Jointly published by Akadémiai Kiadó, Budapest and Kluwer Academic Publishers, Dordrecht
React.Kinet.Catal.Lett. Vol. 76, No. 2, 235-242 (2002)
RKCL3969 THE EFFECT OF HYDRATION/DEHYDRATION ON THE AC ELECTRICAL CONDUCTIVITY OF H4PVMo11O40 AND SOME OF ITS CESIUM SALTS USED AS CATALYSTS Monica Caldararua, Cristian Hornoiua, Georgeta Postolea, Mariana Carataa, Alice Luminita Petrea, Viorel Sascab, Niculae I. Ionescua and Nils I. Jaegerc a
Institute of Physical Chemistry of the Romanian Academy, Splaiul Independentei 202, 77208-Bucharest, Romania b Romanian Academy, Timisoara Branch, B-dul Mihai Viteazul 24, 1900-Timisoara, Romania c Bremen University, Institute of Applied and Physical Chemistry, P.O. Box 330440, D-28334 Bremen, Germany Received September 10, 2001 In revised form March 19, 2002 Accepted April 2, 2002
Abstract Changes in AC electrical conductance G of vanadium substituted 12molybdophosphoric acid and of its (mono)-acid and neutral cesium salts were studied. Depending on the environment, the electrical conductivity is a combination of protonic and electronic conduction, the protonic one dominating at low temperature. Keywords: 1-Vanado-11-molybdophosphoric acid, cesium salts, AC electrical conductivity
INTRODUCTION The heteropolyacids with Keggin structure and their salts are known as bifunctional catalysts due to their strong Brönsted acidity combined with redox properties which can be adjusted by a proper choice of the nature and amount of the constituent elements. Being sensitive to surrounding conditions like relative humidity and temperature [1-3], these compounds have also applications as solid electrolytes in sensors or fuel cells. 0133-1736/2002/US$ 12.00. © Akadémiai Kiadó, Budapest. All rights reserved.
236
CALDARARU et al.: CESIUM SALTS
Electrical properties and acidity of heteropolycompounds depend upon the presence of water in variable amounts. Some heteropolyacids show high protonic conductivity and adsorption of water enhances (in some conditions) the electrical conductivity [4]. For a vanadium substituted molybdophosphoric acid the occurrence of the strongest acid sites is correlated with the amount of water retained by the bulk of the acid phase [5]. The response of electrical conductivity, measured in situ, to the presence of ambient moisture was used previously to follow surface hydration/dehydration of H3PMo12O40•29H2O [6]. In the present paper we report about the hydration/dehydration of H4PVMo11O40 and of two Cs salts followed by using the same technique EXPERIMENTAL H4PVMo11O40•14H2O (HPVM) sample was prepared by following the Tsigdinos method [7,8]. The cesium salts of HPVM were prepared from aqueous solution of the acid with cesium nitrate added in stoichiometric amounts. Cs3HPVMo11O40•7H2O (CsHPVM) and Cs4PVMo11O40•9H2O (CsPVM) samples were dried at 100°C and calcined at 200°C for 2 h in air. The anhydrous HPVM and the cesium salts were stable up to 380 and 400°C, respectively [8]. The installation for in situ AC measurement is composed of needle valves, mass flow controllers (MKS), drying and deoxo units as well. The specially designed conductivity cell [9] consists of two coaxial tantalum cylindrical electrodes fitted into a Pyrex glass tube and supported on a frit, the annular space of which was filled with 2 cm3 of catalyst grains (the fraction between 0.2-0.5 mm). The cell is connected to a high precision RLC bridge TESLA BM 484 and to a quadrupole mass spectrometer (Leybold, PGA 100). AC electrical conductivity of the samples was measured in situ at 1592 Hz and at atmospheric pressure between 20 and 300ºC. The linear heating (2ºC/min) was performed by using a differential temperature programmer Chinoin LP 482. The temperature of the catalyst bed was monitored with a third thermocouple located in the center of the cell. The variation of the conductance G was studied by successive heatingcooling cycles performed on the same sample in gas flow; each run consisted of a heating cooling series preceded by flushing for 30 min with the corresponding gas at room temperature. Cycles in dry argon (DAr) and humid argon (HAr) were performed in the following sequence: DAr-1⇒ DAr-2⇒ HAr-1⇒ HAr-2⇒ DAr-3
CALDARARU et al.: CESIUM SALTS
237
The argon used was research grade 99.999 purity. Humid argon (approx. 2.5 mg of H2O/L gas) was obtained by flushing the gas through a saturator at 25ºC. Surface area measurements (BET) were performed with an automatic surface area analyzer model 4200 (Beta Scientific Corp.) using N2 (purity 99.999) adsorption at - 196ºC, after flushing the samples with a He-N2 mixture at room temperature followed by heating at 200ºC for 2 hours in helium flow.
4
Fig. 1. Variation electrical conductance G of HPVM sample during successive flushing with dry (DAr) and humid argon (HAr)
RESULTS AND DISCUSSION BET surface areas of HPVM, CsHPVM and CsPVM are presented in Table 1.
238
CALDARARU et al.: CESIUM SALTS
Table 1 BET surface areas of HPVM, CsHPVM and CsPVM Sample
Surface area (m2 .g-1)
HPVM CsHPVM CsPVM
15.2 123.9 113.3
These values are in good agreement with the reported ones for heteropolycompounds. Following the classification provided by Mizuno et al. [2] HPVM belongs to the group A (low surface area, up to 15m2 g-1) and CsxH4-xPVMo11O40 with x=3 and 4 to the group B (surface area between 50 and 200 m2g-1). As is shown in Fig. 1, on heating in dry argon (DAr-1 and DAr-2), the conductance of HPVM starts increasing around 30oC, reaches a maximum between 90-100oC, and decreases on further heating between 100 and 200oC. In both cases a new sharp increase is in evidence up to 300oC. The same behavior was observed in DAr-3 run, performed after humid argon (HAr) experiments. A different picture can be observed in humid argon. In this case the initial conductance value is much higher due to water adsorption. The increase of the conductance with temperature for the two successive experiments begins also at 30ºC and the maximum is centered between 60 and 80ºC. Above this temperature, in spite of the permanent presence of water in the flowing gas, the conductance G decreases dramatically on heating between 100 and 200ºC. Beyond 200°C, G increases with the same slope and slightly higher G values (in comparison with the experiments in dry argon), which can be taken as an indication for a similar conduction mechanism in both cases. In solid state, HPVM is composed of heteropolyanions, protons and water (crystallization water and constitution water). The heteropolyacids are usually obtained with a large amount of crystallization water, e.g., in the case of HPVM with up to 29 water molecules [10]. Most of these water species (diffusing from the bulk or adsorbed from ambient at low temperature) become increasingly mobile on heating between 20-95oC and are lost by desorption on heating above (even in presence of humidity in the feed). These two processes are accompanied by the corresponding increase/decrease of conductance and this effect can be used for evaluation of the relative level of hydration/dehydration of the surface.
CALDARARU et al.: CESIUM SALTS
239
,
-
Fig. 2. Variation of the in situ measured electrical conductance G of the CsHPVM sample (a) and CsPVM sample (b), during successive flushing with dry and humid argon
The height of the low temperature maximum (LTM) for CsHPVM (Fig. 2a) in dry atmosphere (centered between 70 and 80ºC) was smaller then that for HPVM. The G-temperature plot presented between 100 and 200ºC another small peak and above 200ºC only a very small increasing trend. In humid atmosphere, due to water adsorption, LTM centered between 50 and 70ºC was much higher then the corresponding peak in dry ambient but lower than its equivalent for HPVM; starting with 100ºC, the plot showed a continuous increasing trend. The second cycle in humid argon reproduced a rather similar pattern, but at lower high temperature (HT) values. By repeating the experiment in dry inert (DAr-3 run) one obtains basically the same behavior as in previous dry argon runs, but with slightly higher HT values. This suggests that the continuous HT increase of G in HAr run must be due to the presence of humidity in the feed.
240
CALDARARU et al.: CESIUM SALTS
A completely different picture was in evidence for CsPVM (Fig. 2b). First of all, the extent of G variation is by far much higher with respect to HPVM and CsHPVM. No maximum of G-temperature plot was observed in LT range in dry atmosphere. In humid feed there is a maximum centered between 5060ºC, the height of which decreases with the number of cycles. Between 100 and 200ºC the change of conductance is quite linear, the values increasing above 200ºC. The final G values are smaller in humid argon with respect to those in dry argon. All curves have similar shapes above 100ºC. It should be noted that the decrease of HT values on progressive wetting of CsPVM sample suggests the incipient surface overlayer transition/restructuring to CsHPVM structure. The activation energies Ea presented in Table 2 were evaluated from linear sections of the lnG vs. 1/T plots. From Table 2 it can be observed that in the LT range there are only small differences in activation energies. Larger differences appear in the high temperature (HT) range. For CsHPVM these differences appear from one run to the other. It is interesting to note that for CsPVM practically no differences in activation energies in dry or humid atmosphere were found. Based on electrical conductivity data, the adsorption and desorption of water was reversible for all the samples studied.
CALDARARU et al.: CESIUM SALTS
241
By definition, the conductivity σ of a compound having protonic and electronic carriers is given by: σ = σp + σe [12], where σp represents the protonic conductivity contribution and σe the electronic one. The maximum observed in G-temperature plots at low temperature in dry ambient can be explained by contribution of proton conductivity (“vehicle mechanism”) [11]. The favored conductivity path would involve protons as charge carriers transported by water molecules, facilitated by molecular water adsorption, through hydrogen bonds. This should be related to the presence of the acidic hydrogen in the molecule, since in the case of neutral CsPVM this peak was not observed in dry ambient. Between 100-200oC, σp decreases on increasing the temperature because the number of vehicles/carriers diminishes by water desorption. At higher temperature, hopping of the proton between OH groups must occur. At the same time on heating, σe increases by electron activation from lattice defects, which must be important in CsPVM [3, 14]. The ratio between σp and σe determines the shape of G-temperature plots. Our in situ conductivity data suggest that the compounds studied show mixed (protonic and electronic) conductivity. At low temperature (up to 100ºC) in presence of water molecules (either crystallization water and/or water adsorbed in presence of humidity) the conduction is predominantly protonic. The loss of water on heating corresponds to the decrease of the number of “vehicle” molecules facilitating the proton movements and the loss of some mobile protons leading to decrease of contribution of ionic conductivity. The increasing trend for conductivity above 200°C must be due to the change of the conduction mechanism from predominantly protonic to a mixed (protonic and electronic) one. The role of mobile protons (particularly in dry ambient) is increasingly smaller and for CsPVM can be neglected. Thus, it can be concluded that between 200-330°C CO and isobutyraldehyde oxidation on HPVM [13] occur by mechanisms involving electron transfer.
Acknowledgements. We thank the Deutsche (AZ436RUM113/13) for financial support.
Forschungsgemeinschaft
242
CALDARARU et al.: CESIUM SALTS
REFERENCES 1. 2. 3. 4. 5. 6.
7. 8. 9. 10. 11. 12. 13.
14.
M. Misono: Catal. Rev. Sci. Eng., 29, 269 (1987). N. Mizuno, M. Misono: Chem. Rev., 98, 199 (1998). S. Berndt, D. Herein, F. Zemlin, E. Beckmann, G. Weinberg, J. Schütze, Y. Mestl, R. Schlögl: Ber. Bunsenges. Phys. Chem., 102, 763 (1998). N. Azuma, R. Ohtsuka, Y. Morioka, H. Kosugi, J. Kobayashi: J. Mater. Chem., 1, 989 (1991). E.M. Serwicka, E. Broklawik, K. Bruckman, J. Haber: Catal. Lett., 2, 351 (1989). M. Caldararu, A.L. Petre, M. Gruia, N.I. Ionescu, N.I. Jaeger: Proc. 9th Int. Symp. Heterogeneous Catalysis, Varna (L. Petrov, Ch. Bonev, G. Kadinov, Eds.), p. 283. M. Drimov Academic Publishing House, Sofia 2000. G.A. Tsigdinos, C.J. Hallada: Inorg. Chem., 7, 437 (1968). V. Sasca, M. Stefanescu, A. Popa: J. Therm. Anal. Cal., 56, 569 (1999). M. Caldararu, D. Sprinceana, V.T. Popa, N.I. Ionescu: Sensors Actuators, B30, 35 (1996). V. Sasca, A. Popa, M. Stefanescu: Chem. Bull. Politehnica Univ. (Timisoara) 43, 275 (1998). P. Columban, A. Novak: Proton Conductors, p. 38 (P. Colomban, ed) Cambridge University Press 1992. A.B. Walters, B.-K. Na, C.-C. Liu, M.A. Vannice: J. Mol. Catal. A: Gen., 162, 287 (2000). V. Sasca, N.I. Jaeger, A. Popa, D. Martin, H. Wellmann, N. Doca: Proc. 9th Int. Symp. Heterogeneous Catalysis, Varna (L. Petrov, Ch. Bonev, G. Kadinov, Eds.), p. 543. M. Drimov Academic Publishing House, Sofia 2000. G. Mestl, T. Ilkenhans, D Spielbauer, M. Dieterle, O. Timpe, J. Kröhnhert, F. Jentoft, H. Knözinger, R. Schlögl: Appl. Catal. A, 13, 210 (2001).