Effects of the Supports on Activity of Supported Nickel Catalysts

Chin. J. Chem. Eng., 15(1) 63—67 (2007)Effects of the Supports on Activity of Supported Nickel Catalysts forHydrogenation of m-Dinitrobenzene to m-PhenylenediamineLIU Yingxin(刘迎新)a,*, CHEN Jixiang(陈吉祥)b and ZHANG Jiyan(张继炎) ba College of Pharmaceuitcal Science, Zhejiang Univer sity of Technology, Hangzhou 310032, Chinab School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, ChinaAbstract The hydrogenation of m-dinitrobenzene to m-phenylenediamine in liquid phase was studied with thenickel catalysts supported on SiO2, TiO2, γ-Al2O3, MgO and diatomite carriers. Based on the experiments of X-rayd iffrac tion (XRD), X-ray photoelectron spectroscopy (XPS), temperature-programmed reduction (TPR),temperature-programmed desorption of hydrogen (H2-TPD) and activity evaluation, the physico-chemical andcatalytic properties of the catalysts were investigated. Among the catalysts tested, the SiO2 supported nickel catalystshowed the highest activity and selectivity towards m-phenylenediamine, over which 97.3% m-dinitrobenzene con-version and 95.1% m-phenylenediamine yield were obtained at 373K under hydrogen pressure of 2.6MPa after re-action for 6 h when using ethanol as solvent. Although TiO2 and diatomite supported nickel catalysts also presentedhigh activity, they had lower selectivity towards m-phenylenediamine. As for γ-Al2O3 and MgO supported catalystswere almost inactive for the object reaction. It was shown that both the activity and selectivity of the catalysts werestrongly depended on the interaction between nickel and the support. The higher activities of Ni/SiO2, Ni/TiO2 andNi/diatomite could be attributed to the weaker metal-support interaction, on which Ni species presented as crystal-lized Ni metal particles. On the other hand, there existed strong metal-support interaction in Ni/MgO andNi/γ-Al2O3, which causes these catalysts more difficult to be reduced and the availability of Ni active sites de-creased, resulting in their low catalytic activity.Keywords hydrogenation, m-phenylenediamine, m-dinitrobenzene, supported nickel catalyst, metal-support in-teraction1 INTRODUCTIONm-Phenylenediamine (MPD) is an important in-termediate for the synthesis of some polymers, aramid fibers, and the production of dyes for textiles, leather and other materials. Additionally, it is also the key component in manufacture of hair dyes, epoxy resins and polyurethane. Traditionally, m-phenylenediamine is manufactured by iron powder reduction method in which the mixture of iron powder and hydrochloric acid are used to reduce m-dinitrobenzene. However, this process is of high cost, lower yield and serious envi-ronmental pollution, therefore, the commercialization of the process is restricted in many countries. Recently, much attention has been focused on the liquid-phase catalytic hydrogenation of m-dinitrobenzene due to its high product yield, mild reaction conditions and re-duced pollution to environment[1]. Usually, the sup-ported noble metals[2—4] and Raney Ni[5]are used as the hydrogenation catalysts. However, the rather rare resource of noble metals and the severe environ-mental pollution caused in the preparation of Raney Ni greatly restrict their industrial application. The sup-ported nickel catalyst has been used for a long time in hydrogenation technologies, due to its lower cost, easy availability, ideal catalytic activity and environmental benefit[6—11]. Up to now, however, little work about hydrogenation of m-dinitrobenzene over supported nickel catalysts has been reported[12].In this work, the supported nickel catalyst has been attempted for the hydrogenation of m-dinitrobenzene. In view of the importance of metal oxide carriers in sup-ported metal catalyst[13—15], using SiO2, TiO2, γ-Al2O3, MgO and diatomite (a silicate natural support) as carriers, a series of supported nickel catalysts were prepared and evaluated in this reaction in order to find an effective catalyst, which was not found in the lit-eratures. Furthermore, the effects of these supports on the physico-chemical properties of the nickel catalysts were studied based on the characterization results of X-ray diffraction (XRD), X-ray photoelectron spec-troscopy (XPS), temperature-programmed reduction (TPR) and temperature-programmed desorption of hydrogen H2-TPD.2 EXPERIMENTAL2.1 Catalyst preparationSome of the metal oxide supports, including SiO2, γ-Al2O3, MgO and diatomite, were obtained from Tianjin Research and Design Institute of Chemical Industry, except for TiO2, which was supplied by Tianjin 3rd Chemical Reagent Plant (analytical grade).All supported nickel catalysts (20% Ni, by mass) were prepared by one-step incipient impregnation method[16]. The supports were saturated in normal aqueous solutions of Ni(NO3)2·6H2O (analytical grade) at room temperature, and then dried at 383K overnight, calcined in air at 823K for 4h, crushed and sieved to a particle size between 0.180 and 0.154mm. Prior to the activity evaluation, the catalyst samples were reduced at 723K for 2h in hydrogen stream, and then cooled to room temperature with nitrogen flow.2.2 Catalyst characterizationXRD characterization of the samples was carried out on the Rigaku D/MAX-2308 instrument using CuReceived 2005-10-31, accepted 2006-07-17.* To whom correspondence should be addressed. E-mail: yxliu@Chin. J. Ch. E. (Vol. 15, No.1)February, 200764K α radiation. The diffraction patterns were recorded over the range of 10—84°. The scanning speed was4°⋅min －1.The electronic states of the nickel species on the surface of the catalysts were determined by XPS on a PHI1600 ESCA system. The pressure in the analyticchamber was maintained below 2.7×10－6 Pa and the Mg K α X-ray excitation source (h ν=1253eV) was used. The electronic binding energy data were amended by compensating charge with C1 peak at 284.6eV .TPR was carried out in a quartz reactor. 100mg of catalyst sample was filled into reactor, heated from298 to 1173K at the rate of 10K ⋅min －1, and swept with 90% nitrogen-10% hydrogen (by volume) stream atflow rate of 50ml ⋅min －1. The effluent hydrogen con-centration was detected by the TCD.H 2-TPD was carried out on the same system as TPR experiments. 150mg of catalyst were reduced at 723K for 2h, and then cooled down and swept with hydrogen stream at room temperature for 30min. After that, the hydrogen in the system was replaced by ni-trogen to conduct the TPD experiment. The heatingrate was kept at 20K ⋅min －1, and the flow rate of ni-trogen was 80ml ⋅min －1.2.3 Activity testThe catalytic activity experiments were carried out in a 250ml stainless steel autoclave at 373K and 2.6MPa hydrogen pressure using 2.72g of m -dinitrobenzene, 100ml of solvent and 0.272g of catalyst. Samples were withdrawn at appropriate intervals and analyzed using gas chromatograph with a flame ionization de-tector (FID). Identification of the products formed during m -dinitrobenzene hydrogenation was also per-formed by GC-MS.3 RESULTS AND DISCUSSION 3.1 XRD characterizationFigure 1 shows the XRD patterns of the sup-ported nickel catalyst precursors. It can be seen that strong diffraction peaks attributed to the support itself and NiO crystallite phase were recorded for the cata-lyst precursors of Ni/SiO 2, Ni/TiO 2 and Ni/diatomite. This suggested that the interaction between NiO andthe support (SiO 2, TiO 2 or diatomite) was weak, and NiO was mainly dispersed in the form of “free-state” on the surface of the support. On the other hand, the XRD patterns of Ni/γ-Al 2O 3 and Ni/MgO showed no apparent diffraction peaks of NiO, but new phases were found. This indicated that NiO interacted strongly with γ-Al 2O 3 or MgO support. For Ni/γ-Al 2O 3, the peaks corresponding to the nickel aluminate spinel (NiAl 2O 4) phase (formed in calcination) were found. As for the XRD patterns of Ni/MgO, in addition to the peaks corresponding to MgO, the peaks corresponding to the solid solution phase Ni x Mg 1-x O could be found, because the lattice parameter and M-O bond length of MgO and NiO were rather close[14].3.2 XPS characterizationAll the catalyst precursors were characterized by XPS in order to know the difference between the nickel catalysts about the interaction of NiO with the support and the existence state of nickel on the cata-lyst surface. The XPS data for studied samples are summarized in Table 1. The binding energies of Ni2p 3/2 for Ni/diatomite and Ni/TiO 2 were close to that for pure NiO (853.8eV[17]). Comparing with pure NiO, the Ni2p 3/2 binding energy for Ni/SiO 2 was 0.8eV higher, indicating that nickel species mainly existed in the oxidative state, but interacted with SiO 2 to some extent. This result corresponded with the con-clusion of Matsumura et al .[18], who indicated that the spectra for Ni2p 3/2 could be separated into two Gaussian peaks, and the middle of broad one was lo-cated at 855.9eV , which nearly equal to that for NiSiO 3 (856.9eV). This indicated a change in the chemical state of Ni from NiO to NiSiO 3, whereas Ni/SiO 2 showed no NiSiO 3 peaks in the above XRD pattern due to the presence of NiSiO 3 in its well-dispersed state. The Ni2p 3/2 binding energy for Ni/γ-Al 2O 3 was consistent with the reported Ni2p 3/2 binding energy for NiAl 2O 4[17], in agreement with the result of XRD mentioned above. The Ni2p 3/2 binding energy for Ni/MgO was much higher than that for pure NiO mainly due to the formation of the solid so-lution Ni x Mg 1-x O as detected in XRD.Table 1 XPS results of various supported nickel-basedcatalyst precursors Catalyst Ni2p 3/2 binding energy, eV ∆E ①, eVNi/SiO 2 854.6 0.8 Ni/TiO 2 854.3 0.5Ni/diatomite 853.9 0.1 Ni/MgO 854.9 1.1 Ni/γ-Al 2O 3855.9 2.1① Energy difference between the Ni2p 3/2 binding energies forthe nickel catalyst precursor and pure NiO.3.3 Temperature-programmed reductionFigure 2 shows the TPR profiles of the supported nickel catalyst precursors. For Ni/diatomite, only one narrow NiO reduction peak at 648K was observed,Figure 1 XRD patterns of supported nickel catalyst pre-cursors after calcination1—Ni/TiO 2; 2—Ni/diatomite; 3—Ni/γ-Al 2O 3;4—Ni/MgO; 5—Ni/SiO 2;SiO 2; ● NiO; ◇ Ni x Mg 1-x O; □ MgO; ▽ NiAl 2O 4;○ γ-Al 2O 3; ■ TiO 2Effects of the Supports on Activity of Supported Nickel Catalysts for Hydrogenation of m -Dinitrobenzene to m -Phenylenediamine Chin. J. Ch. E. 15(1) 63 (2007)65which implied that the interaction between NiO andthe diatomite support was somewhat weaker. For Ni/SiO 2, besides a low temperature reduction peak at 658K ascribed to the reduction of NiO particles, an-other broad reduction peak at higher temperature about 765K appeared, which could be attributed to the reduction of NiO nearly contacted with the oxide support[19]. This result was in agreement with the characterizations of XRD and XPS. As for Ni/γ-Al 2O 3, two reduction peaks were also observed, but both the low temperature peak and high temperature peak were apparently different from those for Ni/SiO 2 in their po-sition, area, and shape. For Ni/SiO 2, the peak area at the higher temperature was much smaller than that at lower temperature, but it was just reversed for Ni/γ-Al 2O 3. This implied that NiO interacted more strongly with γ-Al 2O 3 than with SiO 2, and the nickel species mainly occurred in the form of NiAl 2O 4 as detected in XRD and XPS characterizations. As seen from Fig.2, this NiAl 2O 4 spinal phase was difficult to be reduced to Ni 0 species at the temperature below 890K. For Ni/TiO 2, a third weak reduction peak was observed at 766K, which could be attributed to the partial reduc-tion of TiO 2 support material. Levin et al .[20] reported that TiO 2 could be partially reduced to TiO x at high temperature, which could block the active Ni species on the interface of the metal and support. For Ni/MgO, no significant reduction signal of Ni 2+ ions was de-tected below 1073K. This suggested that Ni 2+ ions were completely incorporated into bulk lattice of MgO[21], and the reduction of the Ni 2+ species in Ni/MgO was too much difficult.3.4 H 2-temperature-programmed desorptionH 2-TPD is one of effective methods to character-ize active surface of a catalyst. As shown in Fig.3, the areas of the H 2-TPD peaks for the nickel catalysts supported on different carriers (MgO, diatomite, TiO 2 and SiO 2) increased in the order of Ni/MgO ＜Ni/diatomite ＜Ni/TiO 2＜Ni/SiO 2, which was just coin-cided with that of their catalytic activities (also see Ta-ble 2), because the areas of H 2-TPD peak could be cor-related well to the amounts of active Ni 0 species. The more hydrogen adsorbed, the higher catalytic activity. As an exception of Ni/γ-Al 2O 3, the amount of hydrogen adsorbed was the largest in this work, whereas its activ-ity was almost the lowest one. This phenomenon probably could be attributed to the hydrogen spillovereffect existed in reduced Ni/γ-Al 2O 3[22]. In this case, the hydrogen was spilt over from metallic surfaces towards the support surfaces, and the spilt-over hy-drogen might be adsorbed onto alumina surface in form of Al-O (S)-H (SP)[23].In the profile of H 2-TPD, the different positions of peaks corresponded to the different adsorption states of hydrogen on surface of catalyst. For Ni/γ-Al 2O 3 and Ni/diatomite, only one weak H 2-TPD peak was ob-served at 345 and 339K, respectively. For Ni/TiO 2, although three different states of hydrogen adsorption were observed, they appeared at lower temperatures (less than 380K), indicating the weaker adsorption of hydrogen on the surface of active nickel. For Ni/SiO 2 and Ni/MgO, a stronger adsorption state of hydrogen was observed, and the H 2-TPD peak temperature (565K) for Ni/MgO was much higher than that for Ni/SiO 2 (439K). This indicated that the adsorptive strength of hydrogen on Ni/MgO was stronger, so that the adsorbed hydrogen also hardly desorbed. On the contrary, hydrogen molecules were adsorbed with a moderate strength on Ni/SiO 2, and were properly ac-tivated.3.5 Catalytic activityIn the reaction of m -dinitrobenzene hydrogena-tion, m -nitroaniline and m -nitrobenzenehydroxylamine were detected as main intermediates besides final product m -phenylenediamine. Other by-products could be detected, but only in trace. As shown in Table 2 when ethanol was used as solvent, both the activity and selectivity of the catalysts were strongly affected by the types of supports. The γ-Al 2O 3 and MgO supported nickel catalysts, which had stronger metal-support interaction, had low activity for hydrogenation. Com-paratively, Ni/TiO 2 and Ni/diatomite, which had weaker interaction of the support with the active spe-cies, showed higher activities, with the conversion of m -dinitrobenzene 83.6% and 50.3%, respectively, whereas the yields of m -phenylenediamine were only at the level of 21.8% and 7.0%, respectively, after 6h of reaction. It was clear that the highest activity and selectivity toward m -phenylenediamine were obtained over the SiO 2 supported nickel catalyst. After 6h of reaction, the conversion of m -dinitrobenzene and the yield of m -phenylenediamine, using ethanol as solvent, were up to 97.3% and 95.1%, respectively.Table 3 shows the stability result of Ni/SiO 2Figure 3 H 2-TPD profiles of supported nickel catalystsFigure 2 H 2-TPR profiles of supportednickel catalyst precursorsChin. J. Ch. E. (Vol. 15, No.1)February, 200766catalyst for m -dinitrobenzene hydrogenation usingethanol as solvent. In a typical catalyst recycle ex-periment, after completion of the first reaction, the reactor was cooled to room temperature. The catalystwas separated from the reaction mixture by filtrationand reused in the next reaction without any treatment. In the first 4 recycles, Ni/SiO 2 catalyst kept high ac-tivity and selectivity to m -phenylenediamine with the conversion of m -dinitrobenzene over 95.0% and theyield of m -phenylenediamine over 92.0% after reac-tion for 6h. In the last two cycles, the catalytic activity of hydrogenation reaction decreased gradually. The effects of solvents on the hydrogenation ofm -dinitrobenzene to m -phenylenediamine over Ni/SiO 2 catalyst were examined and the results are listed in Ta-ble 4. The hydrogenation activity of Ni/SiO 2 catalystwas the highest when using benzene as solvent, while the selectivity to m -phenylenediamine was the lowest. Ethanol was found to be a good solvent for the hydro-genation of m -dinitrobenzene to m -phenylenediamine. However, methanol was less effective. Table 4 Hydrogenation of m -dinitrobenzene to m -phenylenediamine over Ni/SiO 2 catalyst in various organic solvents ①Solvent Conversion, % Yield, %ethanol 97.3 95.1 methanol 86.1 76.7 benzene 98.5 54.7 ① Reaction conditions: hydrogenation pressure 2.6MPa, tem-perature 373K, m -dinitrobenzene 2.72g, solvent 100ml, reac-tion time 6h. 4 CONCLUSIONSBoth the activity and selectivity of supported nickel catalysts for m -dinitrobenzene hydrogenation reaction were strongly affected by the physico-chemical properties of supports.The solid solution phase of Ni x Mg 1-x O inNi/MgO and the spinel phase NiAl 2O 4 in Ni/γ-Al 2O 3were formed, respectively, because of the strong metal-support interaction, which could increase the difficulty in catalyst reduction, thus resulting in lowerdegree of the utilization of metal nickel. The TiO x generated by the partial reduction ofTiO 2 might block the active nickel surface, which was responsible for the low activity of Ni/TiO 2. Somewhatweaker metal-support interaction in Ni/diatomitemade the decreasing of its catalytic activity.Among the catalyst samples studied, the highestactivity and selectivity towards m -phenylenediamine were observed over the Ni/SiO 2 catalyst in solventethanol, suggesting that the Ni/SiO 2 catalyst is a prom-ising catalyst for the synthesis of m -phenylenediamine. This result is attributed to the higher surface area ofactive nickel of the catalyst and the moderately adsorp-tive strength for hydrogen. REFERENCES 1 Bird, A.J., Thompson, D.T., Catalysis in Organic Syn-theses, Academic Press, New York (1980). 2 Yu, Z.K., Liao, S.J., Xu, Y ., Yang, B., Yu, D.R., “Hydro-genation of nitroaromatics by polymer anchored bimetal-lic palladium-ruthenium and palladium-platinum cata-lysts under mild conditions”, J . Mol . Catal ., 120, 247—255(1997).3 Mizuta, H., Nishimura, T., Wada, M., Nagata, T., “Prepa-ration of highly pure m -phenylene-diamine”, Japan Pat., 05331113 (1993).4 Mizuta, H., Nishimura, T., Wada, M., Nagata, T., “Method for the preparation of pure m -phenylenediaminefrom crude m -dinitrobenzene”, Japan Pat., 0609551(1994). 5 Wei, W.D, “Encyclopedia of organic chemical materials (Part B)”, Chemical Industry Press, Beijing (1999).6 Pawelec, B., Daza, L., Fierro, J.L.G ., Anderson, J.A., “Regeneration of Ni-USY catalysts used in benzene hy-drogenation”, Appl . Catal . A , 145(1/2), 307—322(1996).7 Rautanen, P.A., Aittamaa, J.R., Krause, A.O.I., “Liquid phase hydrogenation of tetralin on Ni/Al 2O 3”, Chem . Eng .Sci ., 56(4), 1247—1254(2001).8 Suh, D.J., Park, T.J., Lee, S.H., Kim, K.L., “Nickel–alumina composite aerogels as liquid-phase hy-drogenation catalysts”, J . Non-Cryst . Solids , 285(1/2/3), 309—316(2001).9 Salmi, T., Mäki-Arvela, P., Toukoniitty, E., Neyestanaki, A.K., Tiainen, L.P., Lindfors, L.E., Sjöholm, R., Laine, E., “Liquid-phase hydrogenation of citral over an immo-bile silica fibre catalyst”, Appl . Catal . A , 196(1), 93—102(2000).10 Jovanovic, D., Radovic, R., Mares, L., Stankovic, M.,Markovic, B., “Nickel hydrogenation catalyst for tallow hydrogenation and for the selective hydrogenation of sunflower seed oil and soybean oil”, Catal . Today ,43(1/2), 21—28(1998).11 Toppinen, S., Rantakylä, T. K., Salmi, T., Aittamaa, J.,“The liquid phase hydrogenation of benzene and substi-tuted alkylbenzenes over a nickel catalyst in a semi-batchreactor”, Catal . Today , 38(1), 23—30(1997).12 Telkar, M.M., Nadgeri, J.M., Rode, C.V ., Chaudhari, R.V .,“Role of a co-metal in bimetallic Ni-Pt catalyst for hydro-genation of m -dinitrobenzene to m -phenylenediamine”, Appl . Catal . A , 295(1), 23—30(2005). Table 2 Hydrogenation of m -dinitrobenzene to m -phenylenediamine over various supportednickel catalysts ①Catalyst Conversion, % Yield, %Ni/SiO 2 97.3 95.1Ni/TiO 2 83.6 21.8 Ni/diatomite 50.3 7.0Ni/MgO3.5 0.2 Ni/γ-Al 2O 34.8 0.8 ① Reaction conditions: hydrogenation pressure 2.6MPa, tem-perature 373K, m -dinitrobenzene 2.72g, solvent ethanol 100ml, reaction time 6h. Table 3 Stability results of Ni/SiO 2 catalyst for hydro-genation of m -dinitrobenzene to m -phenylenediamine ①Recycle times Conversion, % Yield, %1 97.3 95.12 97.1 94.93 96.8 93.94 95.5 92.65 87.6 83.26 81.3 77.2① Reaction conditions: hydrogenation pressure 2.6MPa, tem-perature 373K, m -dinitrobenzene 2.72g, solvent ethanol 100ml,reaction time 6h.Effects of the Supports on Activity of Supported Nickel Catalysts for Hydrogenation of m -Dinitrobenzene to m -Phenylenediamine Chin. J. Ch. E. 15(1) 63 (2007)6713 Marginean, P.,Olariu, A., “Influence of metal oxides onthe catalytic activity of nickel”, Appl . Catal . A , 165(1/2), 241—248(1997).14 Wang, S.B., Lu, G .Q.M., “CO 2 reforming of methane onNi catalysts: Effects of the support phase and preparation technique”, Appl . Catal . B , 16(3), 269—277(1998).15 Xu, Z., Li, Y .M., Zhang, J.Y ., Chang, L., Zhou, R.Q.,Duan, Z.T., “Bound-state Ni species —A superior form in Ni-based catalyst for CH 4/CO 2 reforming”, Appl . Catal . A , 210(1/2), 45—53(2001).16 Li, Y .M., Principle of Industrial Catalysis, Tianjin Uni-versity Press, Tianjin (1992).17 Moulder, J.F., Stickle, W.F., Sobol, P.E., Bomben, K.D.,eds., Handbook of X-ray Photoelectron Spectroscopy, Perkin-Elmer, MN (1978).18 Matsumura, Y ., Tanaka, K., Tode, N., Yazawa, T., Haruta,M., “Catalytic methanol decomposition to carbon mon-oxide and hydrogen over nickel supported on silica”, J . Mol . Catal . A Chem ., 152(1), 157—165(2000).19 Diskin, A.M., Cunningham, R.H., Ormerod, R.M., “The oxidative chemistry of methane over supported nickel catalysts”, Catal . Today , 46(2/3), 147—155(1998).20Levin, M., Salmeron, L.V., Bell, A.T., Somorjai, G.A., “The influence of TiO x deposits on a polycrystalline Rh foil on CO adsorption and desorption”, Surf . Sci ., 169, 123—137(1986).21 Bond, G.C., Sarsam, S.P., “Reduction of nickel/magnesia catalysts”, Appl . Catal . A , 38, 365—377(1988).22 Cheng, Z.X., Zhao, X.G., Li, J.L., Zhu, Q.M., “Role of support in CO 2 reforming of CH 4 over a Ni/γ-Al 2O 3 catalyst”, Appl . Catal . A , 205(1/2), 31—36(2001).23Chen, B., Conner, J.L., “Hydrogenation of organic oxy-genates on Ni/Al 2O 3 and Ni/SiO 2 catalysis”, J . Catal ., 147, 72—81(1994).。