Catalytic Activities of Bimetallic Catalysts for Low Temperature Water Gas Shift Reaction

Vincenzo Palma, Marco Martino*, Domenico Pisano, Paolo Ciambelli

Department of Industrial Engineering, University of Salerno, Via Giovanni Paolo II, 13 - 84084 Fisciano (SA), Italy, mamartino@unisa.it

A comparative study on the activity of bimetallic Pt/M/CeZrO₄(M=Na, Mo, Sn, Cu, Zn) catalysts is presented. The catalysts were prepared by wet impregnation method, by loading separately the two metals, with a Pt/M molar ration of 1. All the catalysts were fully characterized by several physical-chemical analytical techniques (ED-XRF, XRD, B.E.T., H₂-TPR) and the performance compared with the corresponding monometallic Pt/CeZrO₄ derivate. The 1Pt/1Sn/CeZrO₄ and 1Pt/1Na/CeZrO₄ catalysts showed a similar CO conversion to that of the monometallic Pt-based catalyst in the temperature range of 473-675 K, however a reduced occurrence ofmethanation, allows a higher production of hydrogen with the bimetallic systems.

1. Introduction

The water gas shift is one of the most studied reaction in the field of industrial chemistry, the thermodynamic profile and the necessity to reach the maximum CO conversion make it necessary to operate with a wasteful two-stage process. The HT Shift is generally performed on Fe/Cr-based catalysts in the temperature range 873-623 K; at this stage the most of CO is converted, although the thermodynamic limitations prevent to reach the complete conversion, and a favourable kinetic can be exploit. The LT Shift is performed on Cu/Zn-based catalysts in the temperature range 453-653 K, at this stage the conversion of CO is thermodynamically favoured and it’s possible to reach few ppm in concentration. These two catalytic systems are inexpensive and ensure the effectiveness of the process, however, they present several drawbacks; in addition to the burden of working with a two-stage system, the Cu/Zn catalysts are not suitable for fuel cell applications. Several alternative have been proposed to replace the conventional catalytic systems, actually used for industrial application; good results were obtained with noble metals such as gold (Yang et al., 2013), platinum (Miao et al., 2016), rhodium (Lombardo et al., 2016), palladium (Shan et al., 2015) and rhenium (Azzam et al., 2013). These catalytic systems are highly active in a large temperature range even at low concentration and are compatible with fuel processor systems, thus they seem to be the best choice for a process intensification (Palma et al., 2016). Comparative studies, on the activity of ceria/zirconia supported noble metals catalysts, showed the highest activity of platinum in the temperature range 473-593 K (Radhakrishnan et al., 2006), moreover in our previous studies the role of the support was investigated (Palma et al., 2014, 2015), highlighting the excellent performance of the ceria/zirconia system, however, these systems suffer of serious deactivation, by sintering and coking (Zalc et al., 2002), in real process conditions and a high methane selectivity (Torrente-Murciano et al., 2015). A strategy to overcome these problems consists in the adding of metallic low cost stabilizers; for example the Pt sintering was inhibited by the sodium promotion on Pt-Na/TiO₂ catalyst (Zhu et al., 2011) obtained by co-impregation of TiO₂ by the two relative salts, otherwise the coke formation was inhibited also by the Ce_0.78Sn_0.2Pt_0.02O_2−δ catalyst, obtained by solution combustion method (Gupta et al., 2010). A molybdenum promotion effect was recently reported, for Pt-carbon supported catalysts, increasing the turnover frequency up to a factor of 4000 (Sener et al., 2016); while a comparative study of platinum catalysts supported on M-doped cerium oxide (M = Ca, La, Mg, Zn, Zr, Yb, Y, Gd) have shown a promotion effect on the activity, depending on the nature of the dopant, varying in the order of Yb>Gd>Zr>Mg>La>CeO₂ (undoped) >Ca>Y>Zn (Panagiotopoulou et al., 2007). Recently Kugai et al., (2015) have demonstrated the stabilizing effect of copper in Pt-Cu/CeO₂ catalysts in oxygen-enhanced water gas shift reaction(OWGS), the bimetallic system showed a higher activity than the monometallic Pt-based catalyst and a higher stability for about 70h in OWGS conditions. In this paper we report a comparative study on the activity of bimetallic 1% Pt/ 1% M/CeZrO₄(M=Na, Mo, Sn, Cu, Zn) catalysts for water gas shift reaction.

2. Experimental

2.1 Catalysts preparation

The CeO₂/ZrO₂ (ActalysR922, 57.4% wt of ceria) support was provided by Rhodia, the Platinum(IV) chloride (99.9%-Pt) was provided by Strem Chemicals, the Tin(II) chloride dehydrate (≥99.99% trace metals basis), Sodium nitrate trihydrate (puriss. p.a., 99-104 %) and Zinc nitrate hexahydrate (98 %) were provided by Sigma-Aldrich. The 1%Pt/CeZrO₄ catalyst was prepared by wet impregnation of the support with a water solution of PtCl₄ (the amount of the salt corresponded to a loading of 1 % of metallic platinum with respect to the ceria-zirconia support), the suspension was vigorously stirred and heated until complete evaporation of the solvent occurred, the resulting suspension was dried at 393 K for three hours and calcined at 873 K for three hours. The bimetallic catalysts were prepared by loading firstly the “etero-metal (M)” and subsequently the platinum, both metals were loaded with the same method of the monometallic catalyst, with an intermediate step of calcination, in all cases the amounts of the salts used were calculated with the intent of obtain a ratio Pt/M = 1 and a total loading of 2% (Pt+M). All the samples were compacted, crushed and sieved to achieve a granulometric distribution 180-355 µm, prior to the H₂-TPR experiments and catalytic activity tests.

2.2 Catalysts characterization

The real metal loading of the catalysts was evaluated by means of Energy Dispersive X-ray Fluorescence spectroscopy with an ARL^TM QUANT’X EDRF Spectrometer; the samples was compressed to form a tablet and analysed by the Fundamental Parameters method for determining the elemental composition. The B.E.T. surface area measurements were performed with a Sorptometer 1,040 K (Constech International); the samples were weighted and inserted in the apparatus tubes, degassed in vacuum for 30 min at 423 K and the specific surface area evaluated with adsorption-desorption isotherms of nitrogen at 77 K. The crystal phases and the crystallite size were evaluated by X-ray Powder Diffraction with a D8 Advance (Brucker); the samples were submitted to a Cu Kα radiation (35kV, 40 mA) in the 2θrange20-80° (Stp = 737; Stp size = 0.0814; t/Stp = 0.5 s). The crystallite sizes were calculated by using the Scherrer equation. The H 2 -TPRexperiments were performed in the same reactor used for the catalytic tests, and immediately before the tests, with a reducing stream of 1,000 Ncc/min of 5% H₂ in N₂, with a heating rate of 10 °C/min, in the temperature range of 300-723 K. The hydrogen uptake was calculated by integration of the curve, expressing the concentration change of hydrogen vs. temperature.

2.3 Catalytic activity tests

All the catalysts were reduced prior the activity tests, as indicated in the previous paragraph. The WGS tests were performed at atmospheric pressure, with a reacting mixture of 8% CO, 30% H₂O, nitrogen balance and a flow rate of 1,000 Ncm³ /min, at a gas hourly space velocity (GHSV) of 10,000 h^-1, in the temperature range 423-673 K, starting form the highest temperature and by applying a descending ramp of 2 K/min. The tests were performed in a fixed bed tubular stainless steel reactor with an internal diameter of 22 mm, 6 cm 3 of each catalysts (180-355 µm) were diluted with quartz (500 - 700 µm) to reach 12 cm 3 of total volume, to minimize the pressure drop and the thermal effects. Products streams were continuously monitored by an ABB system, the nondispersive infrared photometer Uras 14 for CO, CO₂ and CH₄ and the thermal conductivity analyser Caldos 17 for hydrogen. The GHSV was calculated as the ratio between the volumetric flow rate and the volume of the catalyst. The CO conversion was calculated as the ratio between the difference of the moles, between input and output, and the input moles. The hydrogen selectivity was calculated as the ratio between the hydrogen moles and the converted carbon monoxide moles.

Table 1 : Metal loading, specific surface area and crystallite size

3. Results and discussion

3.1 Characterization

In table 1 the effective metal loading, the specific surface area and the crystallite sizes for all the catalysts, are reported. The metal loading, measured on the calcined catalysts, was in a good agreement with the theoretical, while the deviation was attributed to the experimental error. The B.E.T. specific surface area measurements showed that the addition of the two mwtal involved a significant reduction of the s.s.a., only in the 1 Pt/1Na/CeZrO4 catalyst; a similar effect was recently reported for Ni/Na/CeO₂ catalyst (Ang et al.,), and it was attributed to the penetration of sodium into the pores of the support, resulting in the occluding and consequent decrease of the surface area; for the other catalysts this phenomenon was not present, so no significant change in the surface of the support was assumed.

Figure 1: Comparison of the X-ray diffractograms of 1Pt/CeZrO₄ , 1Pt/1Sn/CeZrO₄ and 1Pt/1Na/CeZrO₄ catalysts

The XRD diffractograms of all the catalysts showed the co-existence of two phases for the ceria/zirconia solid solution, the typical cubic fluorite type phase and tetragonal t’ phase, attributed to the broadening of reflections at high-angle profiles (Liotta et al., 2003), in no case reflection of the metals loaded were detected, probably

because of the high dispersion (Figure 1). The average crystallite sizes of CeZrO₄ were calculated from the broadening of the (111) reflection, centred at 2θ 29.36, using the Scherrer equation; the results showed a slight increase in dimensions probably due to the incorporation of the metal into the lattice of the support, resulting in the increasing of the grain dimensions.

Table 2: Experimental H₂ uptake and temperature of the main reduction peak

In Table 2 the experimental H₂ uptake and the temperature of the main reduction peak in TPR experiments, for all catalysts, are reported. The hydrogen uptake was, for all catalysts, higher than the theoretical, due to the occurrence of the “spillover effect” (Norman et al., 2001), moreover, except for the 1Pt / 1Cu/ CeZrO₄ , all the other bimetallic catalysts presented a hydrogen uptake higher than the monometallic 1Pt / CeZrO₄ .Concerning the H₂ -TPR profiles, all catalysts showed a broad peak in the 445 - 550 K temperature range, attributed to the overlapping of the active metal reduction and the metal-induced reduction of the surface of the support, moreover a second broad peak in the 600 - 700 K temperature range was attributed to the bulk reduction of the support. The presence of a single peak at low temperatures, corroborated the hypothesis of the formation of a strong interaction between the two metals species, for all catalysts, inducing a shift on the platinum reduction peak. The presence of copper caused the higher shift towards lower temperatures, but a similar trend was also found with tin and zinc, on the other hand the presence of sodium, and molybdenum shifted the Pt-reduction peak towards higher temperatures.

4. Activity tests

The catalytic activity tests, performed in the conditions above mentioned, showed a strong dependence of the CO conversion from the type of the “etero-metal” used (Figure 2). The 1Pt / 1Cu/ CeZrO₄ catalyst provided the lowest conversion in all the temperature range. Much better performance were obtained with all other bimetallic catalysts, for which, above 560 K, the CO conversions were higher than 85%. However, if we consider the lowest temperatures, only 1Pt / 1Na/ CeZrO₄ and 1Pt / 1Sn / CeZrO₄ showed an activity similar to that of the monometallic 1Pt/CeZrO₄ , approaching the equilibrium above 500 K. For temperature below 500 K the CO conversion obtained with the monometallic Pt-based catalyst was higher than those obtained with the bimetallic systems, in any case.

Figure 2: Comparison of the CO conversion of all the catalysts

More interesting was the trend of the hydrogen selectivity (Figure 3); the monometallic 1Pt /CeZrO₄ catalyst showed a lower H₂ -selectivity with respect to all the bimetallic systems, in all the temperature range studied, and reached the same H₂-selectivity of the bimetallic Pt/Na and Pt/Sn systems, only at 463 K. Between the bimetallic catalysts, the best performance, in terms of hydrogen selectivity were obtained with the 1Pt /1Sn / CeZrO₄ in all the temperature range studied. The reduced hydrogen selectivity of the monometallic Pt-based catalysts was justified by the occurrence of the methanation reaction therefore, despite it showed the same CO conversion of the bimetallic Pt/Sn and Pt/Na catalytic systems, part of the carbon monoxide and hydrogen product, was consumed to produce methane. A large difference of H₂ -selectivity was obtained from above 468 K, with a maximum in the loss of hydrogen at 673 K, in the order of 15%. The stoichiometry on the methanation reaction provides three moles of hydrogen per CO consumed, so a small amount of methane correspond to a large loss of hydrogen. The activity data were cross-checked with the characterization results, however no simple correlation was found, probably due to the different character of the “etero-metal” used.

Figure 3: Comparison of the hydrogen selectivity between the monometallic 1Pt/CeZrO₄ and the bimetallic 1Pt/1Sn/CeZrO₄ and 1Pt/1Na/CeZrO₄ catalysts

Concerning a possible correlation between the activity of the catalysts and the H₂-TPRresults, the differences in the activities of 1Pt / 1Sn / CeZrO₄ and 1Pt / 1Na / CeZrO₄ catalysts, at low temperatures, is probably due both to the lower reducibility and to the lower specific surface area of the latter. Further study seems to be necessary to clarify the role of each “etero-metal” used.

Figure 4: CO conversion trend of 1P t/1Sn / CeZrO₄ catalyst at 513 K

The effect of the GHSV on the activity of the 1Pt/1Sn/CeZrO₄ catalyst, at 513 K, was also investigated (Figure 4); the obtained results showed a slight reduction of the activity at 25,000 h -1 , however the hydrogen selectivity was not affected by the higher space velocity, highlighting the good performance of this catalyst.

5. Conclusion

In this paper we have presented a comparative study on the activity of bimetallic 1%Pt/1%M/CeZrO₄ catalysts (M=Na,Mo,Sn,Cu,Zn), for CO water gas shift reaction. The best performance, in terms of conversion and H₂-selectivity were obtained with the 1Pt /1Sn / CeZrO₄ catalyst, in the temperature range 480-673 K, while at lower temperatures the monometallic 1Pt / CeZrO₄ catalyst showed the best performance. These results highlighted the possibility to enhance the performance of the Pt-based catalysts by using a second metal for alloy formation. Further study are in progress to evaluate the coke formation and sintering resistance of the bimetallic systems.

References

Ang M.L., Oemar U., Saw E.T., Mo L., Kathiraser Y., Chia B.H., Kawi S., 2014, Highly Active Ni/xNa/CeO₂ Catalyst for the Water−Gas Shift Reaction: Effect of Sodium on Methane Suppression, ACS Catalysis, 4, 3237−3248.

Azzam K.G., BabichI V., Seshan K., Mojet B.L.,Lefferts L., 2013. Stable and Efficient Pt–Re/TiO₂ catalysts for Water-Gas-Shift: On the Effect of Rhenium, Chem Cat Chem, 5(2), 557-564.

Gupta A., Hegde M.S., 2010, Ce_0.78 Sn_0.2 Pt_0.02O_2−δ : A new non-deactivating catalyst for hydrogen production via water–gas shift reaction, Applied Catalysis B: Environmental, 99, 279-288.

Kugai J., Fox E.B., Song C., 2015, Kinetic characteristics of oxygen-enhanced water gas shift on CeO₂-supported Pt–Cu and Pd–Cu bimetallic catalysts, Applied Catalysis A: General, 497, 31–41.

Lombardo E.A., Cornaglia C., Múnera J., 2016, Development of an active, selective and durable water-gas shift catalyst for use in membrane reactors, Catalysis Today, 259, 165–176.

Liotta L.F., Macaluso A., Longo A., Pantaleo G., Martorana A., Deganello G., 2003, Effects of redox

treatments on the structural composition of aceria–zirconia oxide for application in the three-way catalysis, Applied Catalysis A: General, 240, 295–307.

Miao D., Goldbach A., Xu H., 2016, Platinum/Apatite Water-Gas Shift Catalysts, ACS Catalysis, 6(2), 775-783.

Palma V., Pisano D., Martino M., Ricca A., Ciambelli P., 2014, Comparative Studies of Low Temperature Water Gas Shift Reaction over Platinum Based Catalysts, Chemical Engineering Transactions, 39, 31-36.

Palma V., Pisano D., Martino M., Ricca A., Ciambelli P., 2015, Catalysts for the Intensification of the Water Gas Shift Process, Springer Proceedings in Energy, 2 nd International Congress on Energy Efficiency and Energy Related Materials (ENEFM2014), Oludeniz, Fethiye/Mugla, (Turkey), 479-484.

Palma V., Ricca A., Meloni E., Martino M., MIccio M., Ciambelli P., 2016, Experimental and numerical investigations on structured catalysts for methane steam reforming intensification, Journal of Cleaner Production, 111, 217-230.

Norman A., Perrichon V., Bensaddik A., LemauxS., Bitter H., Koningsberger D., 2001, Study of the reducibility of Pt or Pd on ceria–zirconia catalystsby XANESmeasured at the Ce L III edge andmagnetic susceptibility measurements, Topics in Catalysis, 16/17 (1–4), 363-368.

PanagiotopoulouP., Papavasiliou J., Avgouropoulos G., IoannidesT., Kondarides D.I., 2007, Water–gas shift activity of doped Pt/CeO₂ catalysts, Chemical Engineering Journal, 134, 16–22.

RadhakrishnanR.,. WilliganR.R, DardasZ., VanderspurtT.H., 2006, Water Gas Shift Activity of Noble MetalsSupported on Ceria-Zirconia Oxides, AIChE Journal, 52 (5), 1888-1894.

SenerC., WesleyT.S., Alba-RubioA.C., KumbhalkarM.D., HakimS.H., RibeiroF.H., MillerJ.T., DumesicJ.A., 2016, PtMo Bimetallic Catalysts Synthesized by Controlled SurfaceReactions for Water Gas Shift, ACS Catalysis, 6, 1334−1344.

ShanJ.-J.,Nguyen L.,Zhang S., TaoF.-F., 2015, Water–Gas Shift on Pd/α-MnO₂ and Pt/α-MnO₂, Catalysis Letters, 145, 1571–1580.

Torrente-Murciano L., Garcia-Garcia F.R., 2015, Effect of nanostructured support on the WGSR activity of Pt/CeO2 catalysts, Catalysis Communications, 71, 1–6.

X. Zhu, M. Shen, L.L. Lobban, R.G. Mallinson, 2011, Structural effects of Na promotion for high water gas shift activity on Pt–Na/TiO₂, Journal of Catalysis, 278, 123-132.

Yang M., Allard L.F., Fleytzani-Stephanopoulos M., 2013, Atomically Dispersed Au-(OH)x Species Bound on Titania Catalyze the Low-Temperature Water-Gas Shift Reaction, Journal of the American Chemical Society, 135, 3768-3771.

Zalc J.M., Sokolovskii V., Löffler D.G., 2002, Are Noble Metal-Based Water–Gas Shift CatalystsPractical for Automotive Fuel Processing?, Journal of Catalysis, 206, 169-171.