Energy and Fuel Azhar 2007

Energy & Fuels 2007, 21, 2481-2488 2481 Investigation of Reforming Catalyst Deactivation by Exposure to Fly Ash from Biomass Gasification in Laboratory Scale Jessica Einvall,† Simone Albertazzi,‡ Christian Hulteberg,§ Azhar Malik,† Francesco Basile,‡ Ann-Charlotte Larsson,† Jan Brandin,§ and Mehri Sanati*,†,| School of Technology and Design, Växjö UniVersity, 351 95 Växjö, Sweden, Dipartimento di Chimica Industriale e dei Materiali, Bologna UniVersity, 40136 Bologna, Italy, Catator AB, Ideon, 223 70 Lund, Sweden, and DiVision of Ergonomics and Aerosol Technology, Lund UniVersity, Box 118, SE-22100 Lund, Sweden ReceiVed December 12, 2006. ReVised Manuscript ReceiVed May 11, 2007 Production of synthesis gas by catalytic reforming of product gas from biomass gasification can lead to catalyst deactivation by the exposure to ash compounds present in the flue gas. The impact of fly ash from biomass gasification on reforming catalysts was studied at the laboratory scale. The investigated catalyst was Pt/Rh based, and it was exposed to generated K2SO4 aerosol particles and to aerosol particles produced from the water-soluble part of biomass fly ash, originating from a commercial biomass combustion plant. The noble metal catalyst was also compared with a commercial Ni-based catalyst, exposed to aerosol particles of the same fashion. To investigate the deactivation by aerosol particles, a flow containing submicrometer-size selected aerosol particles was led through the catalyst bed. The particle size of the poison was measured prior to the catalytic reactor system. Fresh and aerosol particle exposed catalysts were characterized using BET surface area, XRPD (X-ray powder diffraction), and H2 chemisorption. The Pt/Rh catalyst was also investigated for activity in the steam methane reforming reaction. It was found that the method to deposit generated aerosol particles on reforming catalysts could be a useful procedure to investigate the impact of different compounds possibly present in the product gas from the gasifier, acting as potential catalyst poisons. The catalytic deactivation procedure by exposure to aerosol particles is somehow similar to what happens in a real plant, when a catalyst bed is located subsequent to a biomass gasifier or a combustion boiler. Using different environments (oxidizing, reducing, steam present, etc.) in the aerosol generation adds further flexibility to the suggested aerosol deactivation method. It is essential to investigate the deactivating effect at the laboratory scale before a full-scale plant is taken into operation to avoid operational problems. 1. Introduction The objective of having a 5.75% share of biofuels in the transportation sector1 has led to a series of interesting project initiations in the field of renewable energy; one of them is CHRISGAS (Clean Hydrogen-Rich Synthesis Gas, IP-project within FP6). The project focuses on modifying an existing integrated gasification combined-cycle (IGCC) pilot plant situated in Värnamo, Sweden, in order to produce synthesis gas by catalytic upgrading of product gas from the biomass gasifier. A detailed description of the process, with emphasis on the difficulties identified within the proposed catalytic treatment, has already been published.2 The produced synthesis gas is to be used in the production of, e.g., Fischer-Tropsch diesel, dimethyl ether, methanol, or hydrogen.3 * To whom correspondence should be addressed: Phone: +46-470708943. Fax: +46-470-708756. E-mail: [email protected]. † Växjö University. ‡ Bologna University. § Catator AB. | Lund University. (1) Directive 2003/30/EC of the European parliament and of the council, May 8, 2003. (2) Albertazzi, S.; Basile, F.; Brandin, J.; Einvall, J.; Hulteberg, C.; Fornasari, G.; Rosetti, V.; Sanati, M.; Trifirò, F.; Vaccari, A. Catal. Today 2005, 106, 297-300. (3) Tijmensen, M. J. A.; Faaij, A. P. C.; Hamelinck, C. N.; van Hardeveld, M. R. M. Biomass Bioenergy 2002, 23, 129-152. The plant modification of the Värnamo gasifier includes a reforming and water gas shift system. The reforming and water gas shift reactors are however preceded by a hot-gas filter for removal of ashes. This filtration procedure is however not sufficient to totally eliminate impurities present in the flue gas,4 and as the impurities are potential poisons for most reforming and shift catalysts, they can cause deactivation of the catalyst.5 Inorganic particles can also be formed during cooling of the flue gas via condensation, via nucleation,6 and by chemical surface reactions. The steam reforming catalyst in biomass gasification systems will be affected by gaseous sulfur compounds as well as by the emitted inorganic submicrometer compounds present in the flue gas originating from the biomass fuel. The latter kind of compounds have shown a deactivating effect on selective catalytic reduction (SCR) catalysts for abatement of NOx in the flue gas from biomass combustion boilers.7 (4) Strand, M.; Pagels, J.; Szpila, A.; Gudmundsson, A.; Swietlicki, E.; Bohgard, M.; Sanati, M. Energy Fuels 2002, 16, 1499-1506. (5) Forzatti, P.; Lietti, L. Catal. Today 1999, 52, 165-181. (6) Hinds, W. C. Aerosol Technology. Properties, BehaVior, and Measurement of Airborne Particles; John Wiley & Sons, Inc.: New York, 1999; pp 283-292. (7) Moradi, F.; Brandin, J.; Sohrabi, M.; Faghihi, M.; Sanati, M. Appl. Catal., B: EnViron. 2003, 46, 65-76. (8) Rostrup-Nielsen, J. R. Steam reforming catalysts; Teknisk Forlag A/S: Copenhagen, 1975; pp 35, 45-46, 63-76. 10.1021/ef060633k CCC: $37.00 © 2007 American Chemical Society Published on Web 08/22/2007 2482 Energy & Fuels, Vol. 21, No. 5, 2007 EinVall et al. Figure 1. Experimental setup for catalyst exposure to the aerosol particles at 800 °C. Figure 2. Experimental setup for aerosol particle characterization by the SMPS at 150 °C. An early study on reforming catalysts has shown that Ni catalysts are sensitive to sulfur,8 and that sulfur in the flue gas can poison the catalyst, leading to decreased catalyst lifetime. H2S in the gas phase will react with Ni in the reforming catalyst, and NiSO4 and NiS will be formed. Rostrup-Nielsen reported that H2S is dissociatively adsorbed on the surface and found a correlation between H2 and S adsorption on Ni where H and S competitive adsorb on the same sites.8 If the hydrogen content is low in the gas, hydrogen desorbs into the gas phase, leaving NiS as product. The formation of H2S can be favored if the hydrogen concentration is high, and sulfur will then readsorb on the nearest unoccupied nickel surface. RostrupNielsen also reported the difficulty of regenerating sulfur deactivated Ni catalyst in the presence of alkali.8 If a sulfurcontaining catalyst is oxidized, sulfur dioxide is formed, which reacts with alkali and forms sulfates. In reducing conditions sulfur will be reduced into H2S and readsorbed on the catalyst. The regeneration procedure for the Ni catalyst will thus be very slow. The sulfur effect on noble metal reforming catalysts is slightly different compared to the effect on nickel-based catalysts, and it has been reported that Ni catalysts are more sulfur-sensitive than Pt catalysts.9 Noble metal catalysts do not form nonregenerable bulk sulfur; they are however affected by the presence of sulfur, but the effect is reversible.9 The hydrocarbon species to be reacted in the reformer are of two general categories: short-chain alkanes and alkenes (C1-C4) and tar-type compounds (aromatic compounds).2 Rioche et al.10 reported that supported Rh and Pt catalysts were more active for steam reforming of acetic acid, phenol, acetone, and ethanol than other (9) Petersen, E. E.; Bell, T. A. Catalyst deactiVation; Marcel Dekker: New York, 1987; pp 83, 151-152. (10) Rioche, C.; Kulkarni, S.; Meunier, F. C.; Breen, J. P.; Burch, R. Appl. Catal., B: EnViron. 2005, 61, 130-139. catalysts tested in this study. This finding has further been confirmed by Kolb et al.11 in investigating Pt/Rh catalysts for the steam reforming of propane. Rh has shown exceptional ability of aromatic degradation and is therefore a suitable choice as an active phase for this type of gas composition.12,13 The choice of a Pt/Rh catalyst seems to be able to handle both types of hydrocarbons that need to be reacted in this type of reforming system. Earlier studies of particles emitted from biomass combustion in small grate type boilers4,14 showed a fine particle mode of around 100 nm and the main elements in the fine mode to be K, S, and Cl with the main components being K2SO4 and KCl. In the present work, the effect of generated submicrometer aerosol particles of K2SO4 salts and salt derived from fly ash on reforming catalysts was investigated. In the latter case, the fly ash from a commercial biomass combustion plant was dissolved in boiling water, and the catalyst was exposed to the generated particles from the dissolved fly ash solution. This fly ash sample contains a realistic mixture of compounds, which the reforming catalyst is assumed to be, at least to some extent, exposed to in a commercial plant. A new approach for deactivating reforming catalysts in the laboratory scale by aerosol particle deposition is proposed in order to simulate the decay processes, prior to initial operation in full scale. The proposed catalyst deactivation experiment by deposition of aerosol particles at the laboratory scale has previously been verified for deactivation of commercial SCR catalysts in earlier (11) Kolb, G.; Zapf, R.; Hessel, V.; Löwe, H. Appl. Catal., A: Gen. 2004, 277, 155-166. (12) Duprez, D. Appl. Catal., A: Gen. 1992, 82, 111-157. (13) Duprez, D.; Miloudi, A.; Delahay, G.; Maurel, R. J. Catal. 1986, 101, 56-66. (14) Pagels, J.; Strand, M.; Rissler, J.; Szpila, A.; Gudmundsson, A.; Bohgard, M.; Lillieblad, L.; Sanati, M.; Swietlicki, E. J. Aerosol Sci. 2003, 34, 1043-1059. Reforming Catalyst DeactiVation by Exposure to Ash Energy & Fuels, Vol. 21, No. 5, 2007 2483 Figure 3. Schematic view of experimental setup for activity measurement. 2. Experimental Section Figure 4. Number size distribution of generated submicron aerosol particles from K2SO4 and salt leached from fly ash measured by the SMPS. 2.1. Catalyst. A catalyst with a Pt/Rh active phase dispersed on a heat-treated Al/Mg spinel was prepared. A slurry comprised of the sintered material, a binder (Boehmit), and water was dried at 150 °C for 24 h. The dried mixture was calcined for 2 h at 800 °C, and the residue oxide was crushed. The crushed catalyst was fractioned, and the fraction of about 0.1 mm was used for catalyst preparation. The chosen fraction was impregnated with a Pt(NO3)2 solution, dried at 150 °C, and calcined at 800 °C for 2 h. The platinated catalyst was impregnated by RhCl3 and again calcined as previously. The resulting catalyst contained 4.5% noble metal by weight, and had an atomic ratio Pt/Rh equal to 4. The prepared catalyst was compared to a commercial, Ni-MgAl2O4, steam reforming catalyst, having a nominal Ni content of 13 wt %. work.15 A bimetallic catalyst of Pt/Rh has been evaluated as it can provide qualities suitable for steam reforming, and a commercial Ni catalyst was used as comparison. The fresh and aerosol particle exposed catalyst samples were characterized using different techniques, such as BET surface area, XRPD (X-ray powder diffraction), and H2 chemisorption, in order to comprehend the induced catalytic deactivation in biomass gasification system. The effect of the deactivation was also investigated by measurement of the catalyst activity in the steam methane reforming reaction of the Pt/Rh catalyst, before and after K2SO4 aerosol particle exposure. (15) Larsson A.-C.; Einvall, J.; Andersson, A.; Sanati, M. Energy Fuels 2006, 20, 1398-1405. Figure 5. Mass size distribution of generated submicron aerosol particles from K2SO4 and salt leached from fly ash calculated from SMPS measurement using the salt bulk density of potassium sulfate. 2484 Energy & Fuels, Vol. 21, No. 5, 2007 EinVall et al. Figure 6. Calculation of the stability of potassium at aerosol particle generation. (A) Solid phase in N2/H2O environment; (B) potassium in gas phase in N2/H2O environment. 2.2. DeactiVation Species. As K2SO4 is one of the most abundant compounds present in biomass combustion gases, this compound was used to study the effect of emitted biomass associated submicrometer aerosol particles on reforming catalysts. To investigate the influence of other impurities present in fly ash, another case was investigated. The fly ash from a commercial biomass combustion plant was dissolved in boiling water, and the catalyst was exposed to the generated aerosol particles. This fly ash sample contains a realistic mixture of compounds, which the reforming catalyst might be exposed to in a commercial full-scale plant. The collected fly ash from a commercial boiler (circulating fluidized bed, 104 MW) is assumed to contain the most abundant elements emitted from biomass burning. The ash salt solution was prepared by boiling fly ash with deionized water, which was then filtered. The main compounds, oxygen aside, in the ash salt were potassium (38%), sulfur (16%), sodium (5.6%), and calcium (1.9%) analyzed using an inductively coupled plasma atomic emission spectrometer (ICP-AES) and high-resolution inductively coupled plasma mass spectrometry (HR-ICP-MS). One of the major components in the ash salt is K2SO4. The sulfate is stable in an oxidizig environment, i.e., combustion with sulfur present in the gas phase, and it has a relatively high melting point (1067 °C). The stability of K2SO4 was investigated by thermodynamic equilibrium calculations with HSC 5 (manufactured by Outokumpu Research Oy) in N2/H2O (aerosol particle experiment) and in reducing (synthesis gas) environments, respectively. Table 1. Gas Flow during Activity Measurement Experiment GHSV S:C ratio 50 000 h-1 1.7 50 000 h-1 2.5 100 000 h-1 1.7 100 000 h-1 2.5 CH4 H2O N2 total 92.6 Ndm3/h 157.4 Ndm3/h 250 Ndm3/h 500 Ndm3/h 74.6 Ndm3/h 178.6 Ndm3/h 250 Ndm3/h 500 Ndm3/h 185.2 Ndm3/h 314.8 Ndm3/h 500 Ndm3/h 1000 Ndm3/h 149.2 Ndm3/h 357.1 Ndm3/h 500 Ndm3/h 1000 Ndm3/h Reforming Catalyst DeactiVation by Exposure to Ash Energy & Fuels, Vol. 21, No. 5, 2007 2485 Table 4. Deposited Amount of Potassium after Aerosol Particle Exposure Analyzed by AA catalyst samples K concn (wt %) Ni catalyst K2SO4 exposed ash salt exposed Pt/Rh catalyst K2SO4 exposed ash salt exposed 0.0088 0.0023 0.0036 not detected Table 5. H2 Chemisorption of Fresh and Particle Exposed Catalysts Figure 7. Possible mechanism for clogging of a porous catalyst surface at K2SO4 aerosol particle exposure. Table 2. Characteristics of Generated K2SO4 and Ash-Salt Aerosol Particles bulk densitya (g/cm3) concn in reactor (mg/m3N) mean particle mass size (nm) concn in reactor (number/cm3N) mean particle number size (nm) K2SO4 ash salt 2.66 37 231 7.4 × 106 119 2.66 18 187 6.5 × 106 98 a The bulk density of potassium sulfate was used in the mass concentration calculation. Table 3. BET Surface Areas of Fresh and Aerosol Particle Exposed Samplesa catalyst samples Ni catalyst fresh reference K2SO4 exposed ash salt exposed Pt/Rh catalyst fresh reference K2SO4 exposed ash salt exposed surf. area (m2/g) 28 10 11 10 16 12 12 18 a The reference samples have been exposed to the same conditions as the aerosol particle exposed samples but with no salt present. 2.3. DeactiVation by Exposure of the Reforming Catalysts to Generated Potassium and Ash Salt Particles in Laboratory-Scale Experiments. The generation and deposition of aerosol particles was carried out in a setup shown in Figure 1. The aerosol particles were generated by a pneumatic atomizer (Palas GmbH, AGK-2000). The atomizer was connected to pressurized nitrogen with a gauge pressure of 1 bar, and the flow rate from the atomizer was 4 Ndm3/min. The atomizer reservoir contained the salt solution. The concentration of the K2SO4 solution used for aerosol generation was 50 g of salt/dm3 of deionized water, and the mass of salt in the ash solution was 16.5 g/dm3 (originating from 665 g of fly ash). The flow of nitrogen and generated aerosol particles passed through an inertial impactor separating large liquid droplets before entering an electrically resistance heated oven. A catalyst bed consisting of 5.5 g of approximately 0.1 mm diameter catalyst pellets was placed in the center of a tubular reactor, which was heated to 800 °C at atmospheric pressure. A continuous flow of 0.5 Ndm3/min containing the generated aerosol particles was forced through the catalyst bed for 2 h. The relative humidity at deposition was less than 1% measured with a TH-Calc 8720 (TSI Inc.). The generated aerosol particles were physically characterized according to their electrical mobility using a scanning mobility particle sizer (SMPS; TSI Inc.) consisting of a differential mobility analyzer (DMA; TSI Inc. Model 3081) and a condensation particle counter (CPC; TSI Inc. Model 3010), shown in Figure 2. The SMPS measures the particle number size distribution, and from the bulk density, the mass size distribution can be calculated. The number size distribution for a size range of 20-478 nm of the aerosol particle flow was measured, and the flow led through the catalyst samples Ni catalyst fresh K2SO4 exposed ash salt exposed Pt/Rh catalyst fresh K2SO4 exposed ash salt exposed dispersion (%) metallic surf. area (m2/gsample) 5.1 2.4 3.0 3.07 1.44 1.81 3.6 0.5 0.5 0.48 0.06 0.07 instrument was 0.3 Ndm3/min. Particle characterization was carried out at 150 °C in order to avoid losses of particles during sampling. Particle sampling was carried out isokinetically using a glasssampling probe. To decrease the temperature, the particle concentration, and the relative humidity prior to analysis, the sample was diluted with particle-free, dry pressurized air (20 °C) to give a ratio of 1/27. The relative humidity after dilution was measured with a TH-Calc 8720 (TSI Inc.) to 4%. 2.4. Chemical Characterization of Fresh and Aerosol Particle Exposed Catalyst Samples. A Perkin-Elmer 4100 atomic absorption spectrometer (AA) was used to determine the potassium content of the catalyst. For this purpose, the catalyst was partially dissolved in boiling HNO3 and water, and the sample was diluted with water before the measurement. Adsorption of N2 at the temperature of liquid nitrogen (the BET method) was measured on a Sorpty 1750 instrument, to give the surface area of fresh and particle exposed catalyst samples. A fresh catalyst was also exposed to the steam and nitrogen without the presence of poison, in order to investigate the stability of the catalyst. In the remainder of this paper this catalyst is designated “reference”. The metallic surface area, dispersion, and metal particle size were measured by H2 chemisorption (Micrometrics, ASAP 2020). Before adsorption of H2 the Ni catalyst was preheated at 750 °C in He and then in H2. The Pt/Rh catalyst was preheated at 350 °C in He and then in H2. The measurement of H2 adsorption for both catalysts was made at 35 °C. The metal crystallite size was determined by XRPD by a Philips Analytical X-Pert Instrument with Cu KR radiation at 35 kV and 35 mA using the peak broadening in conjunction with the DebyeScherrer equation. 2.5. ActiVity Measurements of the Pt/Rh Catalyst. The activity was measured for fresh and K2SO4 exposed Pt/Rh catalyst in the steam reforming reactions of methane, eqs 1 and 2. CH4 + H2O T CO + 3H2 (1) CH4 + 2H2O T CO2 + 4H2 (2) The main parts of the catalytic system consist of an oven, mass flow controllers, and gas mixing equipment (Figure 3). All flows are presented at normal conditions: 0 °C and 1.01 × 105 Pa. The reaction gases are mixed and preheated in a vessel and then led into a tubular reactor, where the reaction takes place. The gas composition at different gas hourly space velocities (GHSVs) is reported in Table 1. Ten milliliters of catalyst pellets was used in the experiments, and the temperature was varied between 600 and 800 °C. The conversion has been calculated as X ) (CO + CO2)out/CH4,in. 2486 Energy & Fuels, Vol. 21, No. 5, 2007 EinVall et al. Figure 8. XRPD of fresh Pt/Rh catalysts and catalyst exposed to K2SO4 and ash-salt aerosol particles. Table 6. Average Metal Crystallite Size of Fresh and Exposed Catalyst Samplesa catalyst samples Ni catalyst fresh K2SO4 exposed ash salt exposed Pt/Rh catalyst fresh K2SO4 exposed ash salt exposed relative metal crystallite size (exposed/fresh) metal crystallite size (nm) 1 1.05 1.00 61 61 64 a The results for the Ni catalyst are reported as relative values compared to the fresh sample. 3. Results and Discussion 3.1. Salt Exposed Samples at Laboratory Scale. The number concentration of the generated particles was measured by the SMPS, and the mass concentration was calculated. The number and mass size distributions are presented in Figures 4 and 5. All obtained results are presented at normal conditions: 0 °C and 1.01 × 105 Pa. The mean number particle size was 0.1 µm and the calculated mass concentration was 37 mg/m3 for the K2SO4 aerosol particles. The mass concentration of the aerosol particles generated from the ash salt was lower due to the lower salt concentration in the solution. The mean number particle size was also 0.1 µm, and the calculated mass concentration was approximately 18 mg/m3. The bulk density for the ash salt was estimated to be 2.66 g/cm3 (the bulk density of K2SO4) in the calculation of mass concentration. The results are presented in Table 2. In the thermodynamic calculations an amount of 1 kmol of K2SO4 was calculated to be in contact with a total gas flow of 1 × 106 kmol. The sulfur content (SO2 respectively H2S) in the gas phase in the reducing environment was calculated to 100 ppm, while in the N2/H2O calculation no sulfur in the gas phase was present. In the N2/H2O environment, potassium sulfate is the only stable solid phase up to 750 °C, and above this temperature the solid K2SO4 is in equilibrium with a melt KOH phase and gas phase containing SO2 and KOH(g). The results are presented in Figure 6. In the reducing environment, as during biomass gasification, the calculations were made on synthesis gas: H2O (33 vol %), CO2 (20 vol %), CO (24 vol %), and H2 (23 vol %). There is no solid phase present, only a melt KOH phase which is in equilibrium with H2S and KOH(g) in the gas phase. From the equilibrium calculations it can be concluded that, in the model deactivation experiments with K2SO4 aerosol particles in N2/H2O atmosphere, potassium sulfate will be present and deposited on the catalyst mainly as a solid phase, which differs from the synthesis gas environment where the equilibrium calculation reveals a melted phase KOH(l), or at least a state where the aerosol particles move toward formation of a melted KOH(l) phase. A proposed mechanism is that solid K2SO4 in equilibrium with liquid KOH deposits on the catalyst surface, while KOH(g) and H2S(g) are released; see Figure 7. Reforming Catalyst DeactiVation by Exposure to Ash Energy & Fuels, Vol. 21, No. 5, 2007 2487 Figure 9. Conversion as a function of S:C ratio, 1.7 ([) and 2.5 (9), for fresh Pt/Rh catalyst at GHSV 50 000. Figure 10. Conversion of fresh (9) and K2SO4 ([) particle exposed Pt/Rh catalyst at GHSV 100 000, S:C ) 1.7. The effect of aerosol particles on reforming catalysts is of interest, and thus K2SO4 serves as an aerosol particle model compound in the total evaluation of the impact of aerosol particles on deactivation of reforming catalysts. 3.2. BET Measurement. The surface area measured by BET for the fresh, aerosol particle exposed, and reference catalyst samples are shown in Table 3. The BET surface area decreased for both the Pt/Rh catalyst and the Ni catalyst because of the conditions during exposure. The decrease in surface area was 64% for the Ni catalyst and 25% for the Pt/Rh catalyst. The Pt/Rh is less affected by the environment probably because of the calcination procedure when manufactured. For both catalysts the surface area after K2SO4 aerosol particle exposure decreases to the same magnitude as the reference sample. For the Pt/Rh catalyst, the surface area was slightly increased after ash salt particle exposure, which can be due to formation of needle-shaped crystals. On the other hand, no potassium was detected by AA for the Pt/Rh catalyst exposed to ash salt. The increase in surface area may be contributed from other compounds present in the ash salt. It has previously been reported that the surface area can increase after aerosol particle exposure, at least for solid particles.7 For the Ni catalyst the surface area decreased to a magnitude similar to the reference sample after exposure to ash salt. A decrease in surface area after steam and hydrogen exposure has been reported for a MgAl2O4-based Ni catalyst due to sintering,8 but in the current measurement, no sintering of the Ni or Pt crystallites was observed due to the aerosol particle exposure according to the XRPD measurement. 2488 Energy & Fuels, Vol. 21, No. 5, 2007 3.3. Chemical Characterization of the Samples. The bulk concentration of potassium measured by AA is presented in Table 4. The concentration of accumulated potassium in the exposed Ni catalysts was 0.0088 wt % after K2SO4 aerosol particle exposure and 0.0023 wt % after ash-salt aerosol particle exposure. The accumulation of potassium in the Pt/Rh catalyst was 0.0036 wt % potassium after K2SO4 aerosol particle exposure, while in the case of ash-salt aerosol particle exposure, no potassium deposition was detected. The ratio of the deposited potassium concentration to the active metal concentration is a little higher for the Ni catalyst (7 × 10-4) compared to Pt/Rh catalyst (5 × 10-4). The measured deposition of potassium in the catalysts exposed to K2SO4 aerosol particles compared to the amount forced through the catalyst bed is 45% and 18% in the Ni and Pt/Rh catalyst, respectively. The deposited amount of potassium from the ash-salt aerosol particles is 30% in the Ni catalyst, while no deposition of potassium in the Pt/Rh catalyst was detected. The lower mass deposited of the ash salt in both catalysts may be attributed to the lower mass concentration of the salt solution used for aerosol particle generation. The reason for the lower deposition in the Pt/Rh catalyst could be a slightly larger size of the Pt/Rh catalyst particles. The deposition of aerosol particles is affected by the catalyst particle size, and the difference in deposition could also be attributed to different catalyst particle shapes, particle densities, and surface areas. The mass deposition is lower than compared to previously reported aerosol deposition experiments by an electrostatic field on the SCR catalyst,15 which were in compliance with a fullscale exposure. No data for a full-scale steam reforming plant to upgrade gas after biomass gasification are available, but the low weight percent of deposited material indicates that a longer deposition time is required in order to achieve results close to conditions in a full-scale reforming plant. 3.4. H2 Chemisorption. The metallic surface area and dispersion were measured with H2 chemisorption for fresh and aerosol particle exposed catalysts. The results from the H2 chemisorption are presented in Table 5. The metallic surface area measured by H2 chemisorption for both the Ni and Pt/Rh catalysts exposed to K2SO4 and fly ash salts decreased compared to fresh catalyst. The decrease can be attributed to poisoning by particle coverage of the catalyst surface. Rostrup-Nielsen has previously reported a decrease in Ni surface area measured by H2S chemisorption after steam and hydrogen streaming of a MgAl2O4-based Ni catalyst,8 while Pt and Rh catalysts are stable in inert and reducing environments.9 3.5. XRPD of the Catalyst Samples. The XRPD spectra of the Pt/Rh catalysts are presented in Figure 8, the Ni catalyst (commercial) is not presented due to the secrecy agreement with the catalyst supplier. The average metal crystallite size calculated by the Scherrer equation from the XRPD patterns is presented in Table 6. Only relative values are shown for the Ni catalyst. Neither the Pt/Rh nor the Ni catalyst crystallite size was affected by the aerosol particle exposure as measured using XRPD. This proves that the reduction of metallic surface area in Table 5 is due to poisoning by pore blockage by the aerosol particles and not due to active phase crystallite sintering during the experiment. View publication stats EinVall et al. 3.6 ActiVity Measurements of the Pt/Rh Catalyst. The activity measurement of the fresh Pt/Rh at different steam to carbon (S:C) ratios indicated an increased conversion at increased S:C ratio. The results at different S:C ratios are presented in Figure 9. The deactivation with aerosol particles of K2SO4 is demonstrated in Figure 10. The results show that exposing the catalyst to the aerosol particles decreases the activity of the catalyst. The conversion of the catalyst decreased from around 8% to 5% at 700 °C after particle exposure, thus showing that the particle exposure had a negative effect on the catalyst activity. The decrease in metallic surface area measured by H2 chemisorption after K2SO4 aerosol particle exposure is 88%, while the decrease in activity is about 50%. This might be explained by an oversizing of the initial noble metal concentration and thus the initial decrease in metal surface area is not proportional to the loss of activity. 4. Conclusion Catalyst deactivation after exposure to aerosol particles on reforming catalysts has been investigated. The AA analysis showed potassium accumulation caused by particle deposition of 0.0088 wt % in the Ni catalyst after exposure to generated K2SO4 aerosol particles, thus verifying the deposition of aerosol particles. The deposition of potassium on the Ni catalyst is 45% of the exposed amount, which confirms that the method is applicable for aerosol particle deposition in a catalyst bed. The conversion of methane in the steam reforming reaction decreased for the Pt/Rh catalyst exposed to K2SO4 aerosol particles compared to fresh catalyst, showing that aerosol particle exposure had a negative effect on the catalyst activity. This decrease is probably due to coverage of the catalyst active sites by the deposited aerosol particles. This is further emphasized by the decrease in metal surface area measured by H2 chemisorption after the aerosol particle deposition despite the fact that the active phase crystallite size is unaltered as measured by XRPD. The deactivation method can be applied to further evaluate the impact of inorganic salts and carbon compounds expected to be present in the product gas fed to a catalytic steam reformer unit following a biomass gasifier, as well as for the investigation of other parameters that influence process conditions. Further investigations can lead to a better understanding of the mechanism of deactivation related to steam reforming catalysts when exposed to a combination of soot and ash compounds from biomass gasification. This developed decay study of catalytic processes has been shown, in conformity with previous studies,7,15,16 to assist in predicting the influence of emitted aerosol particles on catalytic deactivation in a full-scale experiment in thermochemical conversion of biomass. Acknowledgment. Financial support through the EC 6th Framework Program (CHRISGAS Project Contract No. SES6CT2004-502587), the Swedish Energy Agency, and the Swedish Research Council is gratefully acknowledged. EF060633K (16) Larsson, A.-C.; Einvall, J.; Sanati, M. Aerosol Sci. Technol. 2007, 41, 369-379.