Preparation of Ni-MCM-41 by equilibrium adsorption - Catalytic evaluation for the direct conversion of ethene to propene

Article abstract of DOI:10.1016/j.catcom.2010.10.018

Ni-MCM-41 has been prepared by equilibrium adsorption of different nickel precursors. Nickel citrate and nickel nitrate gave the most active catalysts for the direct transformation of ethene into propene above 250 °C at atmospheric pressure. The maximal ethene conversion at 400 °C was 36% while propene selectivity reached 45%. Analysis of product formation spectra at different temperatures, residence times and inlet compositions revealed reaction kinetics consistent with a sequence of ethene dimerization, positional butene isomerization and propene retro-metathesis.

Full text of DOI:10.1016/j.catcom.2010.10.018

Catalysis Communications 12 (2011) 368–374
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
Preparation of Ni-MCM-41 by equilibrium adsorption — Catalytic evaluation for the
direct conversion of ethene to propene
b
b
c
T. Lehmann a, , T. Wolff , V.M. Zahn , P. Veit , C. Hamel a,b, A. Seidel-Morgenstern a,b
⁎
a
Otto-von-Guericke-University, Institute of Process Engineering, Universitätsplatz 2, 39106 Magdeburg, Germany
Max-Planck-Institute for Dynamics of Complex Technical Systems, Sandtorstr. 1, 39106 Magdeburg, Germany
Otto-von-Guericke-University, Institute of Experimental Physics, Universitätsplatz 2, 39106 Magdeburg, Germany
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 21 July 2010
Received in revised form 20 September 2010
Accepted 22 October 2010
Available online 29 October 2010
Ni-MCM-41 has been prepared by equilibrium adsorption of different nickel precursors. Nickel citrate and
nickel nitrate gave the most active catalysts for the direct transformation of ethene into propene above 250 °C
at atmospheric pressure. The maximal ethene conversion at 400 °C was 36% while propene selectivity reached
45%. Analysis of product formation spectra at different temperatures, residence times and inlet compositions
revealed reaction kinetics consistent with a sequence of ethene dimerization, positional butene isomerization
and propene retro-metathesis.
Keywords:
Ethene
© 2010 Elsevier B.V. All rights reserved.
Propene
Metathesis
Nickel catalyst
MCM-41
Equilibrium adsorption
1. Introduction
Lack of a suitable catalyst has been the main obstacle for propene
generation from ethene. Early candidates based on supported
Propene is one of the major feedstocks in the chemical industry
with its prominent role being mainly due to the sustained growth in
polypropylene production numbers. Additionally, large volume
derivatives like cumene, propene oxide and acrylonitrile contribute
to the demand for propene [1]. The primary sources of this olefin have
traditionally been steam cracking of various hydrocarbon streams and
refinery-based catalytic cracking. In these processes propene is
merely a co-product of ethene production and petroleum refining,
respectively. Unfortunately, owing to reaction characteristics and
process economics there is only limited possibility to adjust the
product spectrum in favor of elevated propene yields [2]. A higher
growth rate in the demand for propene compared to ethene therefore
lead to considerable interest in on-purpose propene production
technologies. One of these specific synthesis approaches is the direct
conversion of ethene to propene according to the stoichiometric net
equation 3 C2H4 →2 C3H6. This reaction would be an interesting
means to boost propene capacity, particularly when integrated into
steam cracker plants. An even more attractive reaction scheme results
if the above reaction is combined with the dehydration of ethanol to
ethene. When using bio-ethanol this would open up a propene
production route based on renewable feedstocks [3].
molybdenum hexacarbonyl [4] and tungsten oxide [5] suffered from
activities being too low to justify additional research efforts. Lately, a
number of heterogeneous catalysts have been developed which
showed sufficient activity and selectivity to be subject to further
investigations. These catalyst systems can be classified via the
supposed mechanism of propene formation. First, there are catalysts
which afford propene by ethene oligomerization and consecutive
cracking processes. SAPO-34 [6–8] and ZSM-5 [9] belong to this group
of catalysts. Second, transition metal compounds like alumina
supported tungsten hydride [10] as well as Ni-MCM-41 [11–13]
catalyze the desired reaction. For Ni-MCM-41, Iwamoto and Kosugi
proposed a catalytic pathway consisting of ethene dimerization to 1-
butene, butene double-bond shift and finally cross-metathesis
between ethene and 2-butenes to yield propene [13]. The production
of propene starting with ethanol as reactant is also catalyzed by Ni-
MCM-41 [14].
The focus in the present study is on Ni-MCM-41. Iwamoto and co-
workers [11–13] used primarily a template-ion exchange approach in
their preparation of Ni-MCM-41. However, a preliminary preparation
attempt with [Ni(NH3)x]2+ caused them to suspect that equilibrium
adsorption of nickel precursors might lead to comparably active
catalysts [11]. In this contribution equilibrium-adsorbed Ni-MCM-41
catalysts were prepared with different nickel salts and evaluated with
respect to their catalytic performance in the direct transformation of
ethene into propene. Particular emphasis has been placed on
⁎
Corresponding author. Tel.: +49 391 6718128; fax: +49 391 6712028.
E-mail address: tino.lehmann@ovgu.de (T. Lehmann).
1566-7367/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.catcom.2010.10.018

T. Lehmann et al. / Catalysis Communications 12 (2011) 368–374
369
ethene conversion
1-butene selectivity
propene selectivity
2-butene selectivity
elucidating the influence of temperature, residence time and feed
composition on the product distribution in order to gain further
insight into the nature of the reaction network on these catalysts.
Catalyst stability was another important point of study.
50
40
30
20
10
0
2. Experimental
Catalysts have been prepared using commercially available MCM-
41 as support (Sigma-Aldrich, specific surface area: 1000 m2/g, pore
diameter: 2.3–2.7 nm). Nickel precursors were used as follows: nickel
(II) acetate tetrahydrate (Sigma-Aldrich, purity: N98%), nickel (II)
acetylacetonate (VWR, N98%), nickel citrate hydrate (Alfa Aesar,
N98%) and nickel (II) nitrate hexahydrate (VWR, N99%). In a typical
preparation procedure, an amount of the respective nickel precursor
corresponding to a nickel concentration of 0.02 mol/l was completely
dissolved in 40 g of demineralised water. 2 g of MCM-41 were added
to the solution. The suspension was stirred for 60 min at room
temperature and then stored without stirring for 20 h in a closed
Teflon bottle at 80 °C. Subsequently, the solid material was filtered,
washed with 500 ml of demineralised water and dried at 80 °C.
Finally, the samples underwent calcination in air at 550 °C for 6 h
(heating rate: 1 °C/min).
The resulting catalysts were analyzed by nitrogen physisorption at
−196 °C employing standard BET procedures on a Quantachrome
Nova2200e apparatus. Samples were degassed beforehand for 24 h at
80 °C in vacuum. Small-angle powder X-ray diffractograms were
measured on an X'Pert apparatus by PANalytik (Cu Kα radiation).
Moreover, a number of transmission electron microscopy (TEM)
micrographs were taken on a FEI CM200 set-up operated at 200 kV. To
this end, powdered catalysts were suspended in ethanol for sample
preparation followed by ultrasonic dispersion. A drop of catalyst
suspension was then placed on a coated copper grid and allowed to
dry. An EDAX DX-4 system was used for energy dispersive X-ray
(EDX) analyses. Spectra were collected in spot mode with a spot
diameter of 15 nm.
All catalytic tests were performed at atmospheric pressure in a
fixed bed reactor (quartz tube, inner diameter: 0.6 cm, length of
catalyst bed: ca. 7 cm) loaded with 0.5 g catalyst. The feed consisted of
ethene (Westfalen Gas, purity: 99.95%) diluted with appropriate
amounts of nitrogen (Westfalen Gas, 99.999%) to vary the inlet
concentration of ethene. The fresh catalyst was exposed to flowing
nitrogen at 50 °C for 2 h. Subsequently the reaction feed was directly
sent to the reactor followed by relatively slow stepwise heating
(increment: 25 °C) from 50 °C to reaction temperature. This proce-
dure roughly took 2 h. An on-line gas chromatograph (Agilent 6890
GC/TCD with 5973 MSD) equipped with a HP-PLOT/Q column was
used for analysis of the feed gas and the reactor effluents.
nickel acetate
nickel
nickel citrate
nickel nitrate
acetylacetonate
Fig. 1. Influence of the nickel precursor on conversion and product spectrum at 400 °C;
feed: 5% ethene in N2, total flow rate: 1 l/h.
point of view – catalysts prepared with nickel nitrate (propene yield:
14.8%) and nickel citrate (12.3%) are clearly superior to those based on
the acetate (9.3%) and acetylacetonate (9.5%) precursors. It should be
mentioned that catalysts ex citrate and ex nitrate exhibited almost
equal performance at temperatures below 400 °C. Byproducts were
only formed to a very small degree (respective yields were always
below 1%) and included mainly ethane, iso-butene and hexenes.
The observed product spectrum sheds some light on the operative
reaction mechanism. Hydrocarbon cracking would require a degree of
surface acidity which exceeds the necessary acidity for skeletal butene
isomerization [15]. Accordingly, the lack of substantial amounts of iso-
butene in the reactor effluent can be taken as an indication for the
absence of cracking reactions. This finding is further corroborated by
the negligible formation of saturated hydrocarbons which are usually
observed upon butene cracking [16]. It appears therefore to be rather
unlikely that propene generation over Ni-MCM-41 is the result of an
oligomerization-cracking-scheme occurring as the main reaction
path.
Since our attention in the present study was mainly on the
catalytic performance of the catalysts we did not investigate in detail
the causes for the varying effectiveness of the different precursors.
Some remarks in that direction may nonetheless be appropriate here.
As can be seen in Fig. 2, catalysts prepared with nickel nitrate and
nickel citrate exhibited X-ray diffractograms with characteristic peaks
at 2θ=2.4°, 4.0° and 4.6° assignable to the (100), (110) and (200)
reflections [17]. Only a weak (210)-peak at 2θ=5.9° observable on
the pristine support material vanished after the equilibrium adsorp-
tion treatment.
3. Results and discussion
The following discussion will focus mainly on conversion and
selectivities. These are defined on a carbon basis as
ethene consumed
In addition, nitrogen adsorption measurements on these catalysts
resulted in type-IV-isotherms with a distinct capillary condensation
step at relative pressures of ca. 0.31 (see Fig. 2). Such physisorption
behavior is typical for MCM-41 materials [18]. Decrease of surface
area upon nickel treatment was usually below 80 m2/g (detailed data
are given in Table 1). It can therefore be concluded that these catalysts
retained their structural integrity.
ethene conversion =
propene selectivity =
butene selectivity =
100;
⋅
ethene fed
propene generated
ethene consumed
3
2
100;
⋅
⋅
butene generated
ethene consumed
4
2
100:
⋅
⋅
Yields can be obtained by multiplying conversion with the
respective selectivity.
Activity and selectivity data for different nickel precursors are
depicted in Fig. 1. The differences regarding ethene conversion and
propene selectivity might not seem to be too pronounced. However,
in terms of propene yield – which is the key parameter from a process
On the contrary, catalysts based on nickel acetate and nickel
acetylacetonate lost all diffraction lines except for the dominant
(100)-peak which also decreased. They also suffered a massive
reduction in specific surface area (more than 400 m2/g) which was
accompanied by deformed isotherm shapes lacking the typical sharp
step caused by the mesopores. Obviously, the equilibrium adsorption

370
T. Lehmann et al. / Catalysis Communications 12 (2011) 368–374
700
pristine support
catalyst ex nitrate
600
500
400
300
200
100
ex citrate
ex acetate
ex acetylacetonate
pristine suppport
catalyst ex nitrate
catalyst ex citrate
catalyst ex acetate
catalyst ex acetylacetonate
1,0
2,0
3,0
4,0
5,0
6,0
7,0
8,0
9,0
10,0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
θ
2 , °
Relative Pressure, -
Fig. 2. Structural characterization of support material and prepared catalysts by XRD patterns (left) and nitrogen physisorption isotherms (right).
treatment led in these cases to structural distortion of pore
periodicity. A possible explanation may be the influence of pH. Use
of nickel nitrate and nickel citrate resulted in final pH values of
approximately 3.6 and 4, respectively, whereas the other two
precursors gave rise to higher pH values (nickel acetate: 5.6, nickel
acetylacetonate: 7.2).
TEM investigations revealed another difference. Nickel nitrate and
citrate lead to catalysts with dispersed nickel particles while catalysts
ex acetate and acetylacetonate exhibited layered nickel structures
(see Fig. 3). The latter morphology has already been observed by TEM
on dried Ni/SBA-15 [19] and Ni-MCM-41 [20] samples prepared by
deposition-precipitation; it was identified in both cases as nickel
phyllosilicate. Our own EDX investigations proved that the local
silicon content was much higher in case of the layered structures
(roughly 20 mol%) compared to the nickel particles (ca. 3 mol%). Two
representative spectra are given in Fig. 3. Even though these numbers
should not be taken as a precise quantitative measure, they do
however support the hypothesis that the layered structures are
comprised of a silicate phase.
In summary, variations concerning catalytic performance are likely
to be due to the existence of different nickel phases as well as varying
degrees of support integrity. For the remainder of the discussion, the
focus is on the most active catalysts which were prepared with the
nitrate and citrate precursors.
A noteworthy difference exists between equilibrium-adsorbed Ni-
MCM-41 employed here and the template-ion exchanged catalysts
used by Iwamoto and co-workers. It has been reported in [13] that a
small amount of water was required in the feed gas for stable catalyst
activity. This would be somewhat detrimental since the hydrothermal
stability of MCM-41 is questionable over longer operating periods
[21]. However, no water vapor was needed for the catalysts used in
the present study. Experiments have been conducted with 0.5, 1 and
1.5% water in the feed gas. In all cases the effect of water addition was
an unselective decline in catalytic activity.
It is also worthwhile to point out that no dedicated activation
procedure was necessary to achieve catalytic activity. We tried
several activation strategies by varying pre-treatment duration,
temperatures and gas compositions which also included water
treatments. No significant changes in terms of conversion or
selectivity could be observed. This is in contrast to Ikeda et al. [11]
who found that pre-treatment at 300 °C was crucial to obtain
sufficient activity on template-ion exchanged Ni-MCM-41. All of
these results indicate that the active nickel sites on equilibrium-
adsorbed Ni-MCM-41 might be different from those on template-ion
exchanged catalysts.
A comparison with literature results in terms of catalytic
performance is appropriate at this point. Table 2 summarizes
performance data for the Ni-MCM-41 catalyst ex citrate employed
in this study as well as for the template-ion exchanged and
equilibrium-adsorbed catalysts used by Ikeda et al. [11]. Numbers
were estimated from Fig. 1 in their publication (data after 4 h of
continous operation). Note that the definition of conversions as given
by Iwamoto [12] corresponds to yields in our nomenclature and that
the sum of the (carbon-based) yields is equal to ethene conversion as
defined here; numerical values were accordingly converted. It should
also be noticed that the operation conditions in Table 2 are not
completely comparable, though the differences in reaction tempera-
ture and space time should at least in part compensate each other.
Inspection of Table 2 clearly shows that the sample prepared by
Ikeda et al. using their template-ion exchange technique is superior to
both equilibrium-adsorbed catalysts. This is mainly due to a higher
activity level. Future work on the preparation of Ni-MCM-41 for the
conversion of ethene to propene should therefore focus on template-
ion exchange as the method of choice. Differences between the two
equilibrium-adsorbed catalysts with respect to conversions and
propene yields are comparatively small. The catalyst ex citrate
prepared in the present study exhibits a higher combined propene/
butene selectivity due to less byproduct formation.
The kinetics of the transformation of ethene into propene were
studied next. This is of considerable interest given that the proposed
catalytic pathway for template-ion exchanged Ni-MCM-41 contains a
metathesis step which is supposed to proceed on nickel sites [13].
Typical olefin metathesis catalysts contain molybdenum, tungsten or
rhenium components [22], whereas nickel is not known to catalyze
this kind of reaction effectively [23]. It is therefore important to
elucidate whether the conversion of ethene to propene on equilib-
rium-adsorbed Ni-MCM-41 also adheres to the reaction scheme
comprised of ethene dimerization, butene double-bond migration and
propene retro-metathesis.
Table 1
Textural characteristics of catalysts prepared with different nickel precursors.
Sample
Specific surface Pore diametera Total specific pore
area (m2 g−1
)
(nm)
volume (cm3 g−1
)
Parent MCM-41
1059
953
983
597
534
3.02
3.03
3.05
3.70
3.70
1.01
0.90
0.89
0.82
0.72
Catalyst ex nitrate
Catalyst ex citrate
Catalyst ex acetate
Catalyst ex acetylacetonate
The effect of residence time variations is presented in Fig. 4. As
expected, ethene conversion is consistently increasing with pro-
longed residence time. Moreover, several conclusions can be drawn
a
Based on the desorption branch of the isotherm.

T. Lehmann et al. / Catalysis Communications 12 (2011) 368–374
371
Fig. 3. Nickel phase characterization of freshly calcined Ni-MCM-41 catalyst ex citrate (left) and ex acetylacetonate (right) by TEM micrographs (upper row) and EDX spectra (lower row).
from the illustrated selectivity behavior. It is instructive to concen-
trate first on the temperature range below 225 °C where no
appreciable propene formation takes place; the only detected product
species are the three n-butene isomers. Selectivity to 1-butene is
without exception diminished at higher residence times. The opposite
is true for the 2-butenes. The terminal olefin appears indeed to be the
primary product of ethene dimerization which is followed by double-
bond migration to give 2-butenes. It is thus evident that 1-butene
assumes the role of a preceding intermediate for the formation of 2-
butenes in the reaction sequence.
nonetheless noteworthy that the absolute slope values of propene
and 2-butene selectivities are nearly equal between 300 and
375 °C. It is hence very probable that propene generation is a 2-
butene consuming process.
The information gathered so far about the reaction network can be
summarized in a sequence of reaction products: 1-butene→2-
butenes→propene. This sequence is also in line with the increased
propene selectivity at higher residence times displayed in Fig. 4. The
effect is however only weakly pronounced in the range of residence
times examined here.
Additional insight can be gained from the temperature depen-
dence of 2-butene selectivity. As can be seen in Fig. 4, 2-butene
selectivity starts to decline between 225 °C and 250 °C which
coincides with the onset of propene formation. These changes are
caused by differences in the activation energies of the various
reactions comprising the reaction network and consequently do
not allow for direct conclusions about reaction stoichiometries. It is
A concluding remark should be made regarding the conversions in
Fig. 4. Ethene consumption at temperatures above 250 °C is still
increasing while all C4-selectivities are in decline and considerably
below 100% in total. Formation of propene is hence simultaneously
consuming 2-butene and ethene — a very strong indication that
metathesis between ethene and 2-butene is indeed proceeding over
equilibrium-adsorbed Ni-MCM-41.
Table 2
Catalytic performance of different nickel loaded MCM-41 catalysts.
Catalyst
Ni-M41d
Preparation methoda Reaction temperature (°C) Space timeb (gcat min ml−1
)
Ethene conversion (%) Propene selectivity (%) Butene selectivityc (%)
TIE
400
400
375
0.027
0.027
0.030
59
26
27
41
39
35
31
35
62
Ni-NH3/M41d EA
Ni-MCM-41e
EA
a
TIE: template-ion exchange and EA: equilibrium adsorption.
Defined as catalyst weight divided by total feed flow rate.
Cumulative value for all butene isomers.
Catalysts employed by Ikeda et al. [11], feed contained 9.3% ethene.
Catalyst ex citrate as used here, feed contained 10% ethene.
b
c
d
e

372
T. Lehmann et al. / Catalysis Communications 12 (2011) 368–374
40
30
20
10
0
40
Total flow rate:
0.7 l/h
1 l/h
1.5 l/h
2.5 l/h
30
20
increasing
residence time
10
increasing residence time
0
100
100
150
200
250
300
350
400
150
200
250
300
350
400
Temperature, °C
Temperature, °C
45
40
35
30
25
20
15
10
85
80
75
70
65
60
55
50
45
40
increasing residence time
increasing residence time
100
150
200
250
300
350
400
100
150
200
250
300
350
400
Temperature, °C
Temperature, °C
Fig. 4. Dependence of conversion and product spectrum on residence time; catalyst ex citrate, feed: 10% ethene in N2.
The effect of ethene content in the feed gas on the product spectrum
wasexamined asa secondimportantaspect. Results are givenin Fig. 5. In
general, ethene conversion and butene selectivities rise with increasing
ethene feed fraction even though there is some scatter in the data. This is
what could have been expected considering the arrangement of ethene
dimerization and butene double-bond shift as consecutive reactions.
15
55
50
45
40
Feed fractionof ethene:
2.5%
12
5%
10%
increasing ethene
content in the feed
35
9
30
25
20
15
10
5
6
increasing ethene
content in the feed
3
0
150
0
150
200
250
300
350
400
200
250
300
350
400
Temperature, °C
Temperature, °C
30
75
70
65
60
55
50
45
40
35
25
20
15
10
increasing ethene
content in the feed
increasing ethene
content in the feed
150
200
250
300
350
400
150
200
250
300
350
400
Temperature, °C
Temperature, °C
Fig. 5. Effect of initial ethene concentration on conversion and product spectrum; catalyst ex citrate, total flow rate: 2.5 l/h.

T. Lehmann et al. / Catalysis Communications 12 (2011) 368–374
373
60
50
40
30
20
10
0
50
ethene conversion
40
propene selectivity
1-butene selectivity
2-butene selectivity
30
20
10
0
0
5
10
15
20
25
30
35
40
0
10
20
30
40
50
60
Time, h
Time, h
Fig. 6. Long-time behavior of catalysts ex citrate, total flow rate: 1 l/h; left diagram: 350 °C, 10% ethene, right diagram: 375 °C, 5% ethene.
A more puzzling detail is the lower propene selectivity at higher
initial ethene concentrations. It should be realized in this context that
one of the branches in the suspected reaction network consists of
parallel reactions with respect to ethene, i.e., dimerization and
metathesis:
what faster with time compared to the dimerizing catalyst function.
The observed deactivation behavior could be an indication for
different nickel sites being responsible for the two reactions. This
would be in contrast to Ikeda et al. [24] who speculated that the active
nickel site for dimerization and metathesis may be the same on
template-ion exchanged Ni-MCM-41.
Catalysts colors changed from very light gray (fresh Ni-MCM-41)
to dark gray or black depending on the duration of the deactivation
experiment. The main reaction is therefore accompanied by coking
processes. For the time being, it is unclear whether coking is the
primary reason for catalyst deactivation. More detailed investigations
are required to clarify this point. In addition, it must be stressed that
the good catalyst stability may well be due to the rather high dilution
of the inlet flow. It is left to future research whether this favorable
deactivation behavior can be retained under more severe conditions
like concentrated hydrocarbon feeds and higher conversions.
+ H2C= CH2
H2C= CH2
H2C= C3H6
H4C2= C2H4
2 H2C= C2H4
In view of the two reaction stoichiometries it can be expected that
the ethene reaction order is higher in the case of dimerization. Higher
ethene concentrations favor the formation of butenes instead of
propene in such a scenario. One may argue that this undesired effect
on the metathesis reaction rate can be compensated by higher
concentrations of 2-butenes. However, because of finite reaction rates
only a certain fraction of 1-butene is indeed converted to 2-butene.
Further, compensation would be non-stoichiometric with respect to
the metathesis step even with an infinite isomerization rate: two
ethene molecules are consumed for a single 2-butene molecule while
ethene and 2-butene enter metathesis in equimolar amounts.
Accordingly, this kind of compensation could only make up for the
primary ethene effect if the partial reaction order of the metathesis
reaction with respect to 2-butene were considerably higher than the
order regarding ethene.
In summary, based on the observations made it could be shown
that the order of product formation is 1-butene→2-butenes→pro-
pene. Furthermore, it was confirmed that propene formation is very
likely to be caused by metathesis between ethene and 2-butenes. All
of the kinetic results are compatible with the catalytic pathway
proposed in the literature [13].
The catalytic study was concluded by an investigation of catalyst
stability. Selected results are displayed in Fig. 6. There was hardly any
deactivation at 350 °C over a period of 35 h (left diagram in Fig. 6).
Deactivation behavior was independent of ethene concentrations;
identical runs with 2.5% and 5% ethene in the feed gave the same
stable behavior.
4. Conclusions
Equilibrium adsorption of nickel precursors on MCM-41 leads to
catalysts which effectively catalyze the direct conversion of ethene to
propene in a temperature range between 250 and 400 °C. Nickel
citrate and nickel nitrate proved to be the most effective precursors.
Only minor amounts of byproducts were formed. Propene formation
on these catalysts did not necessitate water vapor. Kinetic measure-
ments supported the catalytic pathway consisting of ethene dimer-
ization, positional butene isomerization and propene retro-
metathesis which has been proposed in the literature. Higher
temperatures as well as higher residence times improved the
selectivity to propene while increased feed concentrations of ethene
led to lower relative amounts of propene. Ni-MCM-41 prepared by
equilibrium adsorption exhibited very good stability at reaction
temperatures below 375 °C. Active nickel sites on equilibrium-
adsorbed Ni-MCM-41 might be different from those on template-
ion exchanged systems.
References
[1] H.A. Wittcoff, B.G. Reuben, J.S. Plotkin, Industrial Organic Chemicals, 2nd ed.John
Wiley & Sons, Hoboken, 2004.
[2] A.M. Aitani, in: S. Lee (Ed.), Encyclopedia of Chemical Processing, Taylor & Francis,
New York, 2006, pp. 2461–2466.
[3] A. Behr, A. Kleyensteiber, U. Hartge, Chem. Ing. Tech. 82 (2010) 201–213.
[4] P.P. O'Neill, J.J. Rooney, J. Am. Chem. Soc. 94 (1972) 4383–4384.
[5] T. Yamaguchi, Y. Tanaka, K. Tanabe, J. Catal. 65 (1980) 442–447.
[6] J.Z. Li, Y. Qi, Z.M. Liu, G.Y. Liu, F.X. Chang, Chin. J. Catal. 29 (2008) 660–664.
[7] H. Oikawa, Y. Shibata, K. Inazu, Y. Iwase, K. Murai, S. Hyodo, G. Kobayashi, T. Baba,
Appl. Catal. A 312 (2006) 181–185.
[8] H. Zhou, Y. Wang, F. Wei, D. Wang, Z. Wang, Appl. Catal. A 348 (2008) 135–141.
[9] B. Lin, Q. Zhang, Y. Wang, Ind. Eng. Chem. Res. 48 (2009) 10788–10795.
[10] M. Taoufik, E. Le Roux, J. Thivolle-Cazat, J.-M. Basset, Angew. Chem. Int. Ed. 46
(2007) 7202–7205.
Moderate deactivation was observed at 375 °C. Ethene conversions
dropped from 36% to 25% after 55 h. The selectivity to 1-butene
remained constant during this time. On the contrary, a slight increase
in 2-butene selectivity could be evidenced which was matched by a
corresponding decrease in propene selectivity. It can therefore be
concluded that in relative terms metathetic sites deteriorate some-

374
T. Lehmann et al. / Catalysis Communications 12 (2011) 368–374
[11] K. Ikeda, Y. Kawamura, T. Yamamoto, M. Iwamoto, Catal. Commun. 9 (2008)
106–110.
[12] M. Iwamoto, Catal. Surv. Asia 12 (2008) 28–37.
[18] S.-C. Shen, S. Kawi, J. Phys. Chem. B 103 (1999) 8870–8876.
[19] R. Gómez-Reynoso, J. Ramírez, R. Nares, R. Luna, F. Murrieta, Catal. Today 107–108
(2005) 926–932.
[13] M. Iwamoto, Y. Kosugi, J. Phys. Chem. C 111 (2007) 13–15.
[14] S. Sugiyama, Y. Kato, T. Wada, S. Ogawa, K. Nakagawa, K.-I. Sotowa, Top. Catal. 53
(2010) 550–554.
[15] M. Gómez-Cazalilla, J.M. Mérida-Robles, A. Gurbani, E. Rodríguez-Castellón, A.
Jiménez-López, J. Solid State Chem. 180 (2007) 1130–1140.
[16] P. Patrono, A. La Ginestra, G. Ramis, G. Busca, Appl. Catal. A 107 (1994) 249–266.
[17] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt, C.T.-W.
Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B. Higgins, J.L. Schlenker, J. Am.
Chem. Soc. 114 (1992) 10834–10843.
[20] R. Nares, J. Ramírez, A. Gutíerrez-Alejandre, R. Cuevas, Ind. Eng. Chem. Res. 48
(2009) 1154–1162.
[21] M.M.L. Ribeiro Carrott, A.J. Estêvão Candeias, P.J.M. Carrott, K.K. Unger, Langmuir
15 (1999) 8895–8901.
[22] K.J. Ivin, J.C. Mol, Olefin Metathesis and Metathesis Polymerization, Academic
Press, San Diego, 1997.
[23] M.V. Baker, D.H. Brown, B.W. Skelton, A.H. White, Aust. J. Chem. 55 (2002)
655–660.
[24] K. Ikeda, T. Yamamoto, M. Iwamoto, Nippon Kagakkai Koen Yokoshu 86 (2006) 185.