J. MATER. CHEM., 1994, 4(12), 1897-1901 Synthesis and Catalytic Properties of Magnesia Fine Powders prepared by Microwave Cold Plasma Heating Kazuo Sugiyama,* Yasushi Nakano, Hirofumi Souri, Eijiro Konuma and Tsuneo Matsuda Faculty of Engineering, Saitama University, 255 Shimo-okubo, Ura wa 338, Japan A fine powder of magnesia was prepared using microwave cold plasma as the heat source, and its crystal structure and catalytic properties were examined. Magnesium hydroxide was used as the raw material, and magnesia samples were prepared with variable parameters of microwave output and treatment time. After the start of heating, the temperature of the plasma immediately rose to 1100 "C.Magnesia fine powders were formed by heating the hydroxide for 1-2 min. The samples obtained had a large specific surface area (e.g.350 m2 g-'>). The morphology of the samples was observed by scanning electron microscopy, and they were seen to be net-like with micropores of a few hundred nanometres in diameter as opposed to the flat morphology obtained in general electric furnace heating. Solid basicity measurements revealed that the magnesia powders had strong surface basic sites of 27.0 6 H -<33.0. When the catalytic properties were examined using the condensation of benzaldehyde and the oxidative coupling of methane, which are typical base-catalytic reactions, the magnesia prepared by plasma heating exhibited high activity in both reactions. Commonly used processes for manufacturing powdered mate- rials include the preparation of fine particles by mechanical grinding of coarse particles,' the nuclear growth of ions or molecules in the liquid pha~e~,~ and the condensation of metal va~our.~ Recently, thermal plasma processes for manufacturing fine powders of carbides and nitrides have been st~died.',~ In these processes, solid, liquid and gaseous materials are injected into a thermal plasma torch to produce fine powders.In the plasma spraying method,' for example, a metal salt is sprayed and vaporized in a plasma at several thousand degrees centigrade, and then quenched to form dense particles. It is interesting to investigate the properties of powder materials produced by a cold plasma, because a cold plasma (also called a glow discharge plasma) is several hundred degrees centigrade lower than a thermal plasma.We have previously reported' that almost all types of metal oxide fine powder could be produced quickly by microwave cold plasma heating and that the powders thus produced had a bulky structure and surfaces free of hydroxy groups. Bulky metal oxides with large surface areas are said to have a large number of surface sites that are active in catalytic reactions. For example, properties such as the capability of extracting hydrogen from methane or the H2-D2 exchange capability of magnesia, which has vacant surface sites and oxygen ions, are kno~n.~~'~ Magnesia prepared by microwave cold plasma heating is expected to have active surfaces. In this study, a magnesia catalyst was prepared using microwave cold plasma heating and its application to catalytic reactions was examined.A magnesia fine powder with a large specific surface area, which is difficult to produce by conven- tional methods, was successfully prepared. The sample mor- phologies were observed by scanning electron microscopy and the solid basicity of the powders was determined by the indicator method. We then evaluated catalytic properties on the basis of the condensation of benzaldehyde" and the oxidative coupling of methane,'2"3 typical basic catalytic reactions. Experimental Microwave Plasma Heating The microwave plasma heater used for sample preparation is shown in Fig. 1. Descriptions of each unit of the heater were reported in detail previously.* This apparatus consists of a microwave generator, an impedance adjustment unit, an appl- icator and an evacuation unit.The output power of the microwave could be altered within the range 0-1000 W. To prepare a sample, a quartz reactor (capacity 200 ml), in which was placed the raw material powder, was placed in the applicator. The internal pressure of the reactor was reduced by the evacuation unit and a predetermined level of microwave power was applied. Very rapidly the inside of the reactor attained the plasma state. Microwave irradiation was stopped after a predetermined period of time to complete the heating process. Samples and Physical Measurements Magnesium hydroxide (Konoshima Chemicals) was used as the starting material.The crystal structures of the raw material and the samples prepared were determined with an X-ray powder diffractometer (XRD; Rigakudenki RAD-C). The morphology of each sample was observed with a scanning electron microscope (SEM; Hitachi S-5000) at an acceleration voltage of 1 kV in order to avoid damaging the surface morphology of the samples. The specific surface area was determined by the BET method using nitrogen gas adsorption. The plasma temperature was measured using an optical fibre thermometer (Accufiber M-lo), which had a small black-body sensor at the end of a sapphire rod.14 The basicity of samples was determined by the indicator Dower vacuum vacuum Pump liquid N,trap plunger Fig. 1 Schematic diagram of microwave cold plasma heating apparatus method,15 in which the Hammett indicator was used to colour 0.1 g of the sample in 30 ml of dehydrated benzene.This operation was carried out in a dry nitrogen gas atmosphere. The hydroxy group modes of magnesia were observed with an FTIR spectrometer (JEOL JIR-100). The desorption behaviour of carbon dioxide gas from the samples was studied by temperature-programmed desorption. In this method, carbon dioxide gas adsorbed on prepared samples that had previously been evacuated at 100"C for 1h. Then the desorbed gas was analysed with a quadrupole mass spectrometer (Varian MAG-5152). The amount of base in each sample was determined by the pulse adsorption method. A known volume of carbon dioxide was repeatedly adsorbed until the sample was saturated with the gas.The amount of adsorbed gas was estimated by TCD gas chromatography. Catalytic Reactions For plasma heating, the catalyst preparation conditions were 200 W output power and 1.5 min heating time, while for evacuation heating, it was 600°C for 2 h, typical calicination conditions16 for generating maximum amounts of basic sites and the highest base strength on the magnesia. The condensation of 7 ml of benzaldehyde was carried out at 180°C on 0.07 g of catalyst in a batch reactor. The benzaldehyde (Wako Pure Chemical Industries) was purified by distillation under reduced pressure. The reactants and products were analysed using an FID gas chromatograph equipped with an OV1 capillary column.A fixed-bed flow system reactor was used for the oxidative coupling of methane supplied at a rate of 36mmol h-l on 1.Og of catalyst at 750°C. The methane (purity 99.95%, Takachiho Chemicals) was decarbonated by passing it through a soda lime column before it was used in the reaction. The molar ratio of methane to oxygen was 5 :1. Helium was used as the carrier at a flow rate of 30 ml rnin-l. Reactants and products were analysed with a TCD gas chromatograph. An activated carbon column was used for analysis of oxygen and carbon monoxide, while a Porapak-Q column was used for other products. Results and Discussion Preparation of Magnesia A stable orange plasma was produced in the reactor by irradiating 200 W of microwave power into the reactor filled with 10 g of magnesium hydroxide.Similar plasma states were produced with 300 and 400 W microwaves. Fig. 2 shows the XRD patterns of the raw material and the samples prepared by plasma heating. Magnesia was produced within 1.5 min when a microwaves output power of 200 W was used. Although the results are not shown here, magnesia could be prepared in 1min when the hydroxide was heated with microwaves at output powers of 300 or 400 W, i.e. a very short period of time compared with a conventional evacuation heating in an electrical furnace at 600 "C, which requires more than 2 h for magnesia preparation. The products obtained by 1.5 rnin of heating at 200 W had broader XRD peaks than the raw material, implying that the product is made up of smaller particles than the raw material.The specific surface area of the magnesia prepared by plasma heating was 382 m2 g-', ca. ten times greater than that of the raw material (37 m2 g-'). This specific surface area is also larger than that obtained by evacuation heating, which does not generally exceed 250 m2 g-' l7even when the preparation is carried out with great care. The surface area of magnesia prepared by evacuation heating in an electric furnace at 600 "C for 2 h was 205 m2 g-'. J. MATER. CHEM., 1994, VOL. 4 llA1,,J0. 20 30 40 50 60 70 80 219ldegrees Fig.2 X-Ray diffraction patterns for Mg(OH), (Cb)and MgO (0) prepared by microwave cold plasma heating. (a) Raw material; and after heating for (b)0.5 min, (c) 1min, (d) 1.5min.Fig. 3 shows the specific surface areas of magnesia samples obtained by plasma heating under various heating time and microwave output power conditions. When the microwave output power was 200 W, the specific surface area increased rapidly after 1rnin of heating and attained the maximum value, 382 m2 g-', in 1.5 min. When output powers of 300 and 400 W were used, maximum values of the specific surface areas were observed between 1and 2 rnin of heating, but these maximum values were lower than that obtained at 200 W. Thus, using this heating method, magnesia with maximum specific surface area can be prepared by controlling the microwave output power and heating time. Fig. 4 shows the change in plasma temperature with time from the start of heating.With microwaves of 200 W, the plasma reached a temperature of 1050 "C 2 rnin after the start of heating. The plasma temperature was 1180 "C for 300 W microwaves. When using microwaves at 400 W output power, the plasma temperature measurement was stopped after 1 min -400r pb, 0123456 tlmin Fig. 3 Relationship between specific surface area and heating time under different output power conditions: (0)200 W, (A)300 W, (0)400 W. Sample: MgO; 10 g. J. MATER. CHEM., 1994, VOL. 4 9c 0:0 246810 tlmin Fig. 4 Temperature vs. time plots for plasma heating under different output power conditions: (0)200 W, (A) 300 W, sample: MgO; 10 g. because it reached 1200 "C, the upper limit of the temperature sensor used.The plasma temperature could be adjusted by controlling the internal pressure of the reactor and the output power of electromagnetic waves.18 In the microwave cold plasma heat- ing process, the initial pressure in the reactor after inserting the samples was 130 Pa. When the thermal decomposition of the sample began, the reactor pressure rose to ca. lo3 Pa due to an increase in gas desorbed from the sample. The rate of increase in temperature after the start of heating and the maximum attainable temperature depended on both the microwave output power and the increase in pressure in the reactor. The internal pressure of the reactor attained its peak with the progress of sample decomposition by heating, which in turn raised the temperature to its maximum level.Samples obtained around the maximum temperature had larger specific surface areas. When heating was continued after the maximum temperature, the specific surface areas gradually reduced. The specific surface area of magnesia obtained by microwave cold plasma heating is therefore strongly influenced by the plasma temperature and heating time. Morphological Observations Fig. 5 and 6 show SEM micrographs of the surfaces of magnesia prepared by evacuation heating in an electric furnace and by microwave cold plasma heating. The magnesia pow- ders were prepared by heating the hydroxide for 2 h at 600°C in an electric furnace, or by plasma heating for 1.5min at 200 W.The magnesia prepared in the electric furnace had the familiar flat morph~logy,'~ while the magnesia prepared by plasma heating had a net-like morphology with a very large number of micropores, 100-300 nm in diameter. The higher magnification photographs of the magnesia prepared by plasma heating showed that the areas surrounding the micro- pores are also porous. Magnesium hydroxide has a hexagonal structure in which layers of OH and Mg alternate vertically along the c axis.20 Decomposition due to dehydration on heating occurs on the planes that include the OH layers which are between Mg ions. When magnesium hydroxide is decomposed by heating it in an electric furnace for several hours, nuclei of magnesia grow within the layers while the hexagonal structure is maintained.21,22 In microwave plasma heating, however, these layers of the magnesium hydroxide structure undergo two heating actions,8 i.e.dielectric heating from the inside of the samples induced by the microwaves and heating on the surfaces of samples by the plasma gas molecules, within only 1 or 2min. This rapid dehydration prevents the magnesium hydroxide from retaining its layer structure. 0.6pm Fig. 5 SEM photographs of MgO prepared by heating under vacuum in an electric furnace: (u) x 10000, (h) x 50000 Solid Basicity Magnesia, a typical basic oxide, has been extensively studied in terms of its solid basicity by means of the indicator method and the gas adsorption method et~.~~In general, the highest base strength of magnesia16 is 18.4<H-<22.3, nhich is obtained when, for example, magnesium hydroxide is decom- posed by heating it in air at 550-600T.In this study, magnesia prepared in an electric furnace adsorbed 4-ni troanil- ine (pK, = +18.4) and diphenylamine (pK, = +22.3) and developed the basic colour in both cases, but it did not adsorb aniline (pK, = +27.0) and did not colour. On the other hand, magnesia prepared by plasma heating at 200 W for 1.5 min purple-red, the basic colour of aniline, but it was not coloured with triphenylmethane (pK, = +33.0). A similar basicity was observed for magnesia prepared by heating the hydroxide for >1min with 300 or 400 W microwaves. There are therefore strong basic sites of 27.0 <H-< 33.0 on the surface of magnesia prepared by plasma heating.The desorption of carbon dioxide and the amounts of carbon dioxide adsorbed were measured for magnesia samples pre- pared by microwave cold plasma heating and by evacuation heating in an electric furnace. Fig. 7 shows the TPD profiles of carbon dioxide. There were differences in the desorption peaks at higher temperatures: the magnesia prepared in an electric furnace had peaks at 200 and 3OO0C, while that prepared by plasma heating had additional peaks at 380 and 470 "C.This supports the assertion that the magnesia prepared by plasma heating has stronger basic sites than the magnesia prepared by conventional evacuation heating. (a 1 0.6pm Fig. 6 SEM photographs of MgO prepared by plasma heating: (a) x 10000,(b) x 50000 J 100 200 300 400 500 60 desorption TI0C Fig.7 TPD profiles of COz adsorbed on magnesia: (0)MgO pre- pared by plasma heating, (a)MgO prepared by heating under vacuum in an electric furnace The amount of carbon dioxide adsorption was 0.31 mmol g-' for magnesia prepared in an electric furnace and 0.65 mmol g-' for magnesia prepared by microwave cold plasma heating (Table 1). Therefore that the magnesia pre- pared by plasma heating has a larger number of basic sites. Fig. 8 shows the IR absorption spectra. The samples pre- pared by evacuation heating in an electric furnace had a peak24 attributable to carbonyl groups (1300-1450 cm-') and a weak broad peak of hydroxy groups (near 1650cm-'), while samples prepared by plasma heating did not exhibit these peaks and showed superior IR transmittance.J. MATER. CHEM., 1994, VOL. 4 Table 1 Effect of preparation conditions on base amounts of MgO preparation method conditions base amount: mmol g-l evacuation heating 600 'C, 2 h 0.31 plasma heating 200 W, 1.5 min 0.65 4000 3000 2000 1600 1200 wave nu mber/cm-' Fig. 8 1R spectra of (a)Mg(OH)zraw material, (b)MgO prepared by heating under vacuum in an electric furnace, (c) MgO prepared by plasma heating Decarbonation and dehydration by microwave cold plasma heating results in the solid basicity of magnesia. Catalytic Properties Benzaldehyde condensation was carried out in the presence of magnesia powder catalysts prepared by microwave cold plasma heating and the observed reaction activity is shown in Fig.9. This reaction has an induction period which lasts until the catalytic surface is saturated with intermediate benzyl alcoholate." The products of the reaction consisted mainly of benzyl benzoate and a small amount of benzyl alcohol. The magnesia prepared by plasma heating had a higher activity by weight in terms of conversion to benzyl benzoate than the magnesia prepared by evacuation heating in an electric furnace. The kinetics of benzaldehyde condensation were first order" with respect to benzaldehyde after the induction period. The rate constant per surface area of the magnesia catalyst pre- pared by evacuation heating was 1.1 x lop5min-' m-2, while 2o r I A 0fl 0 1 2 3 4 reaction time/h Fig.9 Condensation activity of benzaldehyde on MgO catalysts. Reaction temperature 180"C; 7 mI benzaldehyde. 0.07 g catalyst. (0)Prepared by plasma heating, (e)prepared by heating under vacuum. J. MATER. CHEM., 1994, VOL. 4 h 40r 01 23 45 reaction timeh Fig. 10 Methane conversion and C, selectivity in the oxidative coup- ling reaction of methane over MgO catalysts. Reaction temperature 750 'C; 1.0 g catalyst. (0)Conversion, (A) C2 selectivity of MgO prepared by plasma heating, (0)conversion, (A)C, selectivity of MgO prepared by heating under vacuum. that of the magnesia prepared by plasma heating was 2.6 x min-' m-'. The catalytic oxidative coupling of methane was examined on the magnesia samples.The products of this reaction were carbon monoxide, carbon dioxide, ethane and ethene. As shown in Fig. 10, the magnesia catalyst prepared by plasma heating had a higher activity and C, selectivity than the sample prepared by evacuation heating. These results agree with those reported in ref. 25 in that the catalyst with a strong basicity yielded the a activity to C, products. The magnesia catalyst prepared by plasma heating had strong basic sites, which would be effective for activities of both the benzaldehyde condensation and the oxidative coup- ling of methane. Conclusions (1) Magnesia fine powders with large specific surface areas could be prepared by microwave heating the hydroxide for a few minutes.(2) The magnesia powders had a net-like surface mor-phology with micropores of a few hundred nanometres in diameter, and also had strong basic sites of 27.0 6H-< 33.0. (3) The high catalytic activity of magnesia prepared by this method was observed both in the condensation of benz- aldehyde and the oxidative coupling of methane. References 1 Q.-Q.Zhao and G. Jimbo, Adv. Powder Technol., 1991,2,91. 2 A. G. Walton, J. Phys. Chem., 1964,67,1920. 3 J. J. F. Scholten, A. M. Beers and A. M. Kiel, J. Catal., 1975, 36, 23. 4 K. S. Mazdiyasni, C. T. Lynch and J. S. Smith, J. Am. Ceram. SOC., 1965,48, 372. 5 P. Kong, T. T. Huang and E. Pfender, IEEE Trans. Pltrsma Sci., 1986, 14, 357. 6 K. Watari, K. Ishizaki and T.Fuyuki, J. Mater. Sci. Lett., 1989, 8, 641. 7 T. Ono, M. Kagawa and Y. Shono, J. Mater. Sci., 1985, 20, 2483. 8 K. Sugiyama, Y. Nakano, H. Aoki, Y. Takeuchi and T. Matsuda, J. Muter. Chem., 1994,4, 1497. 9 D. J. Driscoll, W. Martir, J-X. Wang and J. H. Lunsford, J. Am. Chem. SOC.,1985,107,58. 10 M. Boudart, A. Delbouille, E. G. Derouane, V. Indovina and A. B. Walters, J. Am. Chem. SOC.,1972,94,6622. 11 K. Tanabe and K. Saito, J. Catal., 1974,35,247. 12 H. Jrnai, T. Tagawa and N. Kamide, J. Catal., 1987,106, 394. 13 X. D. Peng and P. C. Stair, J. Catal., 1991,128,264. 14 R. R. Dils, J. Appl. Phys., 1983,54, 1198. 15 0.Johnson, J. Phys. Chem., 1955,59,827. 16 K. Tanabe, Solid Acids and Buses, Academic Press, New York, 1970, p. 50. 17 H. Hattori, N. Yoshii and K. Tanabe, Proc. 5th. Int. Congr. CataI., 1973, 10,233. 18 A. von Engel, Electric Plasmas: Their Nature and Uses,Taylor and Francis, London, 1983, p. 162. 19 S. Coluccia and A. J. Tench, J. Chem. SOC., Faraday Trans 1,1979, 75, 1769. 20 L. Pauling, The Nature of the Chemical Bond, Cornell University Press, Ithaca, 3rd edn., 1960,p. 553. 21 F. Freund, Ber. Deutsch. Keram. Ges., 1970,47, 739. 22 R. S. Gordon and W. D. Kingery, J. Am. Ceram. SOC., 1966, 49, 654. 23 K. Tanabe, CATALYSIS-Science and Technology, ed. J. R. Anderson and M. Boudart, Springer-Verlag, Berlin, 1981, p. 241. 24 J. H. Taylor and C. H. Amberg, Can. J. Chem., 1961,39,535. 25 V. R. Choudhary and V. H. Rane, J. Catal., 1991,130,411. Paper 4/02254H; Received 15th April, 1994