Investigation of 4,6-dimethyldibenzothiophene hydrodesulfurization over a highly active bulk MoS2 catalyst

Article abstract of DOI:10.1016/j.jcat.2004.05.001

The hydrodesulfurization (HDS) of 4,6-dimethyldibenzothiophene was investigated in a batch-stirred microautoclave reactor over a laboratory-synthesized MoS₂ catalyst. The reaction was found to proceed through two main parallel pathways: direct desulfurization (DDS) and hydrogenation (HYD). The effect of a wide range of H₂S concentrations on the HDS activity and selectivity was studied. Contrary to the well-known inhibition of HDS caused by H₂S, the global catalytic activity of HDS was enhanced more than two times with inclusion of H₂S in the reaction zone. The increase in the activity is attributed to a significant enhancement in the hydrogenation reaction route, while the direct desulfurization products were only slightly affected. It is suggested that H₂S modifies the active sites so that they can be exploited more efficiently in the hydrogenation reaction.

Full text of DOI:10.1016/j.jcat.2004.05.001

Journal of Catalysis 225 (2004) 531–535
www.elsevier.com/locate/jcat
Research Note
Investigation of 4,6-dimethyldibenzothiophene hydrodesulfurization over
a highly active bulk MoS2 catalyst
Hamdy Farag ∗ and Kinya Sakanishi
Institute for Energy Utilization, AIST, Tsukuba West, Ibaraki 305-8569, Japan
Received 18 February 2004; revised 30 April 2004; accepted 1 May 2004
Available online 7 June 2004
Abstract
The hydrodesulfurization (HDS) of 4,6-dimethyldibenzothiophene was investigated in a batch-stirred microautoclave reactor over a
laboratory-synthesized MoS catalyst. The reaction was found to proceed through two main parallel pathways: direct desulfurization (DDS)
2
and hydrogenation (HYD). The effect of a wide range of H S concentrations on the HDS activity and selectivity was studied. Contrary to
2
the well-known inhibition of HDS caused by H S, the global catalytic activity of HDS was enhanced more than two times with inclusion of
2
H S in the reaction zone. The increase in the activity is attributed to a significant enhancement in the hydrogenation reaction route, while
2
the direct desulfurization products were only slightly affected. It is suggested that H S modifies the active sites so that they can be exploited
2
more efficiently in the hydrogenation reaction.
2004 Elsevier Inc. All rights reserved.
Keywords: Hydrodesulfurization; MoS ; H S; 4,6-Dimethyldibenzothiophene
2
2
1. Introduction
tion pathway. Nevertheless, the contribution percentage of
each reaction route to the ultimate catalytic activity at fixed
reaction conditions is entirely ruled by the catalyst iden-
tity.
Hydrodesulfurization (HDS) is of considerable interest
to the oil refining industry as the global environment be-
comes more concerned. The relative reactivity of sulfur-
containing species varies significantly with the identity of
the molecular structure and the substituent location to the
sulfur atom. Polyaromatic thiophenes are considered to be
the most refractory sulfur species. They constitute the ma-
jor portion of sulfur in the high-boiling fraction of crude
oil, i.e., heavy naphtha, diesel, and light FCC naphtha. The
order of their reactivity has a strong relationship with the
selectivity involved in the reaction [1–6]. In general, the
HDS reaction of thiophene (the most reactive aromatic sulfur
species) occurs much more readily via the direct desulfur-
ization pathway, i.e., hydrogenolysis of the C–S bond. On
the other hand, the HDS of 4,6-dimethyldibenzothiophene
(4,6-DMDBT) (the main refractory target for deep HDS)
operates with a significant contribution from the hydrogena-
On looking for a hydrogenation pathway as a prospective
alternative choice to overcome the steric hindrance barrier
of these alkylated dibenzothiophene species, it is desirable
to obtain as effective a hydrogenation catalyst as possi-
ble [7–11]. In a previous study [12], it was observed that
the HDS of DBT over a bulk MoS2 catalyst occurred prin-
cipally through incorporation of the hydrogenation pathway
in which phenylcyclohexane was the predominant product.
The preliminary results showed the possibility of obtaining
highly active bulk MoS2 hydrogenation catalysts. It is inter-
esting and surprising that the activity of this MoS2 catalyst
for DBT HDS was enhanced remarkably by the presence of
H2S in the feed. Thus, it would be of interest to extrapo-
late, broaden, and demonstrate the HDS investigation over
this catalyst to the more refractory sulfur compounds such as
4,6-DMDBT, and this is the main aim of the present study.
The effect of H2S on the catalytic properties of the synthe-
sized MoS2 is analyzed.
*
Corresponding author.
E-mail address: hamdy-farag@aist.go.jp (H. Farag).
0021-9517/$ – see front matter
doi:10.1016/j.jcat.2004.05.001
2004 Elsevier Inc. All rights reserved.

532
H. Farag, K. Sakanishi / Journal of Catalysis 225 (2004) 531–535
◦
Fig. 1. Hydrodesulfurization reaction network of 4,6-DMDBT over MoS catalyst at 340 C and 3 MPa H .
2
2
2. Experimental
4,6-DMDBT). Traces of isomers of dimethylphenylcyclo-
hexane, dimethylbiphenyl, dimethyldibenzothiophene, and
3,3ꢀ-dimethylbicyclohexane were detected. Interestingly,
traces of dibenzothiophene were detected as well. Their cor-
responding desulfurized species were confirmed. The sug-
gested reaction pathways for the HDS of 4,6-DMDBT are
shown in Fig. 1. The individual reaction rate constants were
obtained from a nonlinear numerical Runge–Kutta fitting of
the experimental data on the assumption of first-order kinet-
ics for each step in this reaction scheme [14].
The mechanism was classified into two reaction routes,
hydrogenation that leads to the partially hydrogenated H4-
4,6-DMDBT and its isomers, and direct desulfurization
that leads to 3,3ꢀ-dimethylbiphenyl. The subsequent re-
action products from these two routes are 3,3ꢀ-dimethyl-
phenylcyclohexane and 3,3ꢀ-dimethylbicyclohexane. The
partially hydrogenated H4-4,6-DMDBT (predominant) and
the other derivatives containing the mother nucleus of diben-
zothiophene were treated collectively as one species.
Details of the MoS2 synthesis procedure have been pre-
sented elsewhere [13]. In brief, the ammonium heptamolyb-
date, AHM, precursor was annealed at 800 ◦C at a rate of
10 ◦C/min with a flowing gas mixture of (10% v/v) H2S/H2,
60 SCCM, from the beginning of the heating. Thereafter,
the catalyst was ground for 24 h in air using a media-
agitating mill that leads to a particle size of ca. 4 nm, as con-
firmed from the X-ray diffraction pattern, and a BET area of
118 m2/g. A Mo (10 wt%)/γ -Al2O3 catalyst was prepared
by conventional impregnation from AHM and sulfided at
400 ◦C by 10% v/v H2S/H2 prior to reaction. Catalytic tests
were performed in a stirred batch 100-ml microautoclave re-
actor. The Feedstock of 0.1 wt% 4,6-DMDBT in decane was
subjected to HDS investigation. All runs were carried out
at 340 ◦C under 3 MPa H2. The reaction was typically per-
formed as follows: The reactor was charged with the catalyst
powder (0.1 g) and 10 g of 0.1 wt% 4,6-DMDBT solution;
0.4 g copper powder was sometimes added to enable the re-
action to be studied in an environment free of H2S (the role
of copper powder is to act as a scrubber for H2S) [3]. The
effect of various concentrations of H2S on the HDS reaction
of 4,6-DMDBT was studied. The external feeding of H2S
was accomplished by inserting a predetermined amount of
sulfur powder into the feed. The measured amounts of H2S
by GASTEC gas detector (Model SG4010) at the end of the
HDS runs refer to the almost fully transformation of sulfur
to H2S. The reaction products were analyzed using GC–FID
(Agilent HP6890) and GC–MS (HP5970) equipped with a
capillary methylsiloxane column (0.32 mm × 50 m).
3.1. 4,6-DMDBT HDS in an environment free of H2S
Fig. 2 shows the product distribution curves for the HDS
reaction of 4,6-DMDBT over the MoS2 catalyst in an en-
vironment almost free from H2S (copper addition run; note
that a considerable concentration of H2S after this HDS run
was measured indicating the limited efficiency of copper
powder to trap H2S). The isomerized and hydrocracked trace
products in HDS of 4,6-DMDBT are a result of the acidic na-
ture of this catalyst. The yield of the partially hydrogenated
H4-4,6-DMDBT exhibits a maximum of about 54% from
the overall conversion. The result implies that the partially
hydrogenated H4-4,6-DMDBT species is an essential inter-
mediate in the hydrogenation reaction route and is not being
thermodynamically limited under the present reaction con-
ditions.
3. Results
The main observed reaction products were 3,3ꢀ-dimethyl-
phenylcyclohexane, 3,3ꢀ-dimethylbiphenyl, and partially
hydrogenatedtetrahydro-4,6-dimethyldibenzothiophene(H4-

H. Farag, K. Sakanishi / Journal of Catalysis 225 (2004) 531–535
533
◦
Fig. 2. Product distribution in the HDS of 4,6-DMDBT over MoS catalyst in approximately free environment of H S at 340 C and 3 MPa H .
2
2
2
MoS2, Mo/γ -Al2O3, and a commercial CoMo/Al2O3 cata-
lysts. The activity was normalized per gram weight of the ac-
tive MoS2 phase. Obviously, bulk MoS2 exhibits the highest
activity. Moreover, in terms of the catalyst volume (MoS2,
4.8–5 g/cm3; CoMo/Al2O3, 0.7–0.9 g/cm3), the catalytic
activity of this bulk MoS2 is highly superior.
4. Discussion
The results described above report a unique interesting
character of H2S in not being an inhibitor but rather a pro-
moter for the catalytic HDS reaction. However, before we
analyze these data further it is important to clarify any sus-
picious assumptions of this performance. Thus, one may
wonder about the possibility of this catalyst being partially
oxidized, in which case the presence of H2S would pos-
sibly bring it into a sulfide form, which is well known to
be more active than the oxide form. Accordingly, the pro-
motion could be due to the resulting sulfide phase and not
essentially to the presence of H2S itself. As this is a reason-
able explanation for the present behavior, we investigated
the reproducibility of the results. Another set of HDS exper-
iments was carried out over the milled MoS2 catalyst that
was freshly sulfided by 10 v/v% H2S/H2 at 400 ◦C prior
to reaction. We have further performed another in situ HDS
experiment in which the MoS2 was resulfided and loaded
into the autoclave along with the reactant solution and iron
powder using a glove box under Ar atmosphere. The results
are included in Fig. 4. As can be seen, almost identical be-
havior was obtained, thus excluding the possibility of the
activity enhancement being due exclusively to the sulfide
form species, especially for runs operated at an H2S level
that is 10 of orders higher than the quantitative needs for
full sulfidation of the catalyst. Moreover, the coincidence be-
tween the catalytic activity of 4,6-DMDBT HDS and H2S is
Fig. 3. Product distribution in the HDS of 4,6-DMDBT over MoS catalyst
2
◦
in a mild(self) environment of H S at 340 C and 3 MPa H .
2
2
3.2. 4,6-DMDBT HDS in an environment of H2S
The influence of a wide range of H2S concentrations was
studied by adding sulfur powder to the feedstock. Fig. 3
shows that when self-produced H2S remains in the reaction
it remarkably enhances the HDS activity of 4,6-DMDBT.
The enhancement in HDS activity has to be attributed to
H2S since the various reaction conditions for all runs were
kept unaltered. The product distributions of 4,6-DMDBT
HDS (run when self-produced H2S from the reaction re-
mained in the feed) showed almost entirely the involvement
of the hydrogenation reaction route. The influence of var-
ious H2S concentrations on the catalytic activity is shown
in Fig. 4. The achievement of a high HDS catalytic ac-
tivity for runs performed with inclusion of H2S basically
originates from the increase in activity in the hydrogenation
route. Furthermore, one can note the steady catalytic activity
in the presence of even a significant amount of H2S. Fig. 5
shows a comparison between the catalytic activity of bulk

534
H. Farag, K. Sakanishi / Journal of Catalysis 225 (2004) 531–535
◦
Fig. 4. Effect of inserted H S level on the ultimate HDS activity and selectivity of 4,6-DMDBT over MoS catalyst at 340 C and 3 MPa H . P, 1: Repro-
2
2
2
duced activity and selectivity results obtained over the milled MoS after fresh sulfidation, respectively. F, 2: In situ HDS activity and selectivity results,
2
respectively.
demonstrated practically in a circular reversible manner, i.e.,
keeping H2S in the reaction medium increases the activity
while taking it out drastically reduces it. The present results
then may indicate the involvement of H2S in the HDS reac-
tion mechanism through modifying the potentially catalytic
active sites, as suggested below.
H
A + H2S
A=S
H
The enhancement in activity with H2S was originally at-
tributed to the tremendous promotion that occurred in the
hydrogenation route, i.e., kHS1, while the direct desulfur-
ization, kD0, was not greatly affected. Fig. 4 describes the
influence of H2S on the reaction selectivity, defined as the ra-
tio of the apparent rate constant for DDS to the apparent rate
constant for HYD, i.e., kD0/kHS1. As the H2S concentration
increases, the selectivity shifts smoothly toward a higher in-
volvement of the hydrogenation route. These results support
the view that two distinct catalytically active sites exist for
each reaction pathway. H2S is proposed to chemically inter-
act with the catalyst producing new sites that are active for
HYD.
Unlike the conventional observation of inhibition result-
ing from H2S in the HDS for both real feed and refractory
sulfur compounds, the present results showed a clear im-
provement in the catalytic activity in the presence of even
a significant amount of H2S. In a trickle bed reactor the oil
stream flows so that the produced H2S accumulates to reach
its highest concentration at the reactor outlet. Consequently,
the downstream part of the catalyst bed has to operate in a
regime with a relatively high concentration of H2S. Thus, a
catalyst that behaves positively with H2S is worthy of more
morphological and structural investigations.
Fig. 5. Activity in the HDS of 4,6-DMDBT over MoS , CoMo/Al O com-
2
2 3
◦
mercial, and Mo/γ -Al O catalysts at 340 C and 3 MPa H . The activity
2
3
2
was normalized on the weight of MoS phase.
2
Inhibition due to the effect of H2S on the HDS activity
of DBT and/or 4,6-DMDBT over a conventional alumina-
supported Mo, CoMo, and/or NiMo catalyst has been ob-
served and documented in the literature [15–18]. The influ-
ence of H2S depends, however, on the HDS conversion and
on the nature of the catalyst. The effect was found to be
more pronounced for the direct desulfurization route than
for the hydrogenation route. Industrially, catalysts for the
HDS of real feedstocks inevitably operate in an environ-
ment containing H2S. The impact of H2S on the catalyst
behavior will permanently remain; hence, it is one of the
HDS by-products. The HDS selectivity depends on the na-
ture of the polyaromatic sulfur species that are processed,
the reaction conditions, and the catalyst. The first two fac-
tors cannot be widely changed because the real gas oil feed
already contains the most refractory sulfur species and ap-
plying severe operating reaction conditions is limited. Thus,
the only factor that eventually rules the selectivity in the deep

H. Farag, K. Sakanishi / Journal of Catalysis 225 (2004) 531–535
535
hydrodesulfurization extent is the catalyst identity. Two op-
References
tions of interest have been described in the literature to over-
come the refractory nature caused by the steric hindrance
of the alkylated substituents close to the sulfur atom: one is
to exploit isomerization, and the other is to focus on better
hydrogenation catalysts [19–25]. The latter option is more
interesting, as the isomerization catalysts can easily be deac-
tivated during reaction. The activity promotion due to H2S,
observed in the present work, is a distinct pioneering phe-
nomenon, specific only to this bulk catalyst.
[1] S. Rossini, Catal. Today 77 (2003) 467.
[2] C. Song, X. Ma, Appl. Catal. B 41 (2003) 207.
[3] H. Farag, I. Mochida, K. Sakanishi, Appl. Catal. A 194–195 (2000)
147.
[4] E. Lecrenay, K. Sakanishi, T. Nagamatsu, I. Mochida, T. Suzuka, Appl.
Catal. B 18 (1998) 325.
[5] S.T. Sie, Fuel Process. Technol. 61 (1999) 149.
[6] R. Shafi, G.J. Hutchings, Catal. Today 59 (2000) 423.
[7] H. Schulz, W. Bohringer, F. Ousmanov, F. Waller, Fuel Process. Tech-
nol. 61 (1999) 5.
[8] B.C. Gates, H. Topsoe, Polyhedron 16 (1997) 3213.
[9] M. Houalla, D.H. Broderick, A.V. Sapre, N.K. Nag, V.H.Z. de Beer,
B.C. Gates, H. Kwart, J. Catal. 61 (1980) 523.
[10] M. Houalla, N.K. Nag, A.V. Sapre, D.H. Broderick, B.C. Gates,
AIChE J. 24 (1978) 1015.
5. Conclusion
[11] X. Ma, K. Sakanishi, I. Mochida, Ind. Eng. Chem. Res. 33 (1994) 218.
[12] H. Farag, K. Sakanishi, M. Kouzu, A. Matsumura, Y. Sugimoto,
I. Saito, Ind. Eng. Chem. Res. 42 (2003) 306.
A clear enhancement in the HDS of 4,6-DMDBT over
MoS2 was observed with incorporation of H2S. The de-
velopment of the activity is mainly attributed to the en-
hancing activity of the hydrogenation pathway, as 3,3ꢀ-
dimethylphenylcyclohexane was the major reaction product.
H2S is suggested to be involved in the HDS reaction mech-
anism. The observation of the positive increase of MoS2
activity with H2S could open a new research area in which
an additional variety of catalytic hydrotreating is feasible.
Processes that take advantage of the high hydrogenation ac-
tivity of this catalyst may be worthwhile to pursue.
[13] H. Farag, Eng. Fuels 16 (2002) 944.
[14] W.H. Teukosky, W.T. Vettering, W.T. Flannery, Numerical Recipes in
Fortran, Cambridge Univ. Press, Cambridge, UK, 1992.
[15] R. Stephan, G. Emig, H. Hofmann, Chem. Eng. Process. 19 (1985)
305.
[16] R.C. Pille, C. Yu, G.F. Froment, J. Mol. Catal. 94, 3 (1994) 369.
[17] E. Olguin, M. Vrinat, Appl. Catal. A 170 (1998) 195.
[18] A. Kabe, Y. Aoyama, D. Wang, A. Ishihara, W. Qian, M. Hosoya,
Q. Zhang, Appl. Catal. A 209 (2001) 237.
[19] P. Michaud, J.L. Lemberton, G. Perot, Appl. Catal. A 169 (1998) 343.
[20] K.G. Knudsen, B.H. Cooper, H. Topsoe, Appl. Catal. A 188 (1999)
205.
[21] F. Bataille, J.L. Lemberton, P. Michaud, G. Perot, M. Vrinat,
M. Lemaire, E. Schulz, M. Breysse, S. Kasztelan, J. Catal. 191 (2000)
409.
[22] K. Sakanishi, T. Nagamatsu, I. Mochida, D.D. Whitehurst, J. Mol.
Catal. A 155 (2000) 101.
Acknowledgment
[23] C. Kwak, J.J. Lee, J.S. Bae, K. Choi, S.H. Moon, Appl. Catal. A 200
(2000) 233.
[24] F. Bataille, J.L. Lemberton, G. Perot, P. Leyrit, T. Cseri, N. Marchal,
S. Kasztelan, Appl. Catal. A 220 (2001) 191.
This work has been entrusted by the New Energy and
Industrial Technology Development Organization under a
subsidy of the Ministry of Economy, Trade, and Industry.
[25] Y. Yoshimura, L. Bihan, Fuel 81 (2002) 491.