INTERNATIONAL CONFERENCE ON ADVANCEMENT OF MATERIALS AND NANOTECHNOLOGY: (ICAMN-2007)

Thermal Behavior Study of the MoVTeNb Oxide Catalyst for SelectiveOxidation ProcessR. Idris and S. B. Abd. Hamid Citation: AIP Conf. Proc. 1136, 16 (2009); doi: 10.1063/1.3160123 View online: http://dx.doi.org/10.1063/1.3160123 View Table of Contents: http://proceedings.aip.org/dbt/dbt.jsp?KEY=APCPCS&Volume=1136&Issue=1 Published by the American Institute of Physics. Related ArticlesSynthesis of silicon oxide nanowires and nanotubes with cobalt-palladium or palladium catalysts J. Appl. Phys. 112, 024312 (2012) Sulfur-mediated palladium catalyst immobilized on a GaAs surface J. Appl. Phys. 111, 124908 (2012) Effect of catalyst nanoparticle size on growth direction and morphology of InN nanowires AIP Advances 2, 022150 (2012) Complementary metal-oxide-semiconductor-compatible and self-aligned catalyst formation for carbon nanotubesynthesis and interconnect fabrication J. Appl. Phys. 111, 064310 (2012) Study on transport pathway in oxide nanowire growth by using spacing-controlled regular array Appl. Phys. Lett. 99, 193105 (2011) Additional information on AIP Conf. Proc.Journal Homepage: http://proceedings.aip.org/ Journal Information: http://proceedings.aip.org/about/about_the_proceedings Top downloads: http://proceedings.aip.org/dbt/most_downloaded.jsp?KEY=APCPCS Information for Authors: http://proceedings.aip.org/authors/information_for_authors

Downloaded 18 Oct 2012 to 113.210.100.196. Redistribution subject to AIP license or copyright; see http://proceedings.aip.org/about/rights_permissions

Thermal Behavior Study of the MoVTeNb Oxide Catalyst for Selective Oxidation Process

R.Idris and S.B.Abd.Hamid

# T T • V fV n i

School of Science & technology, Umversiti Malaysia babah, UMb, Locked Beg No.20/3, 88999 Kota Kinabalu, Sabah, Malaysia

Combinatorial Technology and Catalysis Research Centre, Universiti Malaya, UM, 50603, Kuala Lumpur, Malaysia

*E-mail: [email protected]

Abstract. Several parameters involved in preparing the multi metal oxide (ytMO) catalysts (Mo1V03Te0.23Nb0.12Ox) for selective oxidation of propane to acrylic acid (AA) were investigated. These included the proper pre-calcined and calcinations atmosphere effect on the performance of the catalysts. It was found that each metal element plays a critical role to the performance of an effective catalyst and also the calcinations under a non-flow inert atmosphere. The characterization results from XRD, SEM, TG and DSC show the important differences depending on the activation procedures of the MoVTeNb oxide catalyst. The XRD analysis is used to identify the phase inventory of the MoVTeNb oxide catalysts. The structure of orthorhombic Ml, M2, TeMo5O16, V0.95Mo0.97O5 and Mo5O14 phase was investigated. The orthorhombic Ml phase is the most active and selective phase and is responsible for the major of the efficiently of the best catalyst for selective oxidation process. TGA and DTG allow the identification of the number and types, of reactions involving evaporation of small molecules from removal of ligands and water to condensation or drying processes . From all these analyses it was proven that the activation procedures would affect the performance of the MoVTeNb oxide catalyst.

Keywords: Selective oxidation process, Propane, Acrylic Acid, MoVTeNb oxide catalyst. PACS: 81.16.Hc

INTRODUCTION

Today the major industrial process to produce acrylic acid is by gas-phase catalytic oxidation of propylene. The commercial process used worldwide for making acrylic acid via two-step process [1]. This process starts with propylene and goes through acrolein as the intermediate to make acrylic acid. Alternatively, acrylic acid can be produced from a one-step oxidation of propane in gas phase with molecular oxygen to acrylic acid which is this extensive effort to replace the propylene to propane [2].

Recently the use of multi-component oxides catalysts based on molybdenum, vanadium, niobium and tellurium seems to be a major breakthrough leading to promising developments of the selective oxidation of propane to acrylic acid reviewed by Lin [4]. MoVTeNb oxides catalyst has been reported to be able to give yields of acrylic acid above 40% [1]. The performance of Mo-V-Te-Nb-O catalysts for propane oxidation to acrylic acid has been shown to be significantly better than that of any other multi-component metal oxides (MMO) type of catalysts, or any VPO or HPC

CPl 136, Nanoscience andNanotechnology, International Conference on Nanoscience andNanotechnology, (NANO-Sci-Tech 2008), edited by M. Rusop and T. Soga

© 2009 American Institute of Physics 978-0-7354-0673-5/09/$25.00

16

Downloaded 18 Oct 2012 to 113.210.100.196. Redistribution subject to AIP license or copyright; see http://proceedings.aip.org/about/rights_permissions

type of catalysts. However, it is difficult to prepare active Mo-V-Te -Nb-O catalysts that exhibit the claimed performance in either propane ammoxidation to acrylonitrile or propane oxidation to acrylic acid [3]. In this case, not only the composition of the catalyst is of importance, but also the preparation method also greatly gives varying effectiveness of the catalyst of essentially the same composition. For example, the performance of Mo1 V0.3 Te0.23 Nb0.1 On catalysts can vary from very poor (no yield) [1, 5], mediocre (14% acrylic acid yield) [6], to excellent (more then 40% acrylic acid yield) [1, 5]. These performance differences reflect different structure of the catalysts prepared under different preparation cond itions [1, 5 and 7].

Experimental

The catalyst was prepared via amorphous drying method, which consisted of the preparation of slurry containing salts of four metals: ammonium heptamolybdate, ammonium metavanadate, telluric acid and ammonium niobium oxalate. Initially the 25.7 g of ammonium heptamolybdate was dissolved in 420 g of deionized water by heating and stirring, followed by the addition 5.1 g of ammonium metavanadate and 7.7 g telluric acids. This was indicated by a change of color from yellow to orange. The suspension was stirred and heated at 353 K and then leave to cool to room temperature. An aqueous solution of 17.34 mmol niobium oxalate was added to this solution. The obtained slurry was stirred vigorously for 10 minute, and then dried using vacuum rotavapor at 323 K for 2 hours. The catalyst precursor with chemical composition Mo1.0V0.3Te0.23Nb0.12Ox was divided into four parts and precalcined at four different temperature 448 K (MoVTeNbOx I), 498 K (MoVTeNbOx II), 548 K (MoVTeNbOx III), and 598 K (MoVTeNbOx IV) under air and hold for 1 hour. The catalyst precursor was calcined at temperatures 873 K under flow of argon and hold for 2 hours to generate the active catalyst.

Result and Discussion

Temperature is an important factor for catalyst deactivation; at high temperature the catalyst starts to lose catalytic surface area due to crystal growth of the catalytic phase. These results are in agreement with the XRD diffractogram, which shows improvement in the crystalline phase when the temperature increases. Figure 1 shows all the final materials contain the famous five reflections but almost no intensity at the low-angle fingerprint region indicating that the hex phase is the main constituent. The intensity distribution of the survey data indicates, however, that there are variable admixtures of additional Mo-V oxide phases. Amongst them Mo5O14 [JCPDS, 31-1437] and V0.95Mo0.97 O5 [JCPDS, 1-77-649] are clearly identified components, others my be present but cannot be identified at this level of pattern quality.

17

Downloaded 18 Oct 2012 to 113.210.100.196. Redistribution subject to AIP license or copyright; see http://proceedings.aip.org/about/rights_permissions

(a)

(c)

(b)

(d)

I

tttitttfftt

IA ,

- J In .

ru i^nr - j -c

n-inhrta-ninK

• I

I H I L T l K M Z n - H * ? ! ] -

3-UK4I

j . m n. >a i i H

FIGURE 1. X-ray diffraction patterns of Mo1V0.3Te0.23Nb0.12Ox catalysts system. (a) catalyst was precalcine at 448 K then calcine at 873 K , (b) catalyst was precalcine at 498 K then calcine at 873 K,

(c) catalyst was precalcine at 548 K then calcine at 873 K and (d) catalyst was precalcine at 448 K then calcine at 873 K.

Figure 2 show the thermal behavior of Mo1V0.3Te0.23Nb0.12Ox catalysts system, where this result obtains by TGA and DSC with flow of N2 gas at 50mL min-1, dynamically run at heating rate of 5 K min-1. The arrows indicate the changeover in temperatures from pre-calcination to calcination that were used during performance optimization.

FIGURE 2. TGA, DSC and DTG curve for Mo1V0.3 Te0.23 Nb0.12Ox

catalyst system.

18

Downloaded 18 Oct 2012 to 113.210.100.196. Redistribution subject to AIP license or copyright; see http://proceedings.aip.org/about/rights_permissions

Comparison of this analysis with the choice of pre-calcination temperature allows to draw several conclusions. The catalytically successful choice at 548 K is placed at the final phase transformation of the system into a still supramolecular system. It seems disadvantageous to let a system crystallize that is fully coordinated without any inner reactivity of the solid left. It seems also inadequate to keep redox-active ligands during the main calcination and so to support autoreduction. Thus the compromise between preservation of the oxidation state at high temperatures and maintenance of inner solid state reactivity is right at the end of the removal of the redox-active ligands and before the final condensation step of the inorganic solid.

The temporal assignment of the processes and not the attribution of exact temperatures is done deliberately as the detailed inspection of the DTG and DSC curves gives numerous hints about the metastability of the solid state products and hence towards the inadequacy of phase assignments tacitly done with assigning stability temperatures. This notion is fully in line with the XRD observations of extensive and differing amorphous phases and transient crystallization and re-dissolution of ternary compounds (Anderson phase).

The fact that the optimum pre-calcination temperature in air coincides with a phase-formation event in the thermal analysis done under inert atmosphere points to the redox-labile character of the oxide matrix: the solid is in a position to deliver all oxygen required to remove the ligands present in the standard recipe without being affected in its phase-formation dynamics. The solid acts as a storage medium or sponge for oxygen. This can be seen as an indirect indication of the function of this material as a redox catalyst during which function many moles of oxygen will have to be moved through the gas-solid interface.

FIGURE 3. SEM -image of the final Mo1 V0.3Te0.23Nb0.12 Ox catalysts was pre-calcine at (a) 448, (b) 498, (c) 548, (d) 598 K and a catalyst was calcine at 873K under flow of argon.

Figure 3 show the micromorphology structure of the effect of precalcination MoVTeNbOx

catalyst series by the co-existence of three morphologies: At low calcination temperature (a) needles occur; they shorten at medium temperatures (b) and degenerate to platelets at above 548 K (c & d). A featureless glue phase can be seen in all images covering the large objects and binding small crystallites forming complex aggregates as third morphology. This glue

19

Downloaded 18 Oct 2012 to 113.210.100.196. Redistribution subject to AIP license or copyright; see http://proceedings.aip.org/about/rights_permissions

phase appears as fragment from outer agglomerates in some of the low magnification images as compact featureless platelets. The aggregate phase seems to become gradually less abundant in high-temperature calcination samples. The fact that the glue phase can be seen on both large-scale regular objects (platelets and needles) and on the agglomerates indicates that the two-regular-shaped morphologies are formed first and that the glue phase that must of different chemical composition than the rest is formed last. The lack of porosity except that of intergranular pores further indicates that all the microstructure seen in the SEM has been formed after the pre-calcination step where large amounts of gas have evolved. This observation is fully consistent with the XRD observations showing X-ray amorphous materials after the pre-calcination.

CONCLUSION

The thermal analysis study clearly reveal the complex interplay of ligand removal, polycondensation and phase formation occurring in several intertwined processes separated by temperature. The separation is affected by the redox chemistry of the system. The XRD investigation revealed all samples contain mainly the hex phase of the target systems with the pentagonal bipyramid motif. The ortho phase is present in some samples as a minority contribution. The literature claim about the assignment of phases using the famous five reflections must be taken with great care as evidenced by the reference preparation. The phase formation and phase distribution can be affected by changing the redox reactions during pre-calcination and calcination. This indicates the highly metastable character of the catalytically relevant target phases. The analysis has shown how critical the redox chemistry during calcination affects the final structure and how well the systems remember their thermal pre- history.

ACKNOWLEDGMENTS

Financial support from the Ministry of Science Technology and Innovation (MOSTI) through the projects IRPA RM 8 Grants 3302033010 as well as by the COMBICAT Grants is grateful acknowledged.

REFERENCE

1. M.M. Lin, M.Linsen, and M.William, Europe Patent Application No. 962, 253, A2 (1999). 2. M.M Lin, Applied Catalysis A: General, 207, 1-16 (2001). 3. R.K. Grasselli, Catalysis Today, 49(1 -3), 141-153 (1999). 4. Chemical Week, pp15, September 18 (2001). 5. M.M.Lin, C.Long, M.Linsen, F.Kiser, P.Klugherz, J.McGregor, M.Clark, Catalysis and Surface Science ,

Symposium, 219 th ACS Meeting (2002). 6 . L.W.Lu o, L.Yamashita, L.Labinger, H.Davis, Microporous and Mesoporous Materials, 37(1-2), 5 7 - 65

(2000). 7. H.Watanabe, Y.Koyasu, Applied Catalysis.A: General, (194-195), 479-485 (2000). 8. P.Beato, A.Blume, F.Girgsdies, R.E.Jentoft, R.Schlögl, O.Timpe, A.Trunschke, G.Weinberg, Q.Basher,

F.A.Hamid, S.B.A.Hamid, E.Omar, L.Mohd Salim, Applied Catalysis.A 307, 137 (2006). 9. J. Holmberg, R.K. Grasselli, A. Andersson, Topic in Catalysis, 23(1-4), 55-63, (2003). 10. M. Aouine, J.M.M. Millet, J.L Dubois, Chemical Communication, 13, 1180-1181, (2001)

20

Downloaded 18 Oct 2012 to 113.210.100.196. Redistribution subject to AIP license or copyright; see http://proceedings.aip.org/about/rights_permissions