versión impresa ISSN 0122-5383
C.T.F Cienc. Tecnol. Futuro v.4 n.2 Bucaramanga jul./dic. 2010
EVALUATING THE FUNCTIONALITIES OF NiMo/γ-Al2O3B2O3 CATALYSTS IN NAPHTHALENE HYDRODEAROMATIZATION AND DIBENZOTHIOPHENE HYDRODESFULFURIZATION
Víctor-G. Baldovino-Medrano1, Aristóbulo Centeno1 and Sonia-A. Giraldo1*
1Centro de Investigaciones en Catálisis (CICAT), Escuela de Ingeniería Química, Universidad Industrial de Santander (UIS), Santander, Bucaramanga, Colombia
(Received, Jan. 29, 2010; Accepted, Nov. 30, 2010)
* To whom correspondence may be addressed
The aim of this work is to contribute to the current understanding on the role of the support’s acidic properties in the hydrogenating function of NiMo/γ-Al2O3 type catalysts during hydrodearomatization (HDA) and dibenzothiophene (DBT) type molecules desulfurization. NiMo/γ-Al2O3-B2O3 catalysts of different B2O3 (0, 2, 3, 6 and 8 wt.%) contents were prepared and tested in independent and simultaneous naphthalene (NP) HDA and DBT hydrodesulfurization (HDS) reactions. For HDA the catalytic activity as a function of the B2O3 content followed a volcano-shape trend, with a maximum around 3 wt.% of B2O3. In DBT desulfurization boron was found to have a positive effect in the development of the HYD route of desulfurization possibly due to an increase in total acidity. Conversely, the direct desulfurization route (DDS) was negatively affected by boron addition. The presence of NP during the HDS of DBT was found to have a significant effect in neither total HDS activity nor the HYD/DDS selectivity. The findings in this paper are significant for ultra-deep HDS of heavy oil cuts where increasing in the selectivity to HYD is a must because highly refractory alkyl-DBTs mostly react by this reaction route.
Keywords: NiMo/γ-Al2O3, hydrotreatement, direct desulfurization route, HYD function.
El objetivo de este trabajo es contribuir a entender el papel de la acidez de los soportes sobre la función hidrogenante de catalizadores tipo NiMo/γ-Al2O3 durante la hidrodesaromatización (HDA) y la desulfuración de moléculas tipo dibenzotiofeno (DBT). Se prepararon catalizadores NiMo/γAl2O3-B2O3 con diferentes contenidos de B2O3 (0, 2, 3, 6 y 8% en peso) y se ensayaron independiente y simultáneamente en la HDA del naftaleno (NF) y la hidrodesulfuración (HDS) del DBT. La actividad catalítica en HDA en función del contenido de B2O3 siguió una curva tipo volcán con un máximo alrededor de 3% de B2O3. En la HDS del DBT, se encontró que el boro tiene un efecto positivo sobre el desarrollo de la ruta de hidrogenación (HID) de la desulfurización, posiblemente debido al aumento en la acidez total. Por el contrario, la ruta de desulfurización directa (DDS) fue negativamente afectada por la adición de boro. También se encontró que la presencia de NF durante la HDS de DBT no tiene un efecto significativo sobre la actividad total de HDS ni sobre la selectividad HID/DDS. Los hallazgos realizados en este estudio son importantes para los procesos de desulfurización profunda de cortes pesados de refinería; en los cuales se requiere aumentar la selectividad hacia la ruta HID que es la vía de reacción de las moléculas tipo alquil-DBTs que son altamente refractarias al proceso.
Palabras clave: NiMo/γ-Al2O3, ruta de desulfurización directa, hidrotratamiento, función hidrogenante.
Burning of fossil fuels produces pollutants such as CO2, CO, SOx and NOx. Nevertheless, the world’s energy global market is sustained by this non-renewable resource. To lessen the environmental cost of fossil fuels consumption stringent governmental legislation has been imposed particularly on their sulfur and aromatic contents. Both the US Environmental Protection Agency (EPA, 2008) and the European Directive (1998) have fixed a 10 ppm concentration limit for the sulfur content in diesel oil for 2008-2009, namely Ultra-Low Sulfur Diesel (ULSD). ULSD requires either the development of new refining units combined with more active desulfurization-hydrogenation catalysts (Leliveld & Eijsbouts, 2008). The benefit of high hydrogenating selective catalysts is that their use allows sulfur removal from sterically hindered 4,6-dimethyl-dibenzothiophene (4,6-DMDBT) and parent molecules (Pérot, 2003). Moreover, parallel saturation of aromatic rings (HDA) and hydrodenitrogenation (HDN) can be highly competitive hydrogenation (HYD) reactions (Leliveld & Eijsbouts, 2008). It has been recognized that the acidic function of sulfided CoMo and NiMo/γ-Al2O3 catalysts plays a key role in the hydrodesulfurization (HDS) of refractory 4,6-DMDBT (Pérot, 2003).Asimple and economic way to increase the acidity of the alumina support consists in boron impregnation, to obtain mixed γ-Al2O3-B2O3 type oxides (Sibeijn, Vanveen, Bliek, & Moulijn, 1994; Usman, Takaki, Kubota, & Okamoto, 2005; Torres-Mancera, Ramírez, Cuevas, Gutiérrez-Alejandre, Murrieta, & Luna, 2005; Lewandowski & Sarbak, 2000; Sato, Kuroki, Sodesawa, Nozaki, & Maciel, 1995). By means of this procedure, Torres-Mancera et al. (2005) prepared CoMo and NiMo/γ-Al2O3-B2O3 catalysts exhibiting good activity in the HDS of 4,6-DMDBT. This effect is related to the development of the catalyst’s ability to isomerize the methyl substituents of the 4,6-DMDBT molecule, surpassing steric hindrance, (Pérot, 2003; Torres-Mancera et al., 2005).
Though a huge amount of studies have been carried out to understand the relationship between the acidic function and the isomerization and cracking functionalities of HDT catalysts, the relationship between acidity and the selectivity either to the direct desulfurization (DDS) or HYD route of desulfurization is not fully understood (Pérot, 2003). In the case of γ-Al2O3-B2O3 supports such relationship is more complicated to establish because as boron loading increases catalyst’s acidity also increases, but due to structural changes of the boron oxide species the Co(Ni)-Mo-alumina interaction is affected modifying the active sites of the sulfided phases (Usman et al., 2005; Torres-Mancera et al., 2005; Lewandowski & Sarbak 2000; Sato et al., 1995; Li, Sato, Imamura, Shimada & Nishijima, 1998; Li, Sato, Imamura, Shimada, & Nishijima, 1997; Usman, Kubota, Hiromitsu, & Okamoto, 2007; Ferdous, Dalai, & Adjaye, 2006). Though the structural changes induced by B on the active phase of sulfided CoMo and NiMo have been studied with certain detail (Usman et al., 2005; Li et al., 1998; Usman et al., 2007), few reports have dealt with B effect on the functionalities of such catalysts in HDA and HDS. Furthermore, contradictory results have been presented mainly due to differences in the reaction conditions used in each of these studies (Lewandowski & Sarbak, 2000; Ding, Zhang, Zheng, Ring, & Chen, 2006).
This work aims to contribute to the current understanding on the role of the support’s acidic properties in the HYD function of NiMo/γ-Al2O3-B2O3 type catalysts during aromatics HDA and dibenzothiophene (DBT) type molecules desulfurization. DBT was chosen as a model molecule instead of 4,6-DMDBT because it allows a direct measurement of HYD/DDS selectivity as referred to the removal of the sulfur heteroatom from the thiophenic ring against DBT aromatic backbone saturation (Pérot, 2003). Both molecules share similar desulfurization reaction pathways (Mijoin, Pérot, Bataille, Lemberton, Breysse, & Kasztelan, 2001). Moreover, parallel cracking and isomerization reactions are ruled out. Reactions conditions were selected to avoid thermodynamic constraints in either the HDA or the HYD route of desulfurization of DBT, (Cooper & Donnis, 1996; Ho, 2004).
A series of γ-Al2O3-B2O3 supported NiMo catalysts of various B2O3 contents were prepared by the sequential incipient-wetness method. Procatalyse alumina, Dp = 0,3-0,6 mm, was calcined in air flow at 773 K for 4 h before boron impregnation. Alumina support was first impregnated with a H3BO3 solution, followed by sequential impregnation of Mo and Ni precursors, respectively. The precursors used were ammonium heptamolybdate (NH4)6Mo7O24•4H2O (Merck), and niquel nitrate Ni(NO3)2•6H2O (Aldrich). The amounts of each precursor were calculated from the desired contents of MoO3, 9 wt.%, and NiO, 4,5 wt.%. After each impregnation step, the solids were aged by 24 h. Afterwards, the impregnated solids were dried (393 K for 12 h) and calcined (773 K for 4 h) in air flow. B2O3 nominal contents were: 0, 2, 3, 6 and 8 wt.%. Catalysts were labeled as NiMo-B(x); where, x referred to the B2O3 nominal content.
All of the catalysts (approximately 0,5 g) were in situ activated with a gaseous mixture of H2S (15 vol.%) in H2, at 673 K and atmospheric pressure, during 3 h. The activation mixture flow was (100 mL/min) kept until reaching reaction temperature and beginning H2 pressurization of the reaction system.
BET surface area (SBET), pore volume (VP) and average pore diameter (DP) were measured by the conventional nitrogen adsorption-desorption technique in a NOVA 1200 (Quantachrome) apparatus.
NH3 TPD analysis was used to determine total acidity of the catalysts. A Chembet 3000 (Quantachrome) apparatus was used. Previous to each test, catalyst activation was performed at the same conditions mentioned in the precedent section. Activation was followed by N2 evacuation (15 min) at 373 K. NH3 was adsorbed at this temperature by flowing it during 15 min. Once adsorption step was completed, the evacuation of the system was carried out for 2 h. NH3 TPD analysis was performed by heating the sample (at 10 K/min) until 773 K under N2 flow and, then, collecting desorbed ammonia in a 0,4 vol.% H3BO3 solution. The amount of desorbed NH3 (μeq NH3/g cat.) was calculated by titrating with H2SO4.
Catalytic tests were performed in a continuous highpressure fixed-flow reactor. The volume of the catalytic bed was c.a. 2 mLand it was composed of 0,5 g of catalyst diluted in borosilicate glass-beads. Three types of catalytic tests were performed: (i) naphthalene (NP) HDA, under an H2S atmosphere, (ii) DBT hydrodesulfurization, and (iii) simultaneous DBT hydrodesulfurization and naphthalene HDA. The composition of the liquid feed was 3 wt.% of NP and/or 2 wt.% of DBT respectively, and 2 wt.% of hexadecane, as GC internal standard, diluted in ciclohexane. To generate the H2S atmosphere during the HDA tests, dimethyldisulfide was added to the liquid feed in such a concentration as to obtain 4074 wppm of H2S. Reaction conditions were: T = 563 K for HDA, and T = 583 K for the HDS and the simultaneous HDS and HDA, P = 5 MPa, liquid feed flow of 30 mL/h and an H2/liquid feed flow ratio of 500 NL/L. Under such conditions, the absence of diffusion limitations was verified. Reaction products were identified using an HP 6890 GC, provided with an FID detector and an HP-1 (100 m x 0,25 mm x 0,5 μm) column. All catalytic tests were conducted until steady state (approximately 5 h). The catalytic activity was expressed as the fraction of reactant conversion (X) and as the yield (yj) of products (Equation 1):
Where nj are the moles of different reaction products: cyclohexylbenzene (CHB) and biphenyl (BP) and nDBT,0 are the feed moles of DBT.
3. RESULTS AND DISCUSSION
Table 1 shows the textural characteristics and total acidity of the prepared catalysts.
The results of the textural characterization show that the effect of boron on the catalysts SBET depends on the boron concentration, without causing significant changes in the porous structure (Vp and Dp). As compared to the NiMo catalyst, an increase in SBET is observed for the catalysts containing an amount of B2O3 up to 3 wt.% whereas a decrease in SBET is observed for those with higher B2O3 wt.%. Usman et al. (2005) showed that at low boron concentration, B2O3 is well dispersed on the alumina support without significantly changing its textural properties. Torres-Mancera et al. (2005) ascribed the effect of boron on the support’s textural properties to a partial dissolution of alumina during H3BO3 impregnation. Thus, at low boron concentrations, SBET increases whereas at higher boron load, the boron oxide causes a blocking of the pores.
The trend registered for total acidity (Table 1) agrees with previous reports (Sibeijn et al., 1994; Usman et al., 2005; Torres-Mancera et al., 2005; Lewandowski & Sarbak, 2000; Sato et al., 1995; Li et al., 1998). There is an increase in catalyst’s total acidity after boron incorporation. It has been demonstrated that such acidity increase is due to the generation of Brønsted acidic sites. In addition, it has been observed that the amount of Lewis acidic sites slightly increase at B2O3 concentrations up to 3 wt.%. Such Lewis sites tend to decrease and disappear at B2O3 concentrations higher than 5 wt.%. Sibeijn et al. (1994) showed that boron does not bond to alumina Lewis sites, yet it rather links to its OH- groups. Due to this observation, significant changes in the structure of the sulfided catalyst take place (Usman et al., 2005 and 2007), as it will be discussed later. Sato et al. (1995) studied the acidity of γ-Al2O3-B2O3 supports by pyridine TPD. Their results indicate that the amount and strength of Brønsted acidic sites of such materials is related to the presence of BO4 species, which increases with the load of boron. Similar trends have been reported for NiMo/γ-Al2O3-B2O3 catalysts (Lewandowski & Sarbak, 2000). Lewandowski and Sarbak (2000) found, by means of model reactions, that boron addition to NiMo/γ-Al2O3 catalysts leads to the formation of Brønsted acidic sites of intermediate strength. The above mentioned evidence leads to believe that the increase in the total acidity of the NiMo-B(x) catalysts due to boron incorporation can present two main zones of distribution of acidic sites: at low B2O3 concentrations there is anequilibrium between the numberof Brønsted andLewis acidic sites, and at higher boron concentrations, Lewis acidic sites tend to disappear and, thus mostly Brønsted sites are present on the catalyst’s surface. Such Brønsted sites would possess an intermediate acidic strength, which increases with the amount of BO4 species.
Figure 1 shows the effect of boron concentration in the catalytic performance in naphthalene HDA along with the change in the total acidity of NiMo-B(x). It is observed that the promoting effect of boron in HDA is a function of the B2O3 content of the catalysts. In general, a volcano-shape plot is observed. For B2O3 contents up to 6 wt.% there is an activity increase compared to the NiMo catalyst, whereas for a content of 8 wt.% no significant change is registered. The observed trend has been reported before. Li et al. (1998) reported a volcano-type behavior for the HDA of 1-methylNP for NiMo/γ-Al2O3-B2O3 catalysts. The corresponding maximum was around 1 wt.% of B2O3. They ascribed this behavior to a better dispersion of the oxidic NiO and MoO3 precursors at low boron concentration. Other authors have established that the dispersion and the structure of the active phase of Al2O3-B2O3 supported HDT catalysts is affected by the B2O3 concentration due to changes induced by the conformation of borate ions present in the alumina surface (Sibeijn et al., 1994; Usman et al., 2005). Sato et al. (1995) proved the existence of tetrahedral BO4 monomeric species over the alumina surface at low boron concentrations. Sibeijn et al. (1994) showed that the B2O3 is attached to the hydroxyl groups of Al2O3 and after saturation B2O3 polymeric species are formed. Such species tend to form a monolayer at the alumina surface. The presence of these polymeric species reduces the interaction between Mo and the alumina support, thus increasing the size of the Mo oxide clusters. As a result, a less dispersed MoS2 active phase is formed after sulfidation (Usman et al., 2005). Usman et al. (2007) assuming the Co-Mo-S active phase proposed by Topsøe (2007), suggested a change in the structure of the Co-Mo-S mixed sulfide phase due to a decrease in the number of Co atoms decorating the edges of the MoS2 phase. Nevertheless, these authors could not correlate such efeffect to the catalytic performance of CoMo/Al2O3-B2O3 catalysts in HDT reactions. On the other hand, Li et al. (1997) used XRD, XPS and EXAFS to characterize NiMo/γ-Al2O3-B2O3 catalysts and showed that the main effect of boron is to modify the dispersion of the MoS2 active phase and the Ni, as well as the textural properties of the catalysts. Considering such evidence, it can be said that the HDA trend registered results from a combination of two main effects induced by boron addition to the alumina support. The first would be the conformational structure of B2O3 which modifies the dispersion of the MoS2 active phase and Ni, and the second corresponds to the acidity increase of the catalysts. In this last regard, it seems that not only acidity increase is important but also the distribution and strength of Brønsted and Lewis acidic sites (Figure 1). Considering the work of Sato et al. (1995) it can be speculated that along with the appropriated dispersion of B2O3, MoS2 and Ni, the best catalytic performance of NiMo/γ-Al2O3-B2O3 in HDA reactions is related to an appropriated balance between the relative concentration of Brønsted and Lewis acidic sites of intermediate acidic strength.
Figure 2 shows the influence of B content in the behavior of NiMo-B(x) catalysts in the HDS of DBT as well as over the total acidity. As observed, boron addition at low concentrations does not significantly impact DBT conversion, but at higher boron concentration HDS activity decreases. Other authors have reported similar trends (Lewandowski & Sarbak, 2000; Ferdous et al., 2006). Lewandowski and Sarbak (2000) found that boron addition to NiMo catalysts did not affected the HDS activity of liquid carbon. Ferdous et al. (2006) investigated the effect of the B2O3 concentration in NiMo/Al2O3 catalysts in the HDS of heavy gas-oil, without detecting any effect. A volcano type behavior has been reported for HDS in agreement with the trends registered in HDA in the present work. However, the activity increase reported in such references is not as high as the one reported here (Li et al. 1997 and 1998). Therefore, the overall changes in HDS activity can be ascribed mainly to changes in the dispersion of the MoS2 active phase.
It is much more interesting to analyze the changes in reaction product distribution during DBT hydrodesulfurization as presented in Figure 2. Under the present reaction conditions, biphenyl (BP), resulting from the DDS route, and cyclohexylbenzene (CHB), from the HYD reaction route, were mainly the only detected products. Partially hydrogenated tetrahydro-DBT and hexahydro-DBT intermediates were detected only in very small traces. It is assumed here that no significant further conversion of BP to CHB takes place during DBT hydrodesulfurization (Mijoin et al., 2001). It is observed that boron incorporation has a negative effect in DDS. Conversely, HYD selectivity increases to some extent. Such opposite trends confirm that BP is not being hydrogenated to CHB. Contrary to the present results, Li et al. (1997 and 1998) found an increase in the conversion to BP, with a simultaneous decrease to CHB with the increase in the boron concentration of NiMo/γ-Al2O3-B2O3 catalysts. The differences between their results and ours can be ascribed to the higher temperature employed by those authors (733 K). Under such conditions, thermodynamic limitations for HYD reaction have been predicted (Cooper & Donnis, 1996; Ho, 2004). On the other hand, the trend observed in Figure 2 agrees with the hypothesis that an increase in Brønsted acidic sites of the support of MoS2 based catalysts favors the HYD route of desulfurization (Pérot, 2003). Comparatively, the increase in HYD selectivity, as a function of acidity and B2O3 content, is not as high as the one obtained in HDA. This can be ascribed to the lower aromaticity of the fused rings of NP, as compared with those of DBT. This makes the latter more refractory to HYD (Cooper & Donnis, 1996). Therefore, it is also likely that the development of the HYD route of desulfurization of DBT type molecules over conventional MoS2 based catalysts is more related to the presence of Brønsted sites of higher acidic strength compared to those required for NP hydrogenation.
Simultaneous dibenzothiophene HDS and naphthalene HDA
According to the catalytic results in HDS and HDA as well as to the measured acidic properties for the series of NiMo/γ-Al2O3-B2O3 catalysts prepared, the catalyst labeled as NiMo-B(6) was chosen to perform this test. Figure 3 shows the evolution of the catalytic performance with time on stream for the simultaneous HDS and HDA reaction. The results show only slight differences with those registered for the independent tests. This behavior is in agreement with the generally accepted idea of the existence of different active sites for HDS and HYD (Topsøe, 2007; Grange & Vanhaeren, 1997). Besides, it should be noticed that there are no changes in the selectivity to the desulfurization pathway of DBT. The HYD/DDS selectivity is related to the differences in the adsorption mode of the DBT molecule over the MoS2 active phase (Cristol, Paul, Payen, Bougeard, Hutschka, & Clémendot, 2004; Nag, 1984; Egorova & Prins, 2004). Cristol et al. (2004) theoretically showed that during DDS, the DBT molecule is linked to the coordination unsaturated sites (CUS) of the MoS2 active phase via direct sulfur η1S adsorption (or σ-mode). Conversely, flat π-mode of adsorption of one of the aromatic rings of DBT is required to develop HYD route. The formed π-complex between the benzene ring of DBT and the CUS of MoS2 leads to a hydrogenation-dehydrogenation-sulfur atom bond scission equilibrium resulting in CHB formation (Mijoin et al. 2001; Baldovino-Medrano, Eloy, Gaigneaux, Giraldo, & Centeno, 2009). In agreement with Cristol et al. (2004) and Egorova and Prins (2004) proposed that the C-S-C bond scission step in HYD proceeded only after desorption and readsorption of the partially hydrogenated intermediates over the same active site.
The results of this work are pertinent for deep HDT. In that case, the effect of parallel HDN reactions must also be considered. Works in this sense will be performed in the near future.
4. CONCLUSIONSThe main conclusions of this work are:
- The activity of the NiMo/γ-Al2O3-B2O3 catalysts in HDA follows a volcano-type trend as a function of the boron content. The reason of this behavior is related to changes in the dispersion of the MoS2 active phase, the textural characteristics of the support, and the proportion between the Brønsted and Lewis acidic sites generated by the addition of boron.
- Boron addition caused a decrease in the HDS of DBT activity, particularly at higher boron contents. It was determined that boron presence slightly increases conversion via HYD but negatively affects the DDS route of desulfurization. The latter effect is coherent with an increase in catalysts’ acidity.
- No direct correlation between naphthalene hydrogenation and the HYD pathway was determined.
- A comparison of the results of the independent and simultaneous HDA and HDS catalytic tests indicated that the presence of naphthalene does not particularly hinder the development of the HYD route of desulfurization over NiMo/γ-Al2O3-B2O3.
ACKNOWLEDGEMENTSThis work was possible due to the financial support of Vice rectory for Research and Extension-Universidad Industrial de Santander (UIS), in the frame of the project 5435.Víctor Baldovino-Medrano thanks COLCIENCIAS and UIS for financial support.Authors especially express their gratitude to Engineers L. González and K. Rojas, who carried out some of the presented experiments.
Baldovino-Medrano, V. G., Eloy, P., Gaigneaux, E. M., Giraldo, S.A., & Centeno, A. (2009). Development of the HYD route of hydrodesulfurization of dibenzothiophenes over Pd-Pt/γ-Al2O3 catalysts. J. Catal., 267: 129-139.
Cooper, B. H., & Donnis, B. B. L. (1996). Aromatic saturation of distillates: An overview. Appl. Catal. A: Gen., 137 (2): 203-223. [ Links ]
Cristol, S., Paul, J.-F., Payen, E., Bougeard, D., Hutschka, F., & Clémendot, S. (2004). DBT derivatives adsorption over molybdenum sulfide catalysts: A theoretical study. J. Catal., 224 (1): 138-147. [ Links ]
Ding, L., Zhang, Z., Zheng, Y., Ring, Z., & Chen, J. (2006). Effect of fluorine and boron modification on the HDS, HDN and HDA activity of hydrotreating catalysts. Appl. Catal. A: Gen., 301 (2): 241-250. [ Links ]
Egorova, M., & Prins, R. (2004). Hydrodesulfurization of dibenzothiophene and 4,6-dimethyldibenzothiophene over sulfided NiMo/γ-Al2O3, CoMo/γ-Al2O3, and Mo/γAl2O3 catalysts. J. Catal., 225 (2): 417-427.
European Directive (1998). The Directive 98/70/EC is amended by the European Directive 2003/17/EC of 3 March 2003. [ Links ]
Ferdous, D., Dalai, A. K., & Adjaye, J. (2006). Hydrodenitrogenation and hydrodesulfurization of heavy gas oil using NiMo/Al2O3 catalyst containing boron: Experimental and kinetic studies. Ind. Eng. Chem. Res., 45 (2): 544-552. [ Links ]
Grange, P., & Vanhaeren, X. (1997). Hydrotreating catalysts, an old story with new challenges. Catal. Today, 36 (4): 375-391. [ Links ]
Ho, T. C. (2004). Deep HDS of diesel fuel: chemistry and catalysis. Catal. Today, 98 (1-2): 3-18. [ Links ]
Leliveld, R. G., & Eijsbouts, S. E. (2008). How a 70-yearold catalytic refinery process is still ever dependent on innovation. Catal. Today, 130 (1): 183-189. [ Links ]
Lewandowski, M., & Sarbak, Z. (2000). Effect of boron addition on hydrodesulfurization and hydrodenitrogenation activity of NiMo/Al2O3 catalysts. Fuel, 79 (5): 487-495. [ Links ]
Li, D., Sato, T., Imamura, M., Shimada, H., & Nishijima, A. (1997). Spectroscopic characterization of Ni-Mo/γAl2O3-B2O3 catalysts for hydrodesulfurization of dibenzothiophene. J. Catal., 170 (2): 357-365.
Li, D., Sato, T., Imamura, M., Shimada, H., & Nishijima, A. (1998). The effect of boron on HYD, HC and HDS activities of model compounds over Ni-Mo/γ-Al2O3-B2O3 catalysts. Appl. Catal. B: Env., 16 (3): 255-260.
Mijoin, J., Pérot, G., Bataille, F., Lemberton, J. L., Breysse, M., & Kasztelan, S. (2001). Mechanistic considerations on the involvement of dihydrointermediates in the hydrodesulfurization of dibenzothiophene-type compounds over molybdenum sulfide catalysts. Catal. Lett., 71 (3-4): 139-145. [ Links ]
Nag, N. K. (1984). On the mechanism of the hydrogenation reactions occurring under hydroprocessing conditions. Appl. Catal. A: Gen., 10 (1): 53-62. [ Links ]
Pérot, G. (2003). Hydrotreating catalysts containing zeolites and related materials -Mechanistic aspects related to deep desulfurization. Catal. Today, 86 (1-4): 111-128. [ Links ]
Sato, S., Kuroki, M., Sodesawa, T., Nozaki, F., & Maciel, G. E. (1995). Surface structure and acidity of alumina-boria catalysts. J. Mol. Catal. A: Chem., 104 (2): 171-177. [ Links ]
Sibeijn, M., Vanveen, J. A. R., Bliek, A., & Moulijn, J. A. (1994). On the Nature and Formation of the Active Sites in Re2O7 Metathesis Catalysts Supported on Borated Alumina. J. Catal., 145 (2): 416-428. [ Links ]
Topsøe, H. (2007). The role of Co-Mo-S type structures in hydrotreating catalysts. Appl. Catal. A: Gen., 322 (suppl.): 3-8. [ Links ]
Torres-Mancera, P., Ramírez, J., Cuevas, R., Gutiérrez-Alejandre, A., Murrieta, F., & Luna, R. (2005). Hydrodesulfurization of 4,6-DMDBT on NiMo and CoMo catalysts supported on B2O3-Al2O3. Catal. Today, 107-108: 551-558. [ Links ]
Usman, U., Kubota, T., Hiromitsu, I., & Okamoto, Y. (2007). Effect of boron addition on the surface structure of CoMo/Al2O3 catalysts. J. Catal., 247 (1): 78-85. [ Links ]
Usman, U., Takaki, M., Kubota, T., & Okamoto, Y. (2005). Effect of boron addition on a MoO3/Al2O3 catalyst Physicochemical characterization. Appl. Catal. A: Gen., 286 (1): 148-154. [ Links ]