Development of deep hydrodesulfurization catalysts: II. NiW, Pt and Pd catalysts tested with (substituted) dibenzothiophene
Introduction
In the previous study [1], the performance of CoMo and NiMo catalysts was studied in the desulfurization of (un)substituted dibenzothiophenes (DBT). It was found that the main reaction path was direct hydrogenolysis, a route which is impaired by the presence of substituents, especially when both at the 4- and 6-position. Therefore, the activity of the catalysts for 4MDBT (4-methyl dibenzothiophene) and especially 4E6MDBT HDS (4-ethyl, 6-methyl dibenzothiophene hydrodesulfurization) was clearly lower than for DBT HDS. It was pointed out that appreciable hydrogenation, dealkylation or C–C bond scission activity may lead to more facile pathways for 4E6MDBT HDS. With CoMo and NiMo catalysts, only a small increase in hydrogenation activity could be realized by the use of amorphous silica–alumina (ASA) as a support. In the present study, we will test NiW and noble metal catalysts, which are known for their HDS activity as well as high(er) hydrogenation activity 2, 3. Supports with varying acidity will be used to influence the activity for dealkylation or C–C bond scission. Special attention will be paid to the sensitivity of these catalysts to the presence of H2S in the feed gas. By means of feed flow variation and hydrogenation experiments, information will be obtained about the 4E6MDBT HDS reaction network.
Section snippets
Catalyst preparation
The following supports were used in catalyst preparation (0.25–0.50 mm particles).
Activated carbon (Norit RX3 Extra: pore volume (P.V.) 1.0 ml/g, surface area (S.A.) 1200 m2/g), γ-Al2O3 (Ketjen 001–1.5E: (P.V.) 0.60 ml/g, (S.A.) 276 m2/g), ASA (ex Shell: Si/Al ratio 0.70 at/at, (P.V.) 0.72 ml/g, (S.A.) 389 m2/g), as well as a more acidic variation of the latter ASA, referred to as ASA+.
The NiW/Al2O3, NiW/ASA and NiW/ASA+ catalysts were all obtained from Delft University [4]. They were prepared
DBT HDS
Fig. 1 shows the DBT HDS activities, calculated from experiments at 573 K. For the NiW catalysts, the order in activity is NiW/ASA<NiW/ASA+<NiW/Al2O3; all three are less active than the NiMo/Al2O3 catalysts tested previously [1]. The activity of Pt/ASA is comparable with that of the NiW catalysts; Pt/Al2O3 is clearly less active.
The three NiW catalysts show approximately the same H2S sensitivity: at a level of 0.2% H2S, their activities are all reduced by a factor of six to seven. The Pt
NiW catalysts
The performance of the NiW catalysts is clearly dependent on the support. In 4E6MDBT, 4MDBT and DBT HDS at 573 K (no H2S added), the differences in activity are fairly modest and the order is usually Al2O3>ASA>ASA+. However, the amount of C7–C8 or C6–C7 products increases considerably in the reverse order. From feed flow variation experiments, it appeared that the latter products are formed from (substituted) BP and/or CHB's in a consecutive reaction, possibly on the acidic support. A different
Conclusions
The HDS activity rankings for the catalysts studied appeared quite dependent on the type of reactant. The degree of steric hindrance in the desulfurization reaction increases in the order DBT<4-MDBT<4E6MDBT; therefore, activity for 4MDBT and especially 4E6MDBT HDS is favoured by a high hydrogenation activity, providing an easier reaction pathway. The activity trends, measured without additional H2S (CoMo, NiMo: 4E6MDBT≪4MDBT<DBT; NiW: 4E6MDBT<4MDBT≈DBT; Pt: 4E6MDBT≈4MDBT≈DBT) are therefore
References (8)
- et al.
Appl. Catal.
(1982) - et al.
J. Catal.
(1996) - et al.
J. Catal.
(1994) - W.R.A.M. Robinson, J.A.R. van Veen, V.H.J. de Beer, R.A. van Santen, preceding...