Ni//Mo synergism via hydrogen spillover, in pyridine hydrodenitrogenation

Article abstract of DOI:10.1016/j.catcom.2010.06.002

The pyridine hydrodenitrogenation (HDN) over physically separated stacked Ni//Mo beds was investigated using a continuous-flow high pressure (3 MPa) stainless steel microreactor. Results prove the existence of synergism between physically separated beds of Ni-/γ-Al₂O₃ and Mo/γ-Al₂O₃ catalysts for the pyridine hydrodenitrogenation reaction. This synergism is explained by the formation of hydrogen spillover. Product analysis of the pyridine hydrodenitrogenation over Ni//Mo stacked beds suggested that the hydrogen spillover modified the active sites of the MoS₂ by increasing the hydrogenation sites.

Full text of DOI:10.1016/j.catcom.2010.06.002

Catalysis Communications 11 (2010) 1154–1156
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Ni//Mo synergism via hydrogen spillover, in pyridine hydrodenitrogenation
d
d
F. Valdevenito a, R. García a, N. Escalona a, F.J. Gil-Llambias b,c, , S.B. Rasmussen , A. López-Agudo
⁎
a
Universidad de Concepción, Facultad de Ciencias Químicas, Casilla 160c, Concepción, Chile
Universidad de Santiago de Chile, Casilla 40, Correo 33, Santiago, Chile
Universidad Catolica Silva Henriquez, Santiago, Chile
b
c
d
Instituto de Catálisis y Petroleoquímica, CSIC, Cantoblanco, 28049 Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The pyridine hydrodenitrogenation (HDN) over physically separated stacked Ni//Mo beds was investigated
using a continuous-flow high pressure (3 MPa) stainless steel microreactor. Results prove the existence of
synergism between physically separated beds of Ni-/γ-Al2O3 and Mo/γ-Al2O3 catalysts for the pyridine
hydrodenitrogenation reaction. This synergism is explained by the formation of hydrogen spillover. Product
analysis of the pyridine hydrodenitrogenation over Ni//Mo stacked beds suggested that the hydrogen
spillover modified the active sites of the MoS2 by increasing the hydrogenation sites.
Received 26 March 2010
Received in revised form 25 May 2010
Accepted 1 June 2010
Available online 9 June 2010
Keywords:
HDN
© 2010 Elsevier B.V. All rights reserved.
Hso
Synergism NiMo
Stacked bed
1. Introduction
of the Co(Ni)–Mo–S phase over a catalyst containing Co and Mo can
never be totally excluded. Recent studies in our laboratories using a
The catalysts most frequently used in the industrial hydrotreating
(HYD) processes generally consist of a binary combination of Mo(or W)
and Co(or Ni) sulfides supported on γ-Al2O3 [1,2]. The evolution of the
different structural models proposed to explain the synergy in
hydrodesulfurisation (HDS) catalysts has been summarized in many
reviews, e.g. [1–6]. Among the numerous structural models proposed,
the two models withmajor acceptanceare the Co(Ni)–Mo–S model of H.
Topsøe et. al. [7,8] and the contact synergy/remote control (RC) model
developed by B. Delmon and co-workers [9,10]. In this model the
synergism is related to a hydrogen spillover species (Hso), which
migrates from a donor phase (D=Co9S8 or NiS) to an acceptor phase
(A=MoS2 or WS2) which hereby is activated.
The Co(Ni)–Mo–S model claims that the synergetic effect is due to
the formation of the Co(Ni)–Mo–S structure, which consists of small
MoS2-like domains with the promoter atoms (Co and Ni) located at the
edges of the MoS2 layers [7,8]. As the intrinsic activity of this structure
depends strongly on the catalyst preparation and activation procedure,
two types of Co(Ni)–Mo–S structures, Types I and II, having different
intrinsic activity, were proposed [11]. The Type I is formed after low-
temperature sulfidation and is characterised by incomplete sulfidation
and a lower hydrodesulfurisation (HDS) activity than Type II, which is
formed after high-temperature sulfidation and fully sulfided.
reactor system of physically separated, layered catalyst beds [12–18]
have provided further evidence for the RC model in HDS. It was found
that the activity of two stacked single beds of monometallic supported
Co(or Ni) sulfide (first-layer bed) and Mo sulfide catalysts (second-layer
bed) separated by a 5 mm layer of γ-Al2O3 or SiO2 for the HDS reaction
with gas-oil was higher than the sum of that observed for the
corresponding single-bed monometallic catalysts [12]. This synergistic
effect was found to be strongly dependent on the reaction temperature
[12–16], the separator's nature [13] and the distance between the
separate monometallic catalyst beds [14]. More recently, we detected
synergisms in the HDS reactionwith gas-oil for Mn//Mo Fe//Mo, Co//Mo,
Ni//Mo Cu//Mo and Zn//Mo stacked beds [15]. Also, the positive
influence of phosphorous in the classic Co–Mo/γ-Al2O3 catalyst was
explained by an increase of the Hso mobility between the donor,
D (Co9S8), and the acceptor, A (MoS2), phases induced by an increase of
thesurface acidityof thelayerthatseparatedD and A [16]. Synergismvia
Hso in stacked beds of Ru//Mo,Rh//Mo, Pd//Mo and Pt//Mo has also been
demonstrated [17]. The results of these studies clearly indicated that the
synergy in HDS over conventional Co(or Ni)Mo catalysts is not
necessarily associated uniquely to the formation of Co(Ni)–Mo–S type
structures. Obviously, in such physically separated double-bed catalyst
configurations, no direct interaction between the promoter and the Mo
occur, so the formation of a Co(Ni)–Mo–S type structure can be
unequivocally excluded. The observed synergetic effect can only be
accounted for by spillover hydrogen. It was further demonstrated that
the synergism via Hso observed over double-beds like Co(or Ni)/γ-
Al2O3//SiO2//Mo/γ-Al2O3 in HDS of dibenzothiophene also led to a
higher HYD selectivityas compared tothat of the Mo/γ-Al2O3 single-bed
Direct, unambiguous evidence for the existence for the RC
mechanism was dificult to achieve, since formation of small amounts
⁎
Corresponding author. Tel.: +56 2 4601102.
E-mail address: francisco.gil@ucsh.cl (F.J. Gil-Llambias).
1566-7367/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.catcom.2010.06.002

F. Valdevenito et al. / Catalysis Communications 11 (2010) 1154–1156
1155
Table 1
Table 2
Pyridine HDN conversion of Mo/γ-Al2O3, Ni/γ-Al2O3 single beds and Ni/γ-Al2O3//SiO2//
Mo/γ-Al2O3 stacked beds.
Pyridine HDN conversion of Mo/γ-Al2O3, Ni/γ-Al2O3 single beds and Ni/γ-Al2O3//SiO2//
Mo/γ-Al2O3 stacked beds at 498 K.
Bed
HDN (%)
498 K
Systems
Conversion
(%)
Product formation
Piperidine
523 K
548 K
n-pentane
Ni/Al2O3
0
0
0
Ni/Al2O3
0
–
–
Mo/Al2O3
Ni/Al2O3//SiO2//Mo/Al2O3
3.9
5.5
7.2
Mo/Al2O3
Ni/Al2O3//SiO2//Mo/Al2O3
3.9
7.0
–
3.9
6.6
7.0/1.8a
8.0/1.50a
9.2/1.3a
0.45
a
Promotion factor in HDN of Ni//Mo stacked beds.
tures were 498, 523 and 548 K; the total pressure was 3 MPa; the liquid
feed of 5650 ppm of pyridine (1000 ppm pf nitrogen) dissolved in
decaline; and, the liquid hourly space velocity (LHSV) was 30 h−1 and
H2 GHSV=3600 h−1. Samples were analysed by GC with a FID
(autosystem XL, Perkin Elmer) equipped with a CPSIL-5 CB column. In
addition to unreacted pyridine, piperidine and n-pentylamine were the
only products detected.
system [18]. In view of the above results on separated stacked-bed
experiments, and in order to develop more understanding of the
synergisms observed with HYD catalyst configurations, here we extend
our investigations from hydrodesulfurisation (HDS) to hydrodesnitro-
genation (HDN) reactions. Additionally, changes in the selectivity of the
pyridine HDN by influence of Hso are also studied.
Gas-oil enriched with pyridine HDN conversion was performed using
the same experimental conditions as described above, except the
differences that are summarized below: 0.4 g of Ni/γ-Al2O3 and 0.1 g of
Mo/γ-Al2O3 monometallic catalysts, diluted 1:1 with SiO2 were used. Prior
to the reaction, the catalysts received an in situ sulfidation treatment
during 4 h, at 623 K, total pressure 3 MPa, using 7% CS2 dissolved in gas-
oil. The reaction temperatures were 598, 623 and 648 K. The nitrogen
content in gas-oil was enriched up to 2000 ppm incorporating pyridine to
the initial gas-oil. Total nitrogen was analyzed on an Antek 703C
instrument by chemiluminescence detection. For all tests, a stabilization
period of at least 2 h was allowed before the first sample was collected.
2. Experimental
2.1. Catalysts preparation
The monometallic catalyst samples employed in the present study
were Mo/γ-Al2O3 and Ni/γ-Al2O3 monometallic samples prepared by
wet impregnation, dried overnight at 373 K and calcined at 823 K for
4.0 h. Nickel nitrate (Merck PA) and ammonium heptamolybdate
(Merck PA) were used as precursors. The support was a γ-Al2O3
Girdler T-126 (N2 BET 190 m2 g−1 and pore volume 0.365 cm3 g−1
)
and the separator of both beds was SiO2 BASF D11-10 (BET 154 m2 g−1
and pore volume 0.270 cm3 g−1). The metallic contents of Ni and Mo
were 3.9 and 11.9 g of NiO and MoO3 per 100 g of γ-Al2O3, respectively.
3. Results and discussion
Table 1 shows the pyridine conversion over Ni/γ-Al2O3 and Mo/γ-
Al2O3 single beds compared with the Ni/γ-Al2O3//Mo/γ-Al2O3 stacked
bed at different reaction temperatures. The Ni/γ-Al2O3 material is not
active in the temperature range, but the Mo/γ-Al2O3 catalyst exhibits
a moderate activity.
However, it can be clearly seen that the HDN conversion of the
stacked bed (Ni//Mo) is greater than the sum of the activity of the two
single beds (Ni+Mo). This proves that the synergism Ni–Mo via Hso,
initially reported for HDS reaction [12], also is present for the HDN
reaction. As observed for the HDS reaction [12–17] the synergism via
Hso in HDN decreases with the reaction temperature as shown in Fig. 1.
In close agreement with Kaluža et al. [19], the PF values obtained in
the present study using stacked bed in the HDN are lower than the
values obtained previously in the HDS, using similar experimental
conditions [15]. Certainly in the model reaction HDN of pyridine
2.2. Activity measurements
The pyridine HDN was carried out in a stainless steel continuous-
flow microreactor. A sample of 0.2 g Ni/γ-Al2O3 and 0.2 g of Mo/ γ-Al2O3
of catalyst diluted 1:1 with SiO2 was used in all the tests. These beds
were separated by 2 mmof SiO2. The composite beds are indicated asNi/
γ-Al2O3//SiO2//Mo/γ-Al2O3 and the promotion factor (PF) was calcu-
lated as: ((% conversion)Ni//Mo stacked beds)/(% conversion(Ni+Mo)
)
simple
bed. HDN conversions were defined as percent of total nitrogen, removed
from the initial gas-oil: HDN=((N0−N)/N0)⁎100 (N0=initial nitro-
gen concentration and N=nitrogen concentration after the reaction).
The remaining space in the reactor was filled with SiC particles. The
particle size of all catalysts, SiO2 and SiC, was between 0.42 and
0.85 mm. Prior to the reaction, the catalysts received an in situ
sulfidation with H2/H2S (10%) to 525 K for 4 h. The reaction tempera-
Fig. 2. Reaction scheme for HDN of pyridine.
Table 3
Gas-oil enriched with pyridine HDN conversion of Ni/γ-Al2O3, Mo/γ-Al2O3 single beds
and Ni/γ-Al2O3//SiO2//Mo/γ-Al2O3 stacked beds.
Bed
HDN (%)
598 K
623 K
648 K
Ni/Al2O3
1.0
1.5
3.5
Mo/Al2O3
0.5
1.5
3.5
Ni/Al2O3 +Mo/Al2O3
Ni/Al2O3//SiO2//Mo/Al2O3
1.5
3.0
7.0
2.5/1.7a
4.5/1.5a
7.5/1.1a
Fig. 1. Promotion factor (PF) in the HDN over and Ni//Mo stacked bed at different
reaction temperatures.
a
Promotion factor of Ni//Mo in stacked bed.

1156
F. Valdevenito et al. / Catalysis Communications 11 (2010) 1154–1156
4. Conclusions
These present results clearly indicate that the previous conclusions
about the hydrogen spillover action, observed for various transition
metal based catalysts for the HDS reaction are in fact also present, and
appear to be a general behaviour for the HDN reaction as well, as
depicted in Fig. 3. However, the Hso activated MoS-sites are more
active for the HDS reaction than for the HDN reaction.
Fig. 3. Schematic drawing representing the promotional effect of NiXSY provoked
hydrogen spillover, activating the physically separated Mo/Al2O3 sites for the HDN
reaction.
Acknowledgments
The authors thank CONICYT for the financial support (FONDECYT
Grant No. 1060029) and the Faculty of Chemical Sciences, University
of Conception.
simultaneous with HDS of thiophene, Kaluža et al. [19] find a PF of
5.7–4.6 in the HDS and 1.8–1.6 in the HDN over NiMoS catalyst.
Table 2 shows the product formation in the pyridine conversion
over Ni/γ-Al2O3, Mo/γ-Al2O3 and Ni//Mo at 498 K. As seen, n-pentane
is the only reaction product from the Mo/γ-Al2O3 catalyst, while Ni/γ-
Al2O3//Mo/γ-Al2O3 presents n-pentane and piperidine as reaction
products. Sonnemans et al. [20] claim that the pyridine is first
hydrogenated to piperidine, and then further on to pentylamine by C–
N bond cleavage. Finally the pentylamine is transformed to 1-pentene
by a second C–N bond cleavage, as shown in Fig. 2.
References
[1] P. Grange, Catal. Rev. Sci. Eng. 21 (1980) 135.
[2] H. Topsøe, B.S. Clausen, F.E. Massoth, in: J.R. Anderson, M. Boudart (Eds.),
Hydrotreating Catalysts, Science Technol., vol. 11, Springer, Berlin, 1996.
[3] R. Prins, V.H.J. de Beer, G.A. Somorjai, Catal. Rev. Sci. Eng. 31 (1989) 1.
[4] G.C.A. Schuit, B.C. Gates, AlChE J. 19 (1973) 417.
[5] S. Eijsbouts, Appl. Catal., A: Gen. 158 (1997) 53.
The formation of piperidine in the stacked Ni//Mo bed suggests
that Hso promotes the hydrogenation of the pyridine. This behavior is
in close agreement with the results obtained in the HDS of DBT [18] in
which Hso modified the active sites by increasing the amount of
hydrogenation sites rather than the hydrogenolisis sites of the MoS2.
The HDN results of gas-oil enriched with pyridine by Ni/γ-Al2O3
and Mo/γ-Al2O3 monometallic bed catalysts, as well as the stacked-
bed configuration, are summarized in Table 3. The Ni/γ-Al2O3 and Mo/
γ-Al2O3 monometallic catalysts both exhibit some activity for the gas-
oil HDN under the experimental conditions employed. However, also
for the HDN reaction of gas-oil a clear synergism between Ni/γ-Al2O3
and Mo/γ-Al2O3 in separated beds is observed, since the activity of the
Ni/γ-Al2O3//SiO2//Mo/γ-Al2O3 system is higher than the sum of the
Ni/γ-Al2O3 and Mo/γ-Al2O3 systems. Though the activities and Fso in
Tables 1 and 3 obviously cannot be compared directly to the pyridine
results, the Fso tendencies are similar. The promotion factor decreases
when the reaction temperature increases, as was also repeatedly
observed in all earlier work with the HDS reaction [12–17].
[6] D. Whitehurst, T. Isoda, I. Mochida, Adv. Catal. 42 (1998) 345.
[7] H. Topsøe, B.S. Clausen, R. Candia, C. Wivel, S. Mørup, J. Catal. 68 (1981) 433.
[8] H. Topsøe, B.S. Clausen, Catal. Rev. Sci. Eng. 26 (1984) 395.
[9] M. Karroua, P. Grange, B. Delmon, Appl. Catal. 50 (1989) L5.
[10] M. Karroua, H. Matralis, E. Sham, P. Grange, B. Delmon, J. Catal. 139 (1993) 371.
[11] C. Wivel, R. Candia, B.S. Clausen, S. Mørup, H. Topsøe, J. Catal. 68 (1981) 453.
[12] J. Ojeda, N. Escalona, P. Baeza, M. Escudey, F.J. Gil-Llambías, Chem. Commun.
(2003) 1608.
[13] P. Baeza, M.S. Ureta-Zañartu, N. Escalona, J. Ojeda, F.J. Gil-Llambías, B. Delmon,
Appl. Catal., A: Gen. 274 (2004) 303.
[14] P. Baeza, M. Villarroel, P. Ávila, A. López-Agudo, B. Delmon, F.J. Gil-Llambías, Appl.
Catal., A: Gen. 304 (2006) 109.
[15] M. Villarroel, P. Baeza, N. Escalona, J. Ojeda, B. Delmon, F.J. Gil Llambías, Appl.
Catal., A: Gen 345 (2008) 152.
[16] M. Villarroel, P. Baeza, F. Gracia, N. Escalona, P. Ávila, F.J. Gil Llambías, Appl. Catal.,
A: Gen 364 (2009) 75.
[17] M. Villarroel, A. Méndez, G. Águila, N. Escalona, P. Baeza, F. Gil-Llambías, Catal.
Today (2010), doi:10.1016/j.cattod.2010.01.008.
[18] N. Escalona, R. García, G. Lagos, C. Navarrete, P. Baeza, F.J. Gil-Llambías, Catal.
Commun. 7 (2006) 1053.
[19] L. Kaluža, D. Gulková, O. Šolcová, N. Žilková, J. Čejka, Appl. Catal., A: Gen. 351
(2008) 93.
[20] J. Sonnemans, P. Mars, J. Catal. 31 (1973) 209.