cimino S, Allouis C, Pagliara R, Russo G, Effect of catalyst formulation (, ) on the performance of a natural gas hybrid catalytic burner, Catal Today 171: 72 (211). [5] Hulteberg C, Sulphur-tolerant catalysts in small-scale hydrogen production, a review, Int J Hydrog Energ 37: (212). [6] Bartholomew CH, Mechanisms of catalyst deactivation Appl Catal A 212: 17-6 (21). [7] Bitsch-Larsen A, Degenstein NJ, Schmidt LD, Effect of Sulfur in Catalytic Partial Oxidation of Methane over -Ce Coated Foam Monoliths, Appl Catal B 78: (28). [8] Cimino S, Torbati R, Lisi L, Russo G, Sulphur inhibition on the catalytic partial oxidation of methane over -based monolith catalysts, Appl Catal A 36: (29). [9] Michael BC, Nare DN, Schmidt LD, Catalytic partial oxidation of ethane to ethylene and syngas over and coated monoliths: Spatial profiles of temperature and composition, Chem Eng Sci 65: (21). [1] Donazzi A, Livio D, Maestri M, Beretta A, Groppi G, Tronconi E, Forzatti P, Synergy of homogeneous and heterogeneous chemistry probed by in situ spatially resolved measurements of temperature and composition, Angew Chem Int Ed 5: (211). [11] Bodke A, Olschki D, Schmidt LD, Ranzi E, High Selectivities to Ethylene by Partial Oxidation of Ethane, Science 285: (1999). [12] Lange JP, Schoonebeek RJ, Mercera PDL, van Breukelen FW, Oxycracking of hydrocarbons: chemistry, technology and economic potential, Appl Catal A 283: (25). [13] Cimino S, Donsì F, Russo G, Sanfilippo D, Olefins production by catalytic partial oxidation of ethane and propane over /LaMnO 3 catalyst, Catal Today 157: (21). [14] Cimino S, Donsì F, Russo G, Sanfilippo D, Optimization of ethylene production via catalytic partial oxidation of ethane on /LaMnO 3 catalyst, Catal Lett 122: (28). [15] Graf P, Mojet B, van Ommen J, Lefferts L., Comparative study of steam reforming of methane, ethane and ethylene on, and Pd supported on yttrium-stabilized zirconia, Appl Catal A 332: (27).