Hydrocracking of ethyl laurate on bifunctional micro-/mesoporous zeolite catalysts

Article abstract of DOI:10.1002/cssc.200900234

Olefins from biomass: Metal-modified micro- and mesoporous composite materials are promising catalyst systems for the conversion of ethyl laurate, as a model compound for vegetable oils. The selectivity to light olefins can be enhanced up to 60 wt%.

Full text of DOI:10.1002/cssc.200900234

DOI: 10.1002/cssc.200900234
Hydrocracking of Ethyl Laurate on Bifunctional Micro-/Mesoporous Zeolite
Catalysts
Oliver Busse, Konstantin Rꢀuchle, and Wladimir Reschetilowski*[a]
Fossil raw materials such as crude oil, natural gas, and coal are
still the most important resources for modern chemical indus-
try. Nevertheless, efforts are made world-wide to substitute
them for renewables. A sustainable source that supplies chemi-
cal products and fuels is biomass, the diversity of which ex-
tends from vegetable biomass over fats and oils to biogenic
residues and wastes. From this perspective an important task
for further developing biorefinery technologies is to transfer
knowledge of catalytic and petrochemical processes in oil re-
fining to the ecological conversion of renewable resources.
A promising strategy to cover the requirements for essential
chemical products and fuels can be the use of vegetable oils.
Their high energy content and good compatibility to the pet-
rochemistry infrastructure makes them suitable substitutes for
fossil resources.[1,2] Based on these raw materials a wide
chemistry is accessible. Recent investigations have demonstrat-
ed that oxygen-containing compounds as well as linear,
branched, or aromatic hydrocarbons can be produced.[1] Cur-
rently, the focus is on second-generation biofuels and light ole-
fins such as ethene and propene, which are available by heter-
ogeneously catalyzed conversions of vegetable oils. Controlling
the product selectivity has proven to be problematic. However,
there is a continued demand for research investigations to-
wards the development of suitable, highly selective catalysts
that produce specific and market-relevant chemical products
and fuels based on vegetable oils.
treating, and hydrocracking reactions.[1] By using these condi-
tions defined refinery products such as liquid petroleum gas,
olefins, gasoline, diesel, and kerosene are available.
McCall et al. showed that in the presence of a heterogene-
ous catalyst the selectivity of light olefin formation can be en-
hanced.[4] Vegetable oil was converted into ethene (8.7 wt%)
and propene (22.4 wt%) by using H-ZSM-5 as catalyst.
Throughout the catalytic cracking of vegetable oil on acid zeo-
lite catalysts of the type H-ZSM-5, different reaction pathways
lead to a broad range of products, including heavy hydrocar-
bons, oxygenates, paraffins (gasoline, diesel, kerosene), light
olefins, light paraffins, aromatic compounds, crack gases, and
coke.[5] The cracking of rapeseed vegetable oil under fluid cata-
lytic cracking (FCC) conditions was studied by Dupain et al.[6]
By using a commercial equilibrium FCC catalyst, gasoline- and
diesel-range paraffins, aromatic compounds, and olefins were
available. However, the amount of olefins was low owing to
the high aromatization rate, up to 40 wt%. Bakhshi et al. inves-
tigated the production of C2–C4 olefins via the conversion of
canola oil at atmospheric pressure in a fixed-bed reactor.[7] In
the presence of a potassium-modified H-ZSM-5 catalyst a maxi-
mum yield of 25.8 wt% was obtained at 5008C and a WHSV of
1.8 hÀ1.
Ordered mesoporous materials can be used to improve the
access to the inner surface. The conversion of fatty acids over
mesoporous or composite micro- and mesoporous catalysts
was demonstrated by Ooi et al.[8,9] The conversion as well as
the selectivity to lower olefins was enhanced by using the
composite materials, in comparison to the pure micro- or mes-
oporous materials. Using MCM-41- or SBA-15-coated ZSM-5 re-
sulted in a maximum of the light olefin fraction yield during
the conversion of a fatty acid mixture. ZSM-5 coated with
30 wt% pure-siliceous SBA-15 enhanced the yield of propene
to up to 33.5 wt%. This indicates that composite systems of
both micro- and mesoporous materials, which combine the ac-
cessibility of the pores with a high surface acidity, improve the
selective conversion of vegetable oils towards light olefins.
The reaction mechanism of hydrotreating vegetable oil on
CoMo- and NiMo-catalysts reveals that the hydrogenation of
unsaturated fatty acid residues leads to the formation of dicly-
cerides, monoglycerides, acids, and waxes,[10] which can be
converted into heavy paraffins, propane, CO, and CO2. The hy-
drocarbons can subsequently be isomerized to iso-paraffins or
cracked to light olefins or paraffins.
Zeolites and zeolite-type materials with tailor-made struc-
tures and, by modifications, a wide range surface-chemical
properties play an important role as catalyst components, due
to their extremely large inner surfaces and adsorption capaci-
ties. Furthermore, the nature and concentration of active sites
in the zeolite interior can be adapted to specific applications.
Ordered mesoporous materials such as MCM-41 or SBA-15,
with their larger pore sizes, may be used for heterogeneously
catalyzed reactions of bulky reactants such as vegetable oils.
The pyrolysis of vegetable oil is a potential method for pro-
ducing light olefins. Bakhshi et al. reported the total conver-
sion of canola oil by thermal cracking in the range of 300–
5008C with weight hourly space velocities (WHSVs) between
3.3 and 15.4 hÀ1 [3]
Up to 37.1 wt% of light olefins and
.
13.6 wt% of aromatic compounds could be maintained as
main products. Similarly, biomass can be converted into petro-
leum in a petrochemical refinery by catalytic cracking, hydro-
The synthesis, characterization, and catalytic behavior of
micro- and mesoporous composite materials have been report-
ed by several authors (Van Bekkum et al.,[11] Klemt et al.,[12]
Karrlson et al.,[13] Mintova et al.,[14] and Pinnavaia et al.[15]). The
improved transport of reactants to the active sites via larger
mesopores (MCM-41) and a high acid-catalytic activity as well
[a] O. Busse, Dr. K. Rꢀuchle, Prof. W. Reschetilowski
Institute for Industrial Chemistry
Department of Mathematics and Natural Sciences
Dresden University of Technology
Zellescher Weg 19, 01062 Dresden (Germany)
Fax: (+49)351-463 32658
E-mail: wladimir.reschetilowski@chemie.tu-dresden.de
ChemSusChem 2010, 3, 563 – 565
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
563

as apparent shape selectivity (ZSM-5) are the most important
benefits of these composite materials.
constant in time. The conversion of ethyl laurate increased
with increasing reaction temperature and hydrogen pressure.
In comparison to other support materials NiMo/(Al-MCM-41/
ZSM-5) showed a lower conversion at 4008C (Figure 1) due to
The aim of the present work is to investigate the hydro-
cracking reaction of ethyl laurate (dodecanoic acid ethyl ester,
Scheme 1) by using different bifunctional micro- and mesopo-
Scheme 1. Hydrocracking of ethyl laurate.
rous catalyst systems. This model compound was used be-
cause of its lower viscosity and melting point, in comparison
to vegetable oils. The development of the catalyst is based on
innovative supports. The micro- and mesoporous bifunctional
catalyst samples used in this study are an acidic Al-MCM-41/
ZSM-5 composite as well as Al-MCM-41 and H-ZSM-5, loaded
with nickel (1.73 wt%) and molybdenum (6.67 wt%). The sur-
face acidity of the calcinated samples is characterized by tem-
perature-programmed desorption of ammonia (TPAD; Table 1).
By using these catalyst systems structure–effect relationships
in the hydrocracking of ethyl laurate are investigated.
Figure 1. Influence of support material and temperature on the conversion
of hydrocracking of ethyl laurate (WHSV=10 hÀ1).
its weaker acidity (Table 1). However, at higher temperatures
(5008C) the conversion increased strongly. This can be ex-
plained by a higher Brønsted acidity of the NiMo/(Al-MCM-41/
ZSM-5) sample compared to NiMo/Al-MCM-41. Increasing the
temperature led to an increase of catalyst activity due to an in-
tensification of the proton mobility. Further investigations
showed that NiMo/ZSM-5 reached high conversion rates, even
at lower temperatures of 4008C.
It can be assumed that the acidity of the catalyst systems
should also be reflected in the product distribution. Indeed,
the nature of acid sites strongly influences the cracking mecha-
nism and product distribution. Catalysts with high amounts of
Brønsted acid sites such as NiMo/(Al-MCM-41/ZSM-5) and
NiMo/ZSM-5 led to an improved selectivity towards C3 and C4
hydrocarbons (Table 2). However, a Lewis-acid-rich sample such
NiMo/Al-MCM-41 is more selective to methane formation. The
selectivity to light olefins was increased when using NiMo/(Al-
MCM-41/ZSM-5), due to metal–support interactions that partly
inhibit hydrogen transfer reactions (Table 3).[19] The strength of
the metal–support interactions can be derived from the loca-
tion of peak maxima in temperature-programmed reduction
(TPR) with hydrogen.
Table 1. Acidic properties of the investigated catalyst systems.
Catalyst sample[a,b]
Si/Al[c]
Acidity[d]
Lewis/
Brønsted[d]
(molar ratio) [mmol(NH3)gÀ1
]
NiMo/ZSM-5
NiMo/(Al-MCM-41/ZSM-5) 25.8
NiMo/Al-MCM-41 30.9
26.7
1.14
0.64
1.04
1:2.41
1:1.95
2.54:1
[a] As calcinated. [b] Ni 1.73 wt%, Mo 6.67 wt%. [c] Bulk catalyst, by
atomic absorption spectrometry (AAS). [d] By TPAD, with FT-IR (Brønsted
1450 cmÀ1, Lewis 1620 cmÀ1).[18]
NiMo/Al-MCM-41, NiMo/ZSM-5, and NiMo/(Al-MCM-41/ZSM-
5) were synthesized according to a procedure reported previ-
ously.[16,17] The conversion of ethyl laurate was performed in a
fixed-bed reactor under integral conditions. The influence of
catalyst modifications on the product distribution was studied
in dependence on the reaction temperature (300–5008C) and
process pressure (0–20 bar, hydrogen) at a constant space ve-
locity WHSV=10 hÀ1
.
As a result ethyl laurate was hydrocracked to a product mix-
ture that consisted of organic liquid products, gaseous prod-
ucts, water, and coke. The main
Especially the formation of the light olefins propene, 1-
butene, and ethene was enhanced in the presence of the
products were identified as light
olefins (C2–C4) and aromatic
Table 2. Influence of the catalyst acidity on the yield of selected hydrocarbons at 5008C and 5 bar.
compounds as well as n- and
iso-paraffins in the gasoline and
diesel range. During time-on-
stream all catalyst samples
showed only slight deactivation,
and conversion rates and prod-
uct selectivities remained nearly
Catalyst
C1
[wt%]
C2
[wt%]
C3
[wt%]
C4
[wt%]
C5
[wt%]
S C6–C10
[wt%]
C11
[wt%]
NiMo/ZSM-5
NiMo/Al-MCM-41
NiMo/(Al-MCM-41/ZSM-5)
7.3
30.6
3.9
22.2
11.0
17.9
29.0
4.0
24.6
13.3
4.8
22.1
3.1
5.4
8.6
<1.0
<1.0
<1.0
0.1
9.1
1.8
564
www.chemsuschem.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2010, 3, 563 – 565

Table 3. Influence of the catalyst acidity on the yield of light olefins at 5008C, 5 bar, and a WHSV of 10 hÀ1
.
[a]
Catalyst
Ethene
[wt%]
Propene
[wt%]
1-Butene
[wt%]
iso-Butene
[wt%]
trans-2-Butene
[wt%]
cis-2-Butene
[wt%]
S
Tmax
[8C]
NiMo/ZSM-5
NiMo/Al-MCM-41
NiMo/(Al-MCM-41/ZSM-5)
–
1.0
10.5
–
2.4
27.2
7.3
–
1.0
0.3
2.0
12.9
–
1.2
5.2
–
0.9
3.7
7.6
7.5
60.5
425
425
525
[a] TPR with 10% H2 in Ar, 108CminÀ1 temperature rate.
NiMo/(Al-MCM-41/ZSM-5) mixed system, compared to NiMo/
Al-MCM-41 and NiMo/ZSM-5 (Table 3). For this composite ma-
terial a maximum yield of 60 wt% light olefins (C2–C4) was ob-
tained.
Keywords: biomass · heterogeneous catalysis · olefination ·
porous materials · zeolites
[1] G. W. Huber, A. Corma, Angew. Chem. 2007, 119, 7320; Angew. Chem.
Int. Ed. 2007, 46, 7184.
This work demonstrates that bifunctional micro-/mesopo-
rous zeolite systems can be successfully used for hydrocracking
of ethyl laurate as a model compound for vegetable oils. The
catalytic activity and selectivity of the investigated catalyst
samples depends strongly on their surface acidity. Brønsted
acid sites mainly catalyze the formation of C3 and C4 hydrocar-
bons. On the other hand, Lewis acid sites dominate the forma-
tion of methane. Catalyst systems with strong metal–support
interactions induce an increase in the yield of olefins owing to
[2] K. D. Maher, D. C. Bressler, Bioresour. Technol. 2007, 98, 2351.
[3] R. O. Idem, S. P. R. Katikaneni, N. N. Bakhshi, Energy Fuels 1996, 10, 1150.
[4] M. J. McCall, T. L. Marker, J. Petri, D. Mackowiak, S. Czernik, D. Elliot, D.
Shonnard, 230th American Chemical Society National Meeting, Washing-
ton DC, 2005.
[5] T. Y. Leng, A. R. Mohamed, S. Bhatia, Can. J. Chem. Eng. 1999, 77, 156.
[6] X. Dupain, D. J. Costa, C. J. Schaverien, M. Makkee, J. A. Moulijn, Appl.
Catal. B: Environm. 2007, 72, 44.
[7] S. P. R. Katikaneni, J. D. Adaye, R. O. Idem, N. N. Bakhshi, Ind. Eng. Chem.
Res. 1996, 35, 3332.
[8] Y.-S. Ooi, R. Zakaria, A. R. Mohamed, S. Bhatia, Energy Fuels 2005, 19,
736.
inhibition of hydrogen transfer reactions[19]
.
[9] Y.-S. Ooi, R. Zakaria, A. R. Mohamed, S. Bhatia, Catal. Commun. 2004, 5,
441.
[10] G. W. Huber, P. O’Connor, A. Corma, Appl. Catal. A: Gen. 2007, 329, 120.
[11] K. R. Kloestra, H. W. Zandbergen, J. C. Jansen, H. van Bekkum, Micropo-
rous Mater. 1996, 6, 287.
[12] A. Klemt, W. Reschetilowski, Chem. Ing. Tech. 2001, 73, 872.
[13] A. Karlsson, M. Stçcker, R. Schmidt, Microporous Mesoporous Mater.
1999, 27, 181.
[14] J. Cejka, S. Mintova, Catal. Rev. Sci. Eng. 2007, 49, 457.
[15] Y. Liu, W. Zhang, T. J. Pinnavaia, Angew. Chem. 2001, 113, 1295; Angew.
Chem. Int. Ed. 2001, 40, 1255.
Experimental Section
The synthesis of support materials has been reported earlier.[16,17]
The obtained supports were loaded with nickel (1.73 wt%) and
molybdenum (6.67 wt%) by a wet-impregnation method. The re-
sulting catalyst samples were analyzed by physico-chemical charac-
terization methods (XRD/SAXS, AAS, nitrogen sorption, TPAD, and
TPR). The surface acidity and strength and nature of the acid sites
of the catalyst samples were characterized by temperature-pro-
grammed desorption of ammonia, followed by FT-IR spectroscopy.
The Lewis/Brønsted ratio was determined by integrating the inten-
sities of specific bonds (Brønsted 1450 cmÀ1, Lewis 1620 cmÀ1).
[16] O. Busse, K. Rꢀuchle, W. Reschetilowski, DGMK-Tagungsbericht 2008–3,
35.
[17] O. Busse, M. Neumann, W. Reschetilowski, DGMK-Tagungsbericht 2009–
2, 155.
[18] A. Taouli, A. Klemt, M. Breede, W. Reschetilowski, Stud. Surf. Sci. Catal.
1999, 125, 307.
[19] W. Reschetilowski, Erdçl Erdgas Kohle 1996, 112, 467.
Acknowledgements
O.B. acknowledges financial support for this work from the
Evonik Stiftung.
Received: October 8, 2009
Revised: November 26, 2009
Published online on April 30, 2010
ChemSusChem 2010, 3, 563 – 565
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
565