Al-HMS Nanocatalyst

Ind. Eng. Chem. Res. 2009, 48, 4283–4292

4283

Hydrodesulfurization Activity of NiMo/Al-HMS Nanocatalyst Synthesized by Supercritical Impregnation Mehrdad Alibouri,†,‡ Seyyed M. Ghoreishi,*,‡ and Hamid R. Aghabozorg† Catalysis Research Center, Research Institute of Petroleum Industry, Tehran 18745-4163, Iran, and Department of Chemical Engineering, Isfahan UniVersity of Technology, Isfahan 84156-83111, Iran

The synthesis of NiMo/Al-HMS nanocatalyst (2.3 wt % Ni and 9.4 wt % Mo) via supercritical CO2-methanol deposition and conventional wet coimpregnation was investigated. The characterization of both catalysts by adsorption-desorption of nitrogen, oxygen chemisorptions, XRD, TPR, and TEM indicated that Ni and Mo highly and uniformly dispersed on the Al-HMS support. The results of activity of the NiMo/Al-HMS nanocatalyst in the hydrodesulfurization of dibenzothiophene demonstrated higher conversion for the NiMo/ Al-HMS nanocatalyst in contrast to conventional catalyst. The reaction rate constants at 330 °C for the NiMo/ Al-HMS nanocatalyst and the conventional one were calculated to be 3.65 × 10-5 and 2.20 × 10-5 (mol /g cat. min), respectively. Moreover, the newly developed nanocatalyst is less inhibited than the conventional catalyst by aromatics such as toluene. Introduction The production of clean diesel by hydrodesulfurization (HDS) over metal sulfide catalysts has attracted renewed attention recently due to the introduction of new environmental legislation. The European Union proposed a reduction of sulfur to 50 ppm by 2005. Furthermore, the United States Environmental Protection Agency (US-EPA) has mandated that automotive diesel fuels have no more than 20 ppm by weight (wppm) of sulfur by 2006. In a very near future it will be necessary to reduce these limits until ultra-low-sulfur diesel is obtained (99%) + dibenzothiophene (DBT, Aldrich, >99%), and (2) n-heptane + toluene (Aldrich, >99.5%) + DBT. The DBT concentration in the fuel was 0.5 wt %. For each run, to avoid hot spots, 0.50 g of catalyst was diluted with silicon carbide (SiC, 60 mesh: 0.25 mm) to a constant volume of 1.5 cm3 before being charged into the reactor. Silicon carbide of particle diameter 0.25 mm was used to enclose the catalyst bed at both ends in the reactor. Prior to the performance test, the catalysts were sulfided in situ with a solution of 3 wt % carbon disulfide (CS2, Aldrich, >99%) in cyclohexane (Aldrich, >99%) at 300 °C and 20 bar for 4 h. Upon reaching system steady state after 4 h, the reaction effluents were condensed and analyzed by a gas chromatograph (Model 8700 Perkin Elmer) using a fused silica capillary column (15 m) with a temperature program from 373 to 523 K (10 °C min-1). The temperature range, pressure, H2/feed ratio, and weight hourly space velocity (WHSV) for the HDS reaction were 270-330 °C, 20 bar, 800 Nm3/m3, and 45 h-1, respectively. All measurements in this study were triplicated with the reproducibility of (2.6% standard deviation. The reaction scheme for DBT desulfurization has been proposed.32,33 This reaction proceeds in accordance with the hydrogenolysis pathway, through the direct desulfurization route (DDS), leading to the production of biphenyl (BP), or by a second hydrogenation reaction pathway (HYD), in which one of the aromatic rings of dibenzothiophene is first prehydrogenated, forming tetrahydrodibenzothiophene and hexahydrodibenzothiophene, which is later desulfurized to form cyclohexylbenzene (CHB). In order to apply the appropriate assumption for the experimental reactor in regard to the mass transfer mechanism, it is necessary to investigate the actual geometrical conditions of the reactor and catalyst. While catalyst particles of 0.5-1.0 mm diameter were shown to have minimal mass transfer limitations, some studies34,35 has shown that when catalyst particles are sandwiched between smaller particles of an inert material, it could be assumed that the effective catalyst particle diameter is that of the smaller material. The importance of the aspect ratio (reactor internal diameter/catalyst particle diameter) and axial convective diffusion (reactor length/catalyst particle diameter) parameters for acceptable evaluation of hydroprocessing catalysts has been explained and the minimum for these parameters (25 and 50) recommended.34 The aspect ratio (37.6) and axial convective diffusion (86) parameters for the catalyst particles used in this study exceed the minimum recommended values by 1.5 and 1.7 times, respectively. It is clear that the present study has used acceptable catalyst evaluation conditions. For the synthesized and commercial catalysts, the main products of the reaction were BP and CHB. Based on the conversion and selectivity data, assuming the pseudo first order with respect to DBT, and convective mass transfer mechanism (neglecting axial dispersion coefficient Dz f 0 or in other words, NPec f ∞), the reaction rate constant of HDS (kHDS) was calculated according to the one-dimensional axial mass balance equation for the plug flow reactor: kHDS ) -

F ln(1 - x) W

(2)

4286 Ind. Eng. Chem. Res., Vol. 48, No. 9, 2009

( RTE )

kHDS ) k0 exp -

(3)

in which F is the feed rate of DBT (mol/min), W is the catalyst weight (g), x is the mole fractional conversion of DBT, k0 is the frequency factor (mol/g cat. min), and E is the activation energy (J/mol). Hydrogenolysis (DDS) and hydrogenation (HYD) overall selectivity was calculated according to the formula: So )

CCHB CBP

(4)

Characterization Methods N2 Adsorption-Desorption Isotherms. The nitrogen adsorption-desorption isotherms were determined on an ASAP 2000 Micromeritics equipment. Prior to the experiments, the samples were degassed at 300 °C in vacuum for 5 h. The volume of the adsorbed N2 was normalized to the standard temperature and pressure. Specific surface area (SBET) was calculated by the BET equation applied to the range of relative pressures 0.05 < P/P0 < 0.20. The average pore diameter was calculated by applying the Barret-Joyner-Halenda method (BJH) to the adsorption branches of the N2 isotherms. The cumulative pore volume was obtained from the isotherms at P/P0 ) 0.99. Oxygen Pulse Chemisorptions. Oxygen pulse chemisorptions were used to characterize the dispersion of MoS2 on the catalysts. An automated catalyst characterization unit (Autosorb 2321) was used for the chemisorptions measurement. The samples were pretreated in hydrogen at 250 °C for 2 h prior to oxygen chemisorptions. Oxygen chemisorptions isotherms were obtained at 25 °C. X-ray Diffraction. The X-ray patterns were recorded on a Philips PW1840 diffractometer using a monochromatic Cu KR radiation. The diffractograms were recorded in the 2θ range of 1-90° in steps of 0.02°. Temperature-Programmed Reduction (TPR). The TPR experiments of NiMo supported on Al-HMS were carried out in a semiautomatic Micromeritics TPD/TPR 2900 apparatus interfaced to a microcomputer. TPR profiles were obtained by passing a 5% H2/Ar flow (50 mL/min) through the sample. The temperature was increased at a rate of 10 °C/min, and the amount of H2 consumed was determined with a thermoconductivity detector (TCD). The effluent gas was passed through a cold trap before the TCD in order to remove water from the exit stream. Transmission Electron Microscopy (TEM). Transmission electron microscopy (TEM) in conjunction with energydispersive analysis of X-ray (EDAX) was carried out to determine the dispersion of molybdenum sulfide supported on Al-HMS support and elemental composition. The ground sample was dissolved in ethanol. The powders in the solution were deposited on a grid with a holey carbon copper film. After drying, the compounds were transferred to a Philips CM200 FEG (field emission gun) instrument with 200 kV of acceleration voltage for transmission electron microscopy test. Atomic Absorption Spectrometer (AAS). The metal content of the prepared catalyst was determined using atomic absorption technique (Perkin-Elmer Model AAnalyst 200). Results and Discussion Catalyst Characterization. The physical adsorptiondesorption of nitrogen was used to determine the textural properties of Al-HMS support (H+-Al-HMS) and NiMo/Al-

Figure 3. (a) N2 adsorption-desorption isotherms of support and two NiMo/ Al-HMS catalysts. (b) Pore size distribution of the support and two NiMo/ Al-HMS catalysts.

HMS catalysts. The obtained results are shown in Figure 3a. At low relative pressure, the N2 uptake increases sharply as the N2 pressure increases, which indicates that the support has a very high surface area. In addition, there is another sharp increase in the range of P/P0 ) 0.2-0.3 in the isotherms. This sharp increase in uptake results from the capillary condensation of N2, which suggests that uniform mesopores are present in the Al-HMS support. All the isotherms are type IV shapes and the uptake of N2 decreases with the loading of the metal species on the support. This observation is much more for the catalyst prepared by CIM in contrast to SDM. Figure 3b shows the pore size distributions of the support and two catalysts prepared by SDM and CIM. It should be noted from Figure 3b that the pore size distribution curves are for three different cases: the first one is for the Al-HMS support with a median pore diameter of 2.6 nm, the second for the support with loaded Ni and Mo oxide with a pore diameter of 2.4 nm (SDM), and the third one with a pore diameter of 2.2 nm (CIM). The uniform and narrow pore size distributions of the support and SDM catalyst provide independent verification of retention of pore structure after impregnation of the metal oxide precursors in contrast to CIM catalyst. There appears to be a pore diameter shift to smaller diameter upon SDM and CIM deposition of Ni and Mo on the Al-HMS support. Therefore, the results in Figure 3b indicate a major advantage of using supercritical deposition in which no pore plugging occurs and it further reveals that the internal surfaces of AlHMS pores are slightly reduced.

Ind. Eng. Chem. Res., Vol. 48, No. 9, 2009 4287 Table 1. Pore Textural Parameters of Support and Catalysts and Oxygen Chemisorption Data of MoS2 Catalysts

sample

SBET (m2 g-1)

Vpa (cm3 g-1)

Dpb (nm)

oxygen uptake (mol/g cat.)

H -Al-HMS (support) NiMo/Al-HMS (SDM) NiMo/Al-HMS (CIM)

649 578 485

0.65 0.57 0.48

4.17 4.04 3.87

91 75

+

a

Vp at P/P0 ) 0.99. b Calculated by BJH method.

The pore textural parameters, such as the specific area (SBET), cumulative pore volume (Vp), and average pore diameter (Dpavg) of supports and catalysts, are listed in Table 1. The specific areas of the H+-Al-HMS (support) and the two SDM and CIM catalysts are approximately 649, 578, and 485 m2 g-1, respectively. The BET surface area of Al-HMS support after impregnation via SDM and CIM decreased about 11 and 25%, respectively. In general, the deposition of the Ni and Mo species on the AlHMS support by all conventional methods results in decreasing the BET surface area and pore volume which is well expected. According to the results in Table 1, it is imperative to realize that the decrease in the surface area and pore volume of support by using CIM is significant36-39 in comparison to SDM. Because of high surface tension, the liquid solvent cannot penetrate into the small pores and consequently is not able to deposit the solute on the internal surface area of the pores. Therefore, it is possible that the solutes in the liquid solvents plug the small pores via agglomeration after the solvent removal process. The results derived from Figure 3a,b and Table 1 indicate that the observed phenomenon is due to low surface tension of SCFs which not only permits better penetration into the pores than liquid solvents but also avoids the pore plugging and agglomeration of Mo and Ni species. Dispersion of the active phase plays an important role in catalysis. Catalysis is a surface phenomenon and typically only the fraction of the active phase exposed on the support surface is important and responsible for catalytic activity. This is described and quantified by dispersion. The oxygen chemisorption measurement can provide valuable information in regard to the dispersion of metal on the catalytic support. Thus, in this study, oxygen chemisorptions were conducted on the catalysts synthesized by SDM and CIM and the results are shown in Table 1. The results demonstrate that the oxygen uptake for the synthesized catalyst by SDM (91 mol/g cat.) is higher than the synthesized CIM catalyst (75 mol/g cat.). Higher oxygen consumption for the nanocatalyst synthesized by SDM is the indication of higher dispersed deposition of Ni and Mo on the supports. This incentive has caused 21% higher dispersion for the SDM procedure with respect to CIM. Overall, one can conclude that there is a major improvement in the dispersion of the active phases of the catalyst on the Al-HMS support using SDM in comparison with CIM. Figures 4 and 5 present the X-ray patterns of the Al-HMS samples at different stages and the two NiMo/Al-HMS catalysts preparation that were recorded from 1° to 10° and from 1° to 90°, respectively. Figure 4a shows the XRD pattern characteristic of the powder Al-HMS recorded after normal synthesis. In addition to the most intense low-angle diffraction peak of (1 0 0) at 2.4°, higher order reflections, such as (1 1 0) and (2 0 0) exist between 3° and 5° which can be distinguished in hexagonal lattice symmetry, characteristic of HMS structure.40 Figure 4b shows a pattern of the ion-exchanged proton form of Al-HMS (H+-Al-HMS) with higher intensity peak at lower order

Figure 4. Low-angle X-ray patterns of two supports and two NiMo/AlHMS catalysts.

Figure 5. Wide X-ray patterns of support and two NiMo/Al-HMS catalysts.

reflection (3°-5°) compared with Figure 4a, which is the indication of improved structural property. While the higher order reflections are weak, the sample continues to be fairly ordered. Figure 4, c and d, shows the pattern of the support after deposition of NiO and MoO3 (NiMo/Al-HMS) via SDM and CIM. It is observed that the deposition of the Ni and Mo oxides sample continues to retain a similar structural characteristic compared to Figure 4b. Figure 5 shows the XRD patterns of the samples recorded in the 1°-90° intervals. A broad band is observed between 20° and 30° in the peaks presented in Figure 5, a, b, and c, which corresponds to the amorphous part of the supports. It is observed in Figure 5c that, after supercritical impregnation of two metal salts on the support, the structural characteristics of the catalyst manifest in the higher intensity broad peak and increased width that may be due to higher dispersed deposition of Ni and Mo


Figure 6. TPR curves of two catalysts and of bulk MoO3.

Figure 7. EDAX spectra of two arbitrary points in SDM synthesized NiMo/ Al-HMS nanocatalyst.

that overlaps with the broad peak of the amorphous part of the H+-Al-HMS sample. As observed in Figure 5c, no reflections belonging to molybdenum oxides are realized and the observed peak for nickel oxides is also very weak. Contrary to the obtained results for SDM, Figure 5b shows several peaks for the CIM synthesized catalyst which is the indication of larger particle size for the two metals. Thus, one can conclude that the observed phenomenon deviating the two catalysts in terms of produced particle size is due to fact that the smaller sized metal particles deposit in the pores of the support and, therefore, cannot be detected as specific strong peaks as shown in Figure 5c. It is imperative to realize that the method of metal deposition via SDM can be pinpointed as the most significant and effective variable in the formation of highly dispersed Ni and Mo oxide phases inside the support pores due to special properties of supercritical fluids in terms of low surface tension and viscosity and higher diffusion rate. These special properties of supercritical fluids permit the dispersed deposition of metallic salts on the support surface and avoid the agglomeration of particles in contrast to liquid solvents in the CIM method. The TPR curves of the two SDM and CIM catalysts in the oxide form and of bulk MoO3 are shown in Figure 6. The TPR profile of the bulk MoO3 contains two reduction peaks at 737 and 890 °C,41 which correspond to the two-step reduction of MoO3 (MoO3 f MoO2 f Mo0).42-44 When MoO3 oxide is used in the catalyst support, a significant decrease in the reduction temperature is observed. As shown in Figure 6, the TPR profile of the synthesized SDM NiMo/Al-HMS nanocatalyst contains three peaks at 284, 449, and 704 °C. The peak at 284 °C is ascribed to the reduction of crystalline NiO weakly bound to the support.45-48 The broad and low-intensity peak at 284 °C indicates that the prevailing part of the Ni species is in intimate contact with the Mo species. Comparing the present TPR pattern with TPR of NiMoO4,45,46 the second peak (main peak at 449 °C) is assigned to the reduction of a similar NiMoO4 phase. The peak at 704 °C is ascribed to the second step of reduction of the MoO3 species. Also shown in Figure 6, the TPR profile of the synthesized CIM NiMo/Al-HMS catalyst contains two peaks at 503 and 756 °C

which demonstrate the shift of reduction peaks to higher temperature with respect to SDM catalyst. It is well established that the peaks shift to lower temperature when the support is proton-exchanged46 and the peaks shift to higher temperature when the Al is added to the support (AlHMS).47 Comparison of the TPR pattern in this study for the SDM and CIM catalysts with other TPR profiles in the literature46-48 indicates that the reduction of Ni and Mo species in the supported catalyst synthesized via supercritical deposition occurs at lower temperatures. As shown in Figure 6, the peaks assigned to the reduction of Ni and Mo species in the synthesized catalyst may be due to the interactions between each other as shown by other researchers.45-48 The enhanced reduction behavior of catalyst synthesized via supercritical deposition in terms of lower temperature requirement may be due, on one hand, to the weak interaction of metal oxide species with the support, and, on the other hand, to the decreased size and increased uniform dispersed deposition of Mo and Ni oxide species. In addition to other analyses, the sulfided catalyst used in this study was also characterized using TEM. The main objective for the TEM studies was to determine the dispersion of MoS2 in the Al-HMS supported catalyst. The low density of the AlHMS supported catalyst complicated their characterization using TEM. Despite several sample preparation methods, the Al-HMS supported catalyst could not be anchored to the grids well enough to allow the electron beam to focus and image the catalyst at high magnification. At high magnification, focusing the electron beam on the Al-HMS supported catalyst resulted in rapid movement of the Al-HMS supported catalyst, thus preventing its effective imaging. Therefore, due to the aforementioned obstacles, MoS2 particles cannot be appreciated in this procedure and the obtained results did not provide significant findings. Thus, in order to obtain evidence of the uniform dispersed deposition of active phase in the Al-HMS supported catalyst via SDM, utilization of elemental composition determination using energy-dispersive analysis of X-ray (EDAX) was carried out while performing TEM. Ni and Mo were detected at two arbitrary points of SDM nanocatalyst and their representative EDAX spectra are provided in Figure 7a,b. The quantitative

Ind. Eng. Chem. Res., Vol. 48, No. 9, 2009 4289 Table 2. Reaction Rate Constants for HDS of DBT (mol/g cat. min × 105) catalyst

270 °C

290 °C

310 °C

330 °C

NiMo/Al-HMS (SDM) NiMo/Al-HMS (CIM) NiMo/Al2O3 (commercial)

0.58 0.53 0.58

0.80 0.66 0.66

1.40 1.21 1.35

3.65 2.20 2.42

Table 3. Activation Energy and Frequency Factor for the Synthesized NiMo/Al-HMS Nanocatalyst and NiMo/Al2O3 Commercial Catalyst

Figure 8. DBT conversion as a function of reaction temperature for three catalysts.

analysis of different elements showed 1.98 and 1.53 wt % for the Ni and 9.82 and 7.83 wt % for the Mo content in the two arbitrary chosen points. It is important to realize that the Ni/ Mo ratio in the two arbitrary selected points is 0.202 and 0.195 which are very compatible. The obtained EDAX spectra of the two arbitrary points of the SDM nanocatalyst sample show images with no significant change in relative intensities. The EDAX results provide strong indication that most of the Ni and Mo in the SDM nanocatalyst are uniformly distributed in the Al-HMS support. Catalyst Activity in Terms of HDS Conversion and Selectivity. In the present study, the catalytic performance of sulfided NiMo/Al-HMS nanocatalyst in terms of conversion and selectivity was examined in the HDS of DBT; DBT was chosen as the hydrocarbon model compound because it is one of the most refractory sulfur compounds in gas oil. The GC analysis of the reactor effluent showed that BP and CHB are the only two products of this reaction applying the SDM, CIM NiMo/ Al-HMS, and commercial catalysts. For comparison purposes, CIM synthesized NiMo/Al-HMS catalyst and commercial NiMo catalyst supported on alumina (NiMo/Al2O3) were also evaluated in the HDS of DBT. Since all parameters, such as support, metal loadings, and operating conditions, except the method of impregnation are the same, any changes of activity enhancement in comparison of SDM and CIM synthesized NiMo/Al-HMS catalysts can be corresponded to the method of catalyst preparation. Figure 8 shows the conversion of DBT in the HDS using synthesized SDM NiMo/Al-HMS nanocatalyst, synthesized CIM NiMo/Al-HMS catalyst, and commercial catalyst as a function of temperature. The synthesized NiMo/Al-HMS nanocatalyst shows higher conversion of DBT into BP and CHB. For instance, conversion of 83.4% was obtained using SDM NiMo/ Al-HMS nanocatalyst at 330 °C in contrast to 66.0 and 69.6% for the CIM NiMo/Al-HMS and commercial catalysts, respectively. This increased conversion may be explained in terms of enhancement in the reaction rate constant according to the Arrhenius’ law. As the supporting evidence for this phenomenon, the calculated reaction rate constants of HDS at temperatures of 270-330 °C are summarized in Table 2. The reaction rate constant for the SDM NiMo/Al-HMS nanocatalyst at 330 °C increased about 66% with respect to the CIM catalyst. As shown in Table 3, the activation energy and frequency factor for the synthesized SDM, CIM, and commercial catalysts were calculated. The results show that frequency factor and activation

catalyst

k0 (mol/g cat. min)

E/R (°C)


3.72 × 10 9.74 × 105 1.64 × 106

9852 7902 8158

7

energy is 38 and 1.25 times higher for the new synthesized catalyst with respect to the CIM one. Even though the activation energy of the synthesized SDM catalyst is moderately higher than the CIM catalyst which is considered to be a disadvantage, the positive effect of frequency factor for the new catalyst is much higher on the reaction rate. This incentive may be explained via obtained higher dispersed deposition and uniform distribution of metal active phase on the support by supercritical impregnation method in which higher reaction rate can be achieved using the provided active surface area. According to Arrhenius’ law, increased frequency factor leads to higher number of effective collisions that cause higher conversion of reactants. In accordance with other studies,49,50 the aromatic hydrogenation efficiency is defined as the selectivity of CHB to BP. It is increasing slightly with increasing temperature as shown in Figure 9. The results demonstrate that the CHB/BP selectivity is lower using the synthesized SDM catalyst with respect to the CIM catalyst. The yields of CHB and BP are shown in Figure 10 for three different catalysts. The results indicate that the newly developed SDM catalyst produces more BP compared to CHB. The transformation of DBT occurs mainly through the direct HDS route; therefore, it is expected to observe a low CHB/BP selectivity. The CHB/BP selectivity is lower for the SDM because the higher number of active sites resulting from the higher dispersion favors the direct HDS route. The inhibitive effect of aromatics on the HDS of DBT was studied by the addition of toluene to the feed mixture. The reaction rate constant of HDS at toluene concentration 0-20 wt % in feed at 310 °C is summarized in Table 4. Addition of

Figure 9. Variation of CHB/BP selectivity with temperature for three catalysts.


between toluene and DBT for the hydrogenation sites governs the importance of the different HDS pathways. Toluene competes more with the HYD route of reaction. This leads to a decrease of conversion of DBT. Conclusion

Figure 10. BP and CHB yield as a function of DBT conversion for three catalysts. Table 4. Reaction Rate Constants for HDS of DBT in n-Heptane-Toluene Mixture at 310 °C (mol/g cat. min × 105) toluene concn (wt %)


0

10

20

1.40 1.21 1.35

1.17 0.85 0.95

0.96 0.72 0.85

20 wt % toluene for the application of the synthesized SDM NiMo/Al-HMS catalyst resulted in 31% reduction of the rate constant. Similarly, for the CIM catalyst this reduction was about 40%. The conversions of DBT in the reactions with different amount of toluene are presented in Figure 11. A small addition of toluene decreases the conversion and the inhibition effect gets stronger with the increasing amount of aromatics.28,51,52 The results of Figure 11 show that the inhibition effect of SDM NiMo/Al-HMS nanocatalyst is lower than that of the CIM and commercial catalyst. The effect of toluene clearly seems to be a suppression of the hydrogenation properties of the catalyst. The deactivation of the hydrogenolysis sites does not play a paramount role here and it can be speculated that the competition

The newly developed NiMo/Al-HMS nanocatalyst by supercritical deposition has significant advantages over the CIM and commercial catalysts in terms of (1) increased Mo and Ni dispersed deposition and uniformity on the Al-HMS support (increased active catalytic surface area), (2) increased frequency factor, (3) increased reaction rate constant and conversion, and (4) reduced metal consumption for a specified conversion. It is imperative to realize that all the aforementioned incentives are related to the obtained higher dispersion of metals via SCD method. The major obtained catalyst incentives are mainly due to the method of metal salts impregnation. Supercritical fluid properties such as high diffusivity and density, low viscosity, and low surface tension lead to significant improvement in the catalyst synthesis. High solvation capability of supercritical fluids due to higher density incorporated with an entrainer provides a uniform metal salts solution. Furthermore, high diffusivity, low viscosity, and low surface tension decrease mass transfer resistances such as film and pore diffusion in the supercritical impregnation process. Consequently, the aforementioned improved physical properties lead to higher dispersed deposition and uniformity of metals which avoids any agglomeration on the support. Hydrodesulfurization of DBT in a fixed bed reactor with a hydrocarbon feed model with and without the addition of toluene was used in the evaluation of catalyst activity. At a specified temperature, higher conversion for the new catalyst was obtained in comparison to the CIM and commercial catalysts with or without toluene in the feed. But, the presence of toluene inhibited the catalyst performance much more for the CIM and commercial catalysts compared to SDM catalyst. Acknowledgment The financial support provided for this project by the National Iranian Oil Co. (NIOC) and Isfahan University of Technology (IUT) is gratefully acknowledged. The helpful technical assistance offered by Mr. M. A. Attarnejad and Mrs. A. Tofigh is appreciated. Nomenclature

Figure 11. DBT conversion as a function of amount of toluene in feed for three catalysts.

BJH ) Barret-Joyner-Halenda BP ) biphenyl c ) Ni(NO3)2 · 6H2O concentration (mg salt/g supercritical CO2-methanol solution) CIM ) conventional impregnation method CBP ) concentration of biphenyl (mol/L) CCHB ) concentration of cyclohexylbenzene (mol/L) CHB ) cyclohexylbenzene CS2 ) carbon disulfide Dp ) pore diameter (nm) Dpavg ) average pore diameter (nm) Dz ) axial dispersion coefficient (m2/s) DBT ) dibenzothiophene DDS ) direct desulfurization route E ) activation energy (J/mol) EDAX ) energy-dispersive X-ray analysis EDF ) equilibrium deposition filtration

Ind. Eng. Chem. Res., Vol. 48, No. 9, 2009 4291 F ) feed rate of DBT (mol/min) HDS ) hydrodesulfurization HMS ) hexagonal mesoporous structured HYD ) hydrogenation reaction pathway k0 ) frequency factor (mol/g cat. min) K1 ) Langmuir adsorption constant (mg Ni(NO3)2 · 6H2O/g Al-HMS) kHDS ) reaction rate constant of HDS (mol/g cat. min) NPe ) Peclet number P ) pressure (bar) P0 ) ambient pressure (bar) q ) precursor’s equilibrium uptake by the adsorbent (mg Ni(NO3)2 · 6H2O/mg Al-HMS) Qo ) adsorption capacity (g supercritical CO2-methanol solution/ mg Ni(NO3)2 · 6H2O) R ) universal gas constant (J/(mol °C)) SBET ) specific area by BET method (m2/g) So ) overall selectivity SCF ) supercritical fluid SC-CO2 ) supercritical carbon dioxide SDM ) supercritical deposition method T ) temperature (°C) TCD ) thermoconductivity detector TEM ) transmission electron microscopy TPD ) temperature-programmed desorption TPR ) temperature-programmed reduction Vp ) pore volume (m3/g) W ) catalyst weight (g) WHSV ) weight hourly space velocity (1/h) x ) mole fractional conversion of DBT XRD ) X-ray diffraction

Literature Cited (1) Chen, J.; Ring, Z. HDS Reactivities of Dibenzothiophenic Compounds in a LC-Finer LGO and H2S/NH3 Inhibition Effect. Fuel 2004, 83, 305. (2) Perot, G. Hydrotreating Catalysts Containing Zeolites and Related MaterialssMechanistic Aspects Related to Deep Desulfurization. Catal. Today 2003, 86, 111. (3) Cejka, J. Organized Mesoporous Alumina: Synthesis, Structure and Potential in Catalysis. Appl. Catal., A 2003, 254, 327. (4) Saih, Y.; Segawa, K. Tailoring of Alumina Surfaces as Supports for NiMo Sulfide Catalysts in the Ultra Deep Hydrodesulfurization of Gas Oil: Case Study of TiO2-Coated Alumina Prepared by Chemical Vapor Deposition Technique. Catal. Today 2003, 86, 61. (5) Song, C.; Reddy, K. M. Mesoporous Molecular Sieve MCM-41 Supported Co-Mo Catalyst for Hydrodesulfurization of Dibenzothiophene in Distillate Fuels. Appl. Catal., A 1999, 176, 1. (6) Reddy, K. M.; Wei, B.; Song, C. Mesoporous Molecular Sieve MCM-41 Supported Co-Mo Catalyst for Hydrodesulfurization of Petroleum Resids. Catal. Today 1998, 43, 261. (7) Wang, A.; Wang, Y.; Kabe, T.; Chen, Y.; Ishihara, A.; Qian, W. Hydrodesulfurization of Dibenzothiophene over Siliceous MCM-41-Supported Catalysts I. Sulfided Co-Mo Catalysts. J. Catal. 2001, 199, 19. (8) Yu, K. M. K.; Yeung, C. M. Y.; Thompsett, D.; Tsang, S. C. AerogelCoated Metal Nanoparticle Colloids as Novel Entities for the Synthesis of Defined Supported Metal Catalysts. J. Phys. Chem. B 2003, 107, 4515. (9) Claus, P.; Bruckner, A.; Mohr, C.; Hofmeister, H. Supported Gold Nanoparticles from Quantum Dot to Mesoscopic Size Scale: Effect of Electronic and Structural Properties on Catalytic Hydrogenation of Conjugated Functional Groups. J. Am. Chem. Soc. 2000, 122, 11430. (10) Okitsu, K.; Nagaoka, S.; Tanabe, S.; Matsumoto, H.; Mizukoshi, Y.; Nagata, Y. Sonochemical Preparation of Size-Controlled Palladium Nanoparticles on Alumina Surface. Chem. Lett. 1999, 3, 271. (11) Miller, J. M.; Dunn, B.; Tran, T. D.; Pekala, R. W. Deposition of Ruthenium Nanoparticles on Carbon Aerogels for High Energy Density Supercapacitor Electrodes. J. Electrochem. Soc. 1997, 144, L309. (12) Maldonado-Hodar, F. J.; Moreno-Castilla, C.; Rivera-Utrilla, J.; Ferro Garcia, M. A. Metal-Carbon Aerogels as Catalysts and Catalyst Supports. Stud. Surf. Sci. Catal. 2000, 130, 1007.

(13) Bekyarova, E.; Kaneko, K. Structure and Physical Properties of Tailor-Made Ce, Zr-Doped Carbon Aerogels. AdV. Mater. 2000, 12, 1625. (14) Maldonado-Hodar, F. J.; Ferro-Garcia, M. A.; Rivera-Utrilla, J.; Moreno-Castilla, C. Synthesis and Textural Characteristics of Organic Aerogels, Transition-Metal-Containing Organic Aerogels and Their Carbonized Derivatives. Carbon 1999, 37, 1199. (15) Cushing, B. L.; Kolesnichenko, V. L.; O’Connor, C. J. Recent Advances in the Liquid-Phase Synthesis of Inorganic Nanoparticles. Chem. ReV. 2004, 104, 3893. (16) Alibouri, M. Synthesis of Hydrodesulfurization Catalyst over Hexagonal Mesopores Nano-Structure Support in Supercritical Media. Ph.D. Dissertation, Isfahan University of Technology, Isfahan, Iran, 2009. (17) Ghoreishi, S. M.; Sharifi, S. Modeling of Supercritical Extraction of Mannitol from Plane Tree Leaf. J. Pharm. Biomed. Anal. 2001, 24, 1037. (18) Ghoreishi, S. M.; Akgerman, A. Dispersion Coefficients of Supercritical Fluid in Fixed Beds. Sep. Purif. Technol. 2004, 39, 39. (19) Akgerman, A.; Erkey, C.; Ghoreishi, S. M. Supercritical Extraction of Hexachlorobenzene from Soil. Ind. Eng. Chem. Res. 1992, 31, 333. (20) Watkins, J. J.; McCarthy, T. J. Polymer/Metal Nanocomposite Synthesis in Supercritical CO2. Chem. Mater. 1995, 7, 1991. (21) Ye, X. R.; Lin, Y.; Wang, C.; Engelhard, M. H.; Wang, Y.; Wai, C. M. Supercritical Fluid Synthesis and Characterization of Catalytic Metal Nanoparticles on Carbon Nanotubes. J. Mater. Chem. 2004, 14, 908. (22) Ye, X. R.; Lin, Y.; Wai, C. M. Decorating Catalytic Palladium Nanoparticles on Carbon Nanotubes in Supercritical Carbon Dioxide. Chem. Commun. 2003, 5, 642. (23) Sun, Z.; Liu, Z.; Han, B.; Wang, Y.; Du, J.; Xie, Z.; Han, G. Fabrication of Ru-Carbon Nanotube Nanocomposites in Supercritical Water. AdV. Mater. 2005, 17, 928. (24) Saquing, C. D.; Cheng, T. T.; Aindow, M.; Erkey, C. Preparation of Platinum/Carbon Aerogel Nanocomposites using a Supercritical Deposition Method. J. Phys. Chem. B 2004, 108, 7716. (25) Zhang, Y.; Kang, D.; Saquing, C.; Aindow, M.; Erkey, C. Supported Platinum Nanoparticles by Supercritical Deposition. Ind. Eng. Chem. Res. 2005, 44, 4161. (26) Zhang, Y.; Erkey, C. Preparation of Platinum-Nafion-Carbon Black Nanocomposites via a Supercritical Fluid Route as Electrocatalysts for Proton Exchange Membrane Fuel Cells. Ind. Eng. Chem. Res. 2005, 44, 5312. (27) Zhang, Y.; Kang, D.; Aindow, M.; Erkey, C. Preparation and Characterization of Ruthenium/Carbon Aerogel Nanocomposites via a Supercritical Fluid Route. J. Phys. Chem. B 2005, 109, 2617. (28) Haji, S.; Zhang, Y.; Kang, D.; Aindow, M.; Erkey, C. Hydrodesulfurization of Model Diesel using Pt/Al2O3 Catalysts Prepared by Supercritical Deposition. Catal. Today 2005, 99, 365. (29) Tavana, A.; Chang, J.; Randolph, A. D.; Rodriguez, N. Scanning of Cosolvents for Supercritical Fluids Solubilization of Organics. AIChE J. 1989, 35, 645. (30) Ryoo, R.; Jun, S. Improvement of Hydrothermal Stability of MCM41 using Salt Effects During the Crystallization Process. J. Phys. Chem. B 1997, 101, 317. (31) Warzinski, R. P. Supercritical-Fluid Solubilization of Catalyst Precursors: The Solubility and Phase Behavior of Molybdenum Hexacarbonyl in Supercritical Carbon Dioxide and Application to the Direct Liquefaction of Coal. J. Supercrit. Fluids 1992, 5, 60. (32) Houalla, M.; Broderick, D. H.; Sapre, A. V.; Naga, N. K.; de Beer, V. H. J.; Gates, B. C.; Kwart, H. Hydrodesulfurization of Methyl-Substituted Dibenzothiophenes Catalyzed by Sulfided CoMo/γ-Al2O3. J. Catal. 1980, 61, 523. (33) Lanure-Meille, V.; Schulz, E.; Lemaire, M.; Vrinat, M. Effect of Experimental Parameters on the Relative Reactivity of Dibenzothiophene and 4-Methyldibenzothiophene. Appl. Catal., A 1995, 131, 143. (34) Sie, S. T. Miniaturization of Hydroprocessing Catalyst Testing Systems: Theory and Practice. AlChE J. 1996, 42, 3498. (35) Bej, S. K.; Dabral, R. P.; Gupta, P. C.; Mittal, K. K.; Sen, G. S.; Kapoor, V. K.; Dalai Ajay, K. Studies on the Performance of a Microscale Trickle Bed Reactor using Different Sizes of Diluents. Energy Fuels 2000, 14, 701. (36) Shimada, H.; Sato, T.; Yoshimura, Y.; Hiraishi, J.; Nishijima, A. Support Effect on the Catalytic Activity and Properties of Sulfided Molybdenum Catalysts. J. Catal. 1988, 110, 275. (37) Yue, Y.; Sun, Y.; Gao, Z. Disordered Mesoporous KIT-1 as a Support for Hydrodesulfurization Catalysts. Catal. Lett. 1997, 47, 167. (38) Silva-Rodrigo, R.; Caldero’n-Salas, C.; Melo-Banda, J. A.; Domn´guez, J. M.; Vaz´quez-Rodrg´uez, A. Synthesis, Characterization and Comparison of Catalytic Properties of NiMo- and NiW/Ti-MCM-41 Catalysts for HDS of Thiophene and HVGO. Catal. Today 2004, 98, 123.

4292 Ind. Eng. Chem. Res., Vol. 48, No. 9, 2009 (39) Herrera, J. M.; Reyes, J.; Roquero, P.; Klimova, T. New Hydrotreating NiMo Catalysts Supported on MCM-41 Modified with Phosphorus. Microporous Mesoporous Mater. 2005, 83, 283. (40) Franke, O.; Rathousky, J.; Schultz-Ekloff, G.; Zukal, A. Synthesis of MCM-41 Mesoporous Molecular Sieves. Stud. Surf. Sci. Catal. 1995, 91, 309. (41) Zepeda, T. A.; Halachev, T.; Pawelec, B.; Nava, R.; Klimova, T.; Fuentes, G. A.; Fierro, J. L. G. Hydrodesulfurization of Dibenzothiophene over CoMo/HMS and CoMo/Ti-HMS Catalysts. Catal. Commun. 2006, 7, 33. (42) Noronha, F. B.; Baldanza, M. A. S.; Schmal, M. CO and NO Adsorption on Alumina-Pd-Mo Catalysts: Effect of the Precursor Salts. J. Catal. 1999, 188, 270. (43) Abello, M. C.; Gomez, M. F.; Ferreti, O. Mo/γ-Al2O3 Catalysts for the Oxidative Dehydrogenation of Propane: Effect of Mo Loading. Appl. Catal., A 2001, 207, 421. (44) Jacobs, J. L.; Zhang, G. Y.; Das, T.; Davis, B. H. Fischer-Tropsch Synthesis: Effect of Small Amounts of Boron, Ruthenium and Rhenium on Co/TiO2 Catalysts. Appl. Catal., A 2002, 223, 195. (45) Lo´pez Cordero, R.; Gil Llambias, F. J.; Lo´pez Agudo, A. Temperature-Programmed Reduction and Zeta Potential Studies of the Structure of Mo/O3Al2O3 and Mo/O3SiO2 Catalysts Effect of the Impregnation pH and Molybdenum Loading. Appl. Catal. 1991, 74, 125. (46) Damyanova, S.; Spojakina, A.; Jiratova, K. Effect of Mixed TitaniaAlumina Supports on the Phase Composition of NiMo/TiO2Al2O3 Catalysts. Appl. Catal., A 1995, 125, 257.

(47) Guimon, C.; El Horr, N.; Romero, E.; Monzon, A. Characterization of the Active Sites of Ni-Si-Al Sol-Gel Hydrogenation Catalysts. Stud. Surf. Sci. Catal. 2000, 130, 3345. (48) Zhang, X.; Yu, G.; Gong, Y.; Jiang, D.; Xie, Y. Oxidative Dehydrogenation of Ethane to Ethylene over NiO/Al2O3 Catalyst. Stud. Surf. Sci. Catal. 2000, 130, 1835. (49) Landau, M. V.; Berger, D.; Herskowitz, M. Hydrodesulfurization of Methyl-Substituted Dibenzothiophenes: Fundamental Study of Routes to Deep Desulfurization. J. Catal. 1996, 159, 236. (50) Ho, T. C.; Sobel, J. E. Kinetics of Dibenzothiophene Hydrodesulfurization. J. Catal. 1991, 128, 581. (51) Robinson, W. R. A. M.; van Veen, J. A. R.; de Beer, V.; H, J.; Van Santen, R. A. Development of Deep Hydrodesulfurization Catalysts: II. NiW, Pt and Pd Catalysts Tested with (Substituted) Dibenzothiophene. Fuel Process. Technol. 1999, 61, 103. (52) Jongpatiwut, S.; Li, Z.; Resasco, D. E.; Alvarez, W. E.; Sughrue, E. L.; Dodwell, G. W. Competitive Hydrogenation of Poly-Aromatic Hydrocarbons on Sulfur-Resistant Bimetallic Pt-Pd Catalysts. Appl. Catal., A 2004, 262, 241.

ReceiVed for reView September 25, 2008 ReVised manuscript receiVed February 28, 2009 Accepted February 28, 2009 IE801442C

Al-HMS Nanocatalyst

Recommend Documents