Catalysts for ultra deep hydrodesulfurization and/or aromatics saturation of middle distillates

Transcription

1 Catalysts for ultra deep hydrodesulfurization and/or aromatics saturation of middle distillates Yuji YOSHIMURA a), Makoto TOBA a), Yasuo MIKI a), Yoshihiro MORITA b), Takahisa HORIE b), Yuichi TAKAMORI b), Hisaya ISHIHARA b), Takashi KAMEOKA b). a) National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 5, Higashi, Tsukuba , Japan b) Catalysis Research Center, Catalyst & Chemicals Industries Co., Ltd., 13-2 Kitaminato-Machi, Wakamatsu-Ku, Kitakyushu, , Japan Abstract In developing the ultra deep hydrodesulfurization (HDS) catalysts, lots of efforts have been focusing on how to prepare the so called Type II Co(Ni)MoS structures in the highly dispersed and stabilized conditions. In general, an increase in MoS 2 dispersion will cause a loss of crystallinity of MoS 2 phases and then result in an activity loss due to its agglomeration, etc. We have developed a novel method to prepare the highly dispersed and highly crystalline MoS 2 phases, and finally developed the CoMo and NiMo catalysts for ultra deep HDS, named as LX-NC1. On the other hand, developing of the aromatics saturation catalysts are of interest for further upgrading of sulfur free diesel, and producing the sulfur- and nitrogen-free blend stock from the non straight run feedstocks, such as Light Cycle Oil (LCO). We have developed the highly active and sulfur tolerant Pd-Pt catalysts for aromatics saturation, which catalysts will be used in the 2 nd stage process coupled with the 1 st stage process using the Ni(Co)Mo catalysts. 1. Introduction The need to reduce diesel exhaust emissions requires stringent specifications for diesel fuel. Automotive manufacturers from around the world have proposed future specifications for diesel fuel in Category 4 of the Worldwide Fuel Charter [1]. These 1/12 Aromatics (wt%) High aromatics Low aromatics GTL/BTL/BDF Novel transition metal sulfide catalysts ULSD Novel noble metal catalysts Straight run light gas oil Sulfur (ppm) Fig.1 Trends in low sulfur and/or low aromatics diesel production

2 specifications include lowering levels of sulfur (<10ppm) (Fig.1) and aromatics and lowering the T90 or T95 values. Performances of HDS catalysts are crucial to the sustainable production of ultra low sulfur diesel (ULSD) (S<10 ppm) in the existing units with minimal process modifications. Recent advancement in HDS catalysts is marked, mostly due to the development in the catalyst preparation technology to produce and control so-called Type II sites [2] in order to maximize the HDS activity, in particular, by promoting the hydrogenation pathway for the refractory sulfur compounds. Main concepts involved are to increase the dispersions of MoS 2 crystallites, and to maximize the MoS 2 edge decoration with Co (and Ni). Several approaches have been reported, i.e., controlling/skipping the calcinations step for minimizing the metals-support interaction [3], using the proper chelating agents for controlling the MoS 2 dispersion [3, 4], etc. The support properties, such as pore structures, acid/base properties and sensitivity to sulfur, etc., would also strongly contribute to the selective formation of Type II sites. On the other hand, unsupported MoS 2 catalysts, where all of the edge sites seemed to be the potential Type II sites, are also active in HDS reactions even under the high p (H 2 S) /p (H 2 ) conditions [5]. Over the highly crystalline unsupported MoS 2 particles, H 2 S seemed to promote the hydrogenation route in HDS reaction, which is favorable for HDS of the sulfur compounds with steric hindrances. Therefore, we have introduced a new concept to increase the crystallinity of MoS 2 crystallites in addition to increasing the dispersion of MoS 2 crystallites as well as the utilization of Co (and Ni). We have developed a novel HDS catalyst for ULSD production. Lower levels of aromatics will help to maximize the regeneration cycle of the exhaust gas treatment devices in addition to the reduction of PM emissions. The two-stage hydrotreating has the flexibility to cope with the simultaneous reduction in sulfur and in aromatics [6], i.e., CoMo or NiMo sulfide catalysts in the first stage and sulfur-tolerant noble metal catalysts in the second stage. For reducing aromatics in the straight-run light gas oils (SRLGO), catalytic function of aromatics saturation will be mainly required for the noble metal catalysts. In contrast, for upgrading the low cetane and highly aromatic light cycle oil (LCO) [7,8,9,10] to the low aromatic ULSD or low sulfur gasoline blend stock, two different kinds of catalytic functions will be required depending on the products, i.e., aromatics saturation for ULSD pool, and hydrocracking (HYC)/selective ring opening (SRO) functions [10] for gasoline and ULSD pools. High cost of hydrogen will be required for aromatics saturation of LCO, but some of the cost up will be offset with the volume swell of products. Sulfur-tolerant noble metal catalysts could be used under the mild reaction conditions to bypass the thermodynamic limitations of aromatics saturation, but their increases in sulfur and nitrogen tolerance are still required. Conversion of aromatics into alkyl-benzenes and paraffins by the 2/12

3 HYC/SRO reaction could be advantageous for the hydrogen consumption than the aromatics saturation, but catalytic performances in HYC/ROS selectivity and life should be further improved. In this paper, we have focused on the sulfur-tolerant bimetallic Pd-Pt catalysts for aromatics reduction of the industrial feedstocks such as hydrotreated SRLGO and LCO via. aromatics saturation. 2 Experimental 2.1 Catalysis for ultra deep HDS Catalysts The NiMo/γ-Al 2 O 3 and CoMo/γ-Al 2 O 3 catalysts were prepared by pore filling impregnation with aqueous solutions containing molybdenum and nickel (or cobalt) precursors and chelating agents. The γ-al 2 O 3 was used as a support (Catalysts & Chemical Industries, Co., Ltd.(CCIC), surface area of 209m 2 /g and pore volume of 0.81cm 3 /g). The conventional CoMo/γ-Al 2 O 3 and NiMo/γ-Al 2 O 3 catalysts (calcined) were also used as references. Catalysts were sulfided at 360 o C for 2 hours in a flow of H 2 S(5%)/H 2 (95%) before typical HDS reaction and characterization Characterization of catalyst The numbers of stacking layers and slab lengths of MoS 2 crystallites were measured by TEM analyses (JEOL, JEM-2000EXII). The Mo K-edge EXAFS spectra were measured at the Photon Factory (BL-10B) in the Institute of Materials Structure Science, High Energy Accelerator Research Organization, KEK at Tsukuba in Japan. NO uptake data were obtained by a NO pulse adsorption method (Ohkura Riken, R6015) for the sulfided catalysts with/without post-reduction treatments at 340 o C. DRIFT analyses of adsorbed NO on the sulfided catalysts were done using a FT-IR spectrometer (Thermo Electron, NEXUS 670). All of the catalysts were sulfided at 360 o C Evaluation of catalytic performances HDS runs were done in a high-pressure fixed-bed continuous-flow reactor [11] and in a pilot plant of CCIC. Two kinds of liquid feedstocks were used: (a) model compounds consisting of tetralin (30wt%), n-dodecanne (69.7 wt%), 4,6-dimethyl dibenzothiophene (4,6-DMDBT) (S=1000 ppm), and n-butylamine (N=20 ppm), and (b) SRLGO containing 1.54wt% of sulfur and 130 ppm of nitrogen for long-term tests in a pilot plant. The reaction conditions for the catalytic experiments were as follows: total H 2 pressure of 3.9 MPa, reaction temperature of 593 K, WHSV of 16 h -1, H 2 /feed flow ratio of 500 l (NTP)/l for feedstock (a); total H 2 pressure of 4.9 MPa, reaction temperature of 613 K, LHSV of 1.5 h -1, H 2 /feed flow ratio of 250 l(ntp)/l for feedstock (b). The feedstocks and products were analyzed by using a gas chromatograph (Agilent 6890) for model compounds, a SFC (Dionex 600SFC) for aromatics, a sulfur analyzer (Mitsubshi TS-100V), a GC-SCD (Sievers 355) for the sulfur compounds. 2.2 Catalysis for aromatics saturation 3/12

4 2.2.1 Catalysts Pd-Pt catalysts were prepared using different supports: USY zeolite and Yb-USY zeolites. USY zeolite (SiO 2 /Al 2 O 3 =13.9) was obtained from Tosoh Co.. Yb-USY zeolites were prepared by impregnation of USY zeolite with Yb acetate solution. The bimetallic Pd-Pt catalysts supported on zeolites were prepared by incipient wetness impregnation using [Pd(NH 3 ) 4 ]Cl 2 and [Pt(NH 3 ) 4 ]Cl 2 (mole ratio Pd:Pt=4:1). The total amount of metal loading was 1.2wt%. These impregnated catalysts were dried at 333 K and then calcined in an oxygen stream at 573 K for 3 h, and then reduced in situ in H 2 before the catalytic tests. For LCO upgrading, the higher silica content of USY zeolite (SiO 2 /Al 2 O 3 =80) was used as a support Characterization of catalyst Metal dispersion (D R ) of the Pd-Pt catalysts was determined by using a CO pulse adsorption method. Acid strength and the amounts of each catalyst sample were determined by using microcalorimetric measurements of the differential heat of adsorption of ammonia at 303K after the catalysts were evacuated at 673 K for 4 h. The EXAFS of Pt L III -edge and Pd K-edge were measured for samples of each catalyst using a double-crystal monochromator at the Photon Factory (BL-12C, 10B) of the Institute of Material Structure Science (KEK-PF) Evaluation of catalytic performances Hydrogenation runs were done in a high-pressure fixed-bed continuous-flow reactor under an up-flow mode for model compounds hydrogenation. Three kinds of feedstocks were used: (a) model compounds consisting of tetralin(30wt%), n-hexadecane(69.7 wt%), 4,6-dimethyldibenzothiophene (4,6-DMDBT) (S=300 ppm), and n-butylamine(n=20 ppm), (b) hydrotreated SRLGO containing 263 ppm of sulfur, 26.3% of aromatics(19.4% of mono-ring aromatics, and 6.9% of di- and tri-rings aromatics), and 8 ppm of nitrogen, and (c) LCO obtained from FCC unit where a mixture of DSVGO (60%) and DSVR (40%) was used as a feed oil: density (@15 C) = g/cm 3, sulfur = 6090 ppm, nitrogen = 225 ppm, and cetane index = The conditions for the catalytic experiments were a total H 2 pressure of 3.9 MPa, reaction temperature of 553 K, WHSV of 16 h -1 (for model compounds), WHSV of 4 h -1 (for hydrotreated SRLGO), H 2 /feed flow ratio of 500 l(ntp)/l. Reaction conditions for LCO upgrading are shown later. 3. Results and Discussion 3.1 Catalysis for ultra deep HDS Structure of the developed catalysts Fig.2 shows the TEM picture of the developed NiMo/γ-Al 2 O 3 catalyst. Average slab length of 4.4 nm and average number of stacking layer of 1.7 indicated that MoS 2 crystallites were highly dispersed with minimal stacking [12]. These values were smaller than those of the conventional NiMo/γ-Al 2 O 3 and CoMoγ-Al 2 O 3 catalysts. 4/12

5 The Mo3d and Ni2p XPS spectra for the developed NiMo/γ-Al 2 O 3 and the conventional NiMo/γ-Al 2 O 3 catalysts (calcined) showed that sulfidation of Mo and Ni phases were more advanced for the developed NiMo catalyst, i.e., Mo 4+ ratio>80%. Mo K-edge EXAFS data showed that average coordination numbers of S and Mo around Mo atom were 5.9 and 3.7, respectively for the developed NiMo catalyst, comparing 5.3 and 3.4, respectively for the conventional NiMo catalyst (calcined). This indicated that Fig.2 TEM picture of sulfided highly crystalline MoS 2 crystallites were NiMo/γ-Al 2 O 3 catalyst dispersed on the developed NiMo catalyst. Fig.3 shows the NO uptake for the sulfided catalysts after post-reduction treatments. NO is adsorbed on the sulfur coordinatively unsaturated sites (CUS). New CUS sites were formed for each of catalysts with increasing the reduction temperature, but highly crystalline MoS2 structure over the developed NiMo catalyst retarded the CUS formation, i.e., stronger Me-S affinity. However, this indicated the lowest amounts of CUS compared with the conventional NiMo catalyst Fig.3 NO uptake for the sulfided catalysts (calcined) and CoMo catalyst. NO absorbance spectra were also obtained for the developed and conventional NiMo/γ-Al 2 O 3 and CoMo/γ-Al 2 O 3 catalysts. NO molecules adsorbed on CUS on Co sulfide phases (absorption band at about 1865 cm -1 ) and MoS 2 edge sites (1700 cm -1 ) [13]. Interestingly, frequency shifted downward, i.e., weaker N-O bond, for the developed catalyst. This is indicative of the increase in the electron density of CUS sites. This might be due to an increase in basic properties of the neighboring S 2- to donate the electron to CUS sites after the crystallinity improvement of MoS 2. This increase in 5/12

6 basicity of S 2- and CUS sites might be linked with the abstracting ability of H + from S-compounds in the E 2 elimination reaction [14,15] HDS performances of the developed catalysts Table 1 shows the HDS activity and selectivity for 4,6-DMDBT for the conventional NiMo catalyst (calcined) and developed NiMo and CoMo catalysts. Table 2 shows the HYD activity and selectivity for tetralin. Table 1 HDS performances of the sulfided NiMo and CoMo catalysts Catalysts HDS (%) Conversion (mol%) Total DDS HYD DM-BP NiMo/Al 2 O 3 (conv.) NiMo/Al 2 O CoMo/Al 2 O Reaction schemes of tetralin hydrogenation and HDS of 4,6-DMDBT are shown in Fig.4. The developed NiMo/γ-Al 2 O 3 catalyst showed the highest HDS activity (90.3%) with the most dominant hydrogenation pathway (HYD=71.6%). In contrast, the developed CoMo/γ-Al 2 O 3 catalyst showed the higher HDS activity (83.4%) Product distribution (mol%) TH- DMDBT HH- DMDBT PH- DMDBT Table 2 HYD performances of the catalysts Conversion Selectivity (mol%) (-) Catalysts trans-decalin/ HYD cis-decalin NiMo/Al 2 O 3 (conv.) NiMo/Al 2 O CoMo/Al 2 O with the most dominant direct desulfurization pathway (DDS=35.5%), though about 50% of HDS are still via. HYD pathway. As shown in Table 3, the developed DM- CHB DM- BCH DM-BCH/ DM-CHB Selectivity (-) DM-BCH+DM- CHB)/DM-BP Fig.4 Reaction networks 6/12

7 CoMo/γ-Al 2 O 3 catalyst showed the lowest tetralin HYD activity (4.5%) compared with the developed NiMo catalyst (8.6%). This selective hydrogenation of 4,6-DMDBT compared with aromatic compound of tetralin would be an advantage of the developed CoMo catalyst in minimizing the hydrogen consumption. Table 3 shows the properties of product obtained after hydrotreating SRLGO over the developed NiMo/γ-Al 2 O 3 catalyst. Deeper hydrodenitrogenation and some hydrogenation of aromatic compounds simultaneously occurred. Boiling range shifted to lighter were mainly due to the aromatics hydrogenation. Table 3 Properties of SRLGO and ULSD Feed Product C g/cm Cetane Index Sulfur ppm Nitrogen ppm 105 1> Total aromatics wt% Mono-aromatics wt% Di-aromatics wt% Tri+-aromatics wt% Distillation properties IBP C T10 C T30 C T50 C T70 C T90 C FBP C * Reaction conditions: T=340 C, P=4.9MPa, LHSV=1.5h -1 H 2 /oil=250 Nl/l; product after the time on stream of 50 h HDS performances of the developed catalysts in the pilot plant runs The above-mentioned developed catalysts were finally manufactured industrially after several optimizations. As the new concepts introduced in our catalyst technology were related with the Highly Dispersed Nano Crystalline Type II Active Phases, so the newly HDS catalyst was named as LX-NC1, and was commercialized by Catalyst & Chemical Industries Co., Ltd (CCIC). Fig.5 HDS activity of LX-NC1 7/12

8 Fig. 5 shows the performances of LX-NC1 and the conventional CoMo/γ-Al 2 O 3 catalyst of CDS-LX6 (CCIC) in hydrotreating of SRLGO (S=1.542 wt%, N=130 ppm) in a pilot plant. Improvement in HDS activity was significant, and an improvement was about 20 C at the product sulfur of 7 ppm. HDS activity of LX-NC1 was quite stable, and the temperature compensation rate was about 0.5 C/month during the stable runs. 3.2 Catalysis for aromatics saturation Structure of the developed catalysts Table 4 shows the dispersion data for Pd-Pt/Yb (5wt%)-USY zeolite and Pd-Pt/USY zeolite catalysts: Pd+Pt=1.2 wt%, Pd/Pt atomic ratio=4, SiO 2 /Al 2 O 3 ratio of zeolite=13.9 for both of the catalysts. Yb modification (atomic ratio of Yb to Pd+Pt was 3.1) caused an increase of the dispersion of reduced catalysts (D R ), despite the loss of the strong acidic sites of zeolite [16]. Yb modification of USY zeolite decreased the dispersion of the reduced-sulfided catalysts (D RS ). The basic Yb species might donate electrons to the Pd-Pt phases to promote the surface sulfidation (D R -D RS ), but retard the agglomeration of Pd-Pt particles under the sulfiding and hydrotreating conditions as confirmed with the TEM pictures. Table 4 Effect of Yb modification on catalytic performances Catalysts Pd-Pt/USY(13.9) Pd-Pt/Yb-USY(13.9) Dispersion (D R ) (%) Dispersion (D RS ) (%) =(D R -D RS )/D R (%) HYD of tetraline(%)* trans -decalin/cis -decalin (-) HDS of 4,6-DMDBT (%)* ,3'-DMBCH/3,3'-DMCHB** TOF HYD (10 2 h -1 )*** TOF HDS (h -1 )**** SiO 2 /Al 2 O 3 molar ratio of USY zeolite was 13.9 * HDS: hydrodesulfurization of 4,6-DMDBT; HYD: conversion of tetralin. Feedstock: 4,6-DMDBT(S=300ppm)/n-butylamine (N=20ppm)/tetralin(29.7wt%)/n-C 16 Reaction conditions: T=553K, P=3.9MPa, WHSV=16h -1, H 2 /Feed=500Nl/l. Each activities are after the time on stream of 50h. ** ratio of 3,3'-dimethyldicyclohexyl to 3,3'-dimethylcyclohehyylbenzenne *** TOF calculated based on D RS **** TOF calculated based on D R -D RS HYD performances of the developed catalysts (test reactions with tetralin) Table 4 also shows the tetralin HYD activity and 4,6-DMDBT HDS activity for Pd-Pt/USY and Pd-Pt/Yb-USY zeolite catalysts. Yb modification showed a doubling in the TOF in HYD with the little changes in the TOF in HDS [17]. The highest value of trans-decalin/cis-decalin ratio after Yb modification suggests that adsorption of the hydrogenated tetralin intermediate on the HYD active sites was attenuating, thus 8/12

9 facilitating its desorption and readsorption. The high selectivity of 3,3 -dimethylbicyclohexyl (3,3 -DMBCH) to 3,3 -dimethylcyclohexylbenzene (3,3 -DMCHB) after Yb modification was due to the promotion of the hydrogenation route, which accompanied their high HYD activity. Table 5 shows the comparison in HDS of 4,6-DMDBT and tetralin HYD activities between Pd-Pt/Yb-USY zeolite and the developed NiMo/γ-Al 2 O 3 catalyst. Pd-Pt/Yb-USY zeolite catalyst. The former catalyst showed quite a high HDS and an equivalent HYD activity comparing with the latter catalyst, even under the 40 C lower reaction temperature. Table 5 Catalytic performances of noble metal catalyst and NiMo catalyst. Catalysts Reaction temperature ( C) Hydrogenation of tetralin Hydrodesulfurization of 4,6-DMDBT HDA (%) * HDS (%)* Pd-Pt/Yb-USY(13.9)** NiMo/Al 2 O 3 *** * Each of activities was obtained after the time on stream of 50h. ** Reaction conditions: T=280 C, P=3.9MPa, WHSV=16h -1, H 2 /Feed=500Nl/l. *** Reaction conditions: T=320 C, P=3.9MPa, WHSV=16h -1, H 2 /Feed=500Nl/l. Feedstock: 4,6-DMDBT(S=300ppm)/n-butylamine (N=20ppm)/ tetralin(29.7wt%)/n-c 16 (balance) HYD performances of the developed catalysts in the pilot plant runs (HYD of ULSD) Table 6 shows the HDS and hydrodearomatization (HDA) performances of Pd-Pt/Yb-USY zeolite catalyst in hydrotreating a ULSD (S=6ppm, total aromatics=21.0wt%). Most of the aromatics were converted into the naphthenic compounds after hydrotreating, but the aromatics contents would be easily conditioned by changing the reaction conditions. Table 6 Properties of low aromatic ULSD Properties Units This work Sweden GTL Feed Product Class 1 Diesel Cetane Intex Density 15 C g/cm Sulfur content mass ppm <1 Total aromtics content mass Polyaromatics content (di+tri+) mass 1.7 < Distillation properties 10 C C T90 C Elemental analysis H/C /12

10 An ultra-low aromatics diesel was also produced by hydrotreating a low sulfur diesel (440 ppm sulfur, 22.1% mono-ring aromatics, 5.4% of di- and tri-ring aromatics), and 8 ppm nitrogen) over the Pd-Pt/Yb-USY zeolite catalysts under more severe reaction conditions using a bench-scale reactor, where the reaction conditions were T = C, P = 4.9 MPa, H 2 /Feed = 500 NL/L, LHSV = 2 h 1. Stable production of low aromatics diesel (< 5wt%) was confirmed over this catalyst. Figure 6 shows Z-contrast images of the ultra-microtome cut sample of the spent Pd-Pt/Yb-USY zeolite catalyst used for 2700 h. The bimetallic Pd-Pt particles, which appeared as white spots, were still highly dispersed, even after the time on stream of 2700 h. Fig.6 TEM picture (Z-contrast image) of spent Pd-Pt/Yb-USY zeolite catalysts LCO upgrading over the Pd-Pt catalysts in the pilot plant runs Table 7 shows the catalytic performances of Pd-Pt/USY zeolite ((SiO 2 /Al 2 O 3 =80) based catalyst in hydrotreating the hydrodesulfurized and hydrodenitrogenated LCO [9]. With the 2-stage hydrotreating of LCO, LCO could be successfully converted into the low aromatics (<10wt%), ultra low sulfur (<10ppm) and high cetane index (<42) diesel blend stock. Further modification of catalytic functions, i.e., addition of selective ring opening/hydrocraking functions, are in progress. 4. Conclusions 1. We have introduced a new concept to increase the crystallinity of MoS 2 in addition to increasing the dispersion of MoS 2 crystallites as well as the utilization of Co (and Ni) for the deeper HDS. According to this concept, we have finally developed a novel HDS catalyst of LX-NC1 for sulfur-free diesel fuel production in the existing units operating with almost the same conditions as in the production of <50 ppm sulfur diesel fuel. 10/12

11 Table 7 Properties of feed LCO and hyodrtreated LCO Catalysts Feed LCO 1st reactor product Transition metals sulfide H 2 /Oil NL/L pph 2 MPa LHSV h nd reactor product Noble metal catalyst Temperatute C C g/cm Cetane index Sulfur ppm Nitrogen ppm Total aromatics wt% Monoaromatics wt% Diaromatics wt% Tri+ aromatics wt% Distillation properties IBP C T10 C T30 C T50 C T70 C T90 C The bimetallic Pd-Pt catalyst supported on Yb-modified USY zeolite, i.e., Pd-Pt/Yb-USY zeolite catalyst, showed excellent HDS and HDA activity, and showed high sulfur and nitrogen tolerance. ULSD containing less than 5% aromatics (di+ aromatics <1%) could be produced over these catalysts at moderate reaction conditions, and their activities were confirmed during a time-on-stream of 2700 h in a bench-scale high-pressure plant. 3. LCO could be successfully converted into the low aromatics (<10wt%), ultra low sulfur (<10ppm) and high cetane index (<42) diesel blend stock by the two-stage hydrotreating of LCO, using the Pd-Pt based catalysts in the 2 nd stage. Its sustainable activities were confirmed over more than 7000h in a bench-scale high-pressure plant. Acknowledgement LCO upgrading work has been carried out as a research project of the Japan Petroleum Energy Center (JPEC) with the subsidy of the Ministry of Economy, Trade and Industry, Japan. References [1] [2] R. Candia, O. Sørensen, J. Villadsen, N-Y. Topsøe, B.S. Clausen, H. Topsøe, Bull. Soc. Chim. Belg. 1984, 93(8-9), [3] J.A.R. van Veen, E. Gerkema, A.M. van der Kraan, A. Knoester, J.Chem. Soc., 11/12

12 Chem,. Commun., 1987, [4] Y. Yoshimura, N. Matsubayashi, T. Sato, H. Shimada, A. Nishijima, Appl. Catal. A:General 1991, 79, [5] H.Farag, K.Sakanishi, J. Catal., 225 (2004) [6] B.H.Cooper, B.B.L. Donnis, Appl. Catal. A:General 137 (1996) 203. [7] V.P.Thakkar,S.F.Abdo,V.A.Gembicki,J.F.McGehee, NPRA paper, AM-05-53, (2005). [8] S. McGovern, C.K. Lee, J. Zagorski, NPRA paper, AM-06-08, (2006). [9] J.Futikami, T.Horie, Y.takamori, K.Ohhama, S.Ando, Y.Shitaisi,H.Ishihara, T.Hatano, K.Ookuma, Prepr. Pap.-Am. Chem.Soc., Div. Petr. Chem 2006, 51(2), [10] R.Giardino, V.Calemma, A.Carati, M.Ferrari, Pap.-Am. Chem.Soc., Div. Petr. Chem 2007, 52(2),3-4. [11] H. Yasuda, M. Higo, S. Yoshitomi, T. Sato, M. Imamura, N. Matsubayashi, H. Shimada, A. Nishijima, Y. Yoshimura, Catal. Today, 1997, 39, [12] Y.Yoshimura, M.Toba, T.Matsui, H.Ishihara, T.Ida, T.Kameoka, J.Fuchikami, Y.Furukawa, Prepr. Pap.-Am. Chem.Soc., Div. Petr. Chem 2005, 50(4), [13] N-Y. Topsøe, H. Topsøe, J. Catal. 1983, 84, [14] S. Kasztelan, In Hydrotreating Technology for Pollution Control: Catalysts, Catalysis and Processes, M.L.Occelli and R.Chianelli, Eds.; Marcel Dekker, Inc., New York, 1996; pp [15] V. Meille, E. Schultz, M. Lemaire, M. Vrinat, M. J. Catal., 1997, 170, [16] Y. Yoshimura, H. Yasuda, T. Sato, N. Kijima, T. Kameoka, Appl. Catal., A :General, 207(2001) [17] Y.Yoshimura, M.Toba, T.Matsui, M.Harada, Y.Ichihashi, K.K.Bando, H.Yasuda, H.Ishihara, Y.Morita, T.Kameoka, Appl. Catal. A:General, 322(2007) /12