[(SiO)TaVCl2Me2]: A well-defined silica-supported tantalum(V) surface complex as catalyst precursor for the selective cocatalyst-free trimerization of ethylene

Article abstract of DOI:10.1002/anie.201206272

On the surface of it: In the absence of co-catalyst, a well-defined silica-supported surface organometallic complex [(SiO)Ta ^VCl₂Me₂] selectively catalyzes the oligomerization of ethylene. The use of surface organometallic species allows three different pathways to be determined for the reduction of Ta^V to Ta^III species under pressure of ethylene. Copyright

Full text of DOI:10.1002/anie.201206272

.
Angewandte
Communications
DOI: 10.1002/anie.201206272
Surface Organometallics
V
ꢀ
[
( SiO)Ta Cl Me ]: A Well-Defined Silica-Supported Tantalum(V)
2
2
Surface Complex as Catalyst Precursor for the Selective Cocatalyst-
Free Trimerization of Ethylene**
Yin Chen, Emmanuel Callens, Edy Abou-Hamad, Nicolas Merle, Andrew J. P. White,
Mostafa Taoufik, Christophe Copꢀret, Erwan Le Roux,* and Jean-Marie Basset*
Over the last 50 years, the production of linear a-olefins by
ular analogues, which is in particular due to the absence of
bimolecular reactions between supported metal complexes.
[8]
ethylene oligomerization has gained increasing interest in
[
1,2]
industrial and academic research.
Currently, numerous
When considering the reaction intermediates, and in the view
of developing well-defined silica supported species, we
studies in this field have been reported and developed into
industrial processes: titanium-based catalysts for the dimeri-
[9]
targeted the immobilization of Me TaCl2 onto an inorganic
3
[
3]
zation of ethylene (Alpha-butol, IFP); chromium-based
catalysts for the trimerization of ethylene (Phillips Petro-
carrier, silica, by surface organometallic chemistry (SOMC)
as a catalyst precursor.
[
4]
leum), and more recently chromium-bearing PNP ligand for
Herein, we report the synthesis and the surface character-
ization of the grafted organometallic species (ꢀSiO)-
[
5]
ethylene tetramerization (SASOL). Along with these now-
classical systems, which require co-catalysts such as MAO,
Sen also reported the use of tantalum pentachloride in
combination with an alkylating agent, such as SnMe , ZnMe ,
V
Ta Cl Me 2 in view of its application in ethylene oligome-
2
2
rization. Furthermore, mechanistic studies on this selective
catalytic process have been successfully achieved thanks to
the dynamic reactor used. They indicate three different
pathways for the initiation process.
4
2
AlMe , or MeLi. In this case, the catalysts is assumed to be
3
III
Ta species formed in situ by the reduction of TaMe Cl in the
2
3
[6]
presence of ethylene. In this context, Mashima and co-
workers have shown that Ta active species can be alter-
natively formed by reduction of Ta Cl by 3,6-bis(trimethyl-
SBA-15 was selected because of its ordered mesoporous
network with large surface area (Supporting Information,
III
V
[10]
Figures S1, S2). This porous silica was subjected to partial
5
[
7]
silyl)-1,4-cyclohexadiene derivatives. However the catalytic
performances of these systems require the addition of co-
catalysts. In several instances, site isolation on the oxide
surface has been highly beneficial to the design of efficient
catalysts, compared to inactive or rapidly deactivating molec-
dehydroxylation under vacuum at 7008C to afford SBA-
15(700), which features mostly isolated silanols, as indicated on
À1
the IR spectrum by the characteristic sharp peak at 3747 cm
(Supporting Information, Figure S3). SBA-15(700) was reacted
with TaCl Me 1 (Supporting Information, Figure S4), and the
2
3
resulting powder was characterized to determine the organ-
ometallic species on the surface prior to its catalytic
[
*] Y. Chen, E. Callens, E. Abou-Hamad, J.-M. Basset
KAUST Catalysis Center
King Abdullah University of Science and Technology
Thuwal 23955-6900 (Kingdom of Saudi Arabia)
E-mail: jeanmarie.basset@kaust.edu.sa
[
11]
application. Elemental analysis gave 14.8% Ta, 1.65% C,
and 0.39% H, with a ratio of Ta/C/Cl = 1:1.97:1.88 (theoret-
ical: Ta/C/Cl = 1:2:2).
1
H-MAS NMR spectrum of 2 unexpectedly displays two
N. Merle, M. Taoufik, C. Copꢀret
major signals at 1.27 ppm and 0.85 ppm with a broad peak at
Chimie Organometallique de Surface/Laboratoire de Chimie,
Polymeres et Procedes UMR-C2P2-5265CNRS/ESCPE-Lyon/UCBL)
ESCPE Lyon
F-308-43 Boulevard du 11 Novembre 1918
F-69616, Villeurbanne Cedex (France)
1
.90 ppm and a very weak signal at À0.03 ppm, which is
probably due to methane or a trace amount of ꢀSiMe (see
below for further comments, and the Supporting Information,
Figure S5). The NMR signal at 1.9 ppm most likely corre-
sponds to the small amount of unreacted silanols, in agree-
ment with IR spectroscopy results. Two peaks appear at 1.27
and 0.85 ppm that would be consistent with two inequivalent
methyl groups coming from one species or indicating the
presence of two distinct species.
A. J. P. White
Department of Chemistry, Imperial College London
Exhibition Road, London SW7 2AZ, England (UK)
E. Le Roux
Department of Chemistry, University of Bergen
Realfagbygget, Allegaten 41, 5007, Bergen (Norway)
E-mail: Erwan.leRoux@kj.uib.no
Proton double (DQ)- and triple (TQ)-quantum correla-
tion spectra under 22 kHz MAS (Supporting Information,
Figure S6) confirm that these two signals correspond to
methyl groups, most likely from two different species in view
C. Copꢀret
Department of Chemistry, ETH Zurich
HCI H 229 Wolfgang-Pauli-Strasse 10, 8093, Zurich (Switzerland)
[
12]
of the absence of correlation of diagonal peak. Autocorre-
lation peaks are observed on the diagonal of the 2D DQ
spectrum for all the protons (notably, this shows that
unreacted silanols are in close proximity to each other and
most likely located in micropores). A strong autocorrelation
[
**] We are indebted to Prof. Lyndon Emsley for his enthusiastic
discussion and suggestions. The work is supported by KAUST and
SABIC.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201206272.
1
1886
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 11886 –11889

Angewandte
Chemie
peak is observed in the TQ spectrum for both resonances
attributed to methyl groups (0.85, 1.27 in F2 and 2.55,
coverage. This confirms the presence of two surface species,
possibly resulting from grafting on isolated silanols of differ-
ent environment and reactivity (Supporting Information,
3
.81 ppm in F1). As expected, the silanols do not show a TQ
[
13]
13
correlation. To obtain a better assignment of the structure of
the supported Ta species, a series of NMR spectroscopy
Figure S10). In fact, a CP/Static C NMR spectrum reveals
that these two methyl groups have different dynamics: the
peak at 67 ppm is much broader and is associated with slower
dynamics than the other at 61 ppm (Supporting Information,
Figure S11). This indicates that the former probably interact
more strongly with adjacent siloxane bridges and can be
depicted as 2a (Scheme 1). Variable-temperature experi-
1
3
experiments have been performed, and two batches of C-
1
3
enriched TaCl ( CH ) 1* (at 19% and 97%) onto SBA-
1
2
3 3
1
3
5(700) were synthetized. The C CP/MAS NMR spectrum of 2
displays two peaks at 61 ppm and 67 ppm (Figure 1a),
1
3
quantitative C NMR MAS gives a ratio of these two peaks
of 1:2 (Supporting Information, Figure S8).
V
Scheme 1. Synthesis of (ꢀSiO)Ta Cl Me 2.
2
2
1
3
ments of C CP/MAS NMR, with a temperature increasing
from room temperature to 1608C further support this
interpretation: a new peak at À2 ppm, attributed to methane
(
decomposition of 1) and ꢀSiÀMe, resulting from a methyl
transfer to the surface, appears at the expense of the decrease
of intensity of the peak at 67 ppm (Supporting Information,
Figure S12), while that at 61 ppm remains intact.
The species 2a associated with the signal at 67 ppm most
likely reacts with adjacent siloxane bridges (Supporting
Information, Scheme S1). These results are in agreement
with the presence of two species differing by the presence
[
13,14]
(2a) or not of adjacent siloxane bridges (2b) (Scheme 1).
After extensive solid-state NMR studies to establish the
structure of the grafted organometallic species, the activity of
catalyst 2 for ethylene oligomerization was first investigated
in a semi-batch autoclave in the liquid phase (Table 1). Under
these conditions, this supported catalyst 2 is active for the
ethylene oligomerization without co-catalysts or additives,
with comparable productivity for 1-hexene as compared to
1
3
Figure 1. 1D C PMAS spectrum (a) and contour plot (b) of the
1
13
V
aliphatic region of a H– C HETCOR spectrum of (ꢀSiO)Ta Cl Me
2
2
recorded with a contact time of 0.5 ms. (Experimental details are given
in the Supporting Information.)
[6]
the homogenous system. Increasing the temperature to
1
008C gave the best productivity of 1-hexene, TON 375
The 2D HETCOR spectrum with short contact time
(
0.5 ms) shows a clear correlation between the methyl protons
Table 1: Oligomerization of ethylene catalyzed with 2.
and these two carbon atoms, allowing the assignment of the
carbon–proton pairs to the individual methyl groups
(
[
a]
[e]
Entry Pressure
bar)
T
1-Butene
1-Hexene
Activity
[
b]
[c,d]
(
[8C] [wt%]
[wt%]
Figure 1). Long-range 2D HETCOR with contact times
between 0.5 and 8 ms shows no additional correlation peaks
arising from longer-range dipolar through-space interactions
1
2
3
4
5
6
7
8
9
50
50
50
50
50
50
40
30
20
60
70
80
90
6.0
5.4
5.9
7.3
9.6
7.8
7.1
88.3 (98.1)
82.6 (98.8)
84.3 (99.2)
83.8 (99.0)
82.7 (99.1)
82.6 (99.3)
83.1 (97.7)
80.5 (97.6)
77.8 (97.1)
160
242
322
332
375
273
270
200
148
(
Supporting Information, Figure S9a), confirming that these
two peaks correspond to two different species. When a very
long contact time was employed (10 ms), these peaks start to
correlate, suggesting the proximity of some of the species
100
110
100
100 11.1
100 7.3
(
Supporting Information, Figure S9b). Moreover, grafting
1
3
a varying amount of [TaCl ( CH ) ] 1* per surface silanol (25/
2
3 3
5
0/98% coverage) provided further evidence for the presence
[
a] Reaction conditions: 50 mg of 2, 10 mL of toluene, 30 min. [b] Only 1-
of two distinct species. At low coverage, the peak at 61 ppm
has a greater intensity than the peak at 67 ppm in the
butene was detected. [c] Selectivity of 1-hexene in products. [d] Values
in parentheses are selectivity in 1-hexene from C . [e] Turnovers
mol(ethylene)/mol(Ta)/h.
6
1
3
C NMR spectrum, while the situation is reverse at high
Angew. Chem. Int. Ed. 2012, 51, 11886 –11889
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11887

.
Angewandte
Communications
(
58 g1-hexene/g h) at an optimal pressure of 50 bar of ethylene
Ta
(
Table 1, entries 1–5 and 7–10; Supporting Information, Fig-
ure S13). Performing the catalytic test at 1108C led to
a decrease of productivity (Table 1, entry 6).
Moreover, the productivity for 1-hexene is linearly
correlated to the pressure of ethylene indicating a first-
order reaction (Supporting Information, Figure S14, Scheme
S2). The major oligomer produced is 1-hexene (> 80%) along
with 1-butene (< 10%) and polyethylene. Solid-state NMR
analysis of the residual catalyst at the end of the reaction
showed mainly polyethylene, which may be the cause of the
catalyst ageing (Supporting Information, Figure S15).
III
Previous results postulated Ta intermediates, which
leads, via oxidative coupling of ethylene, to tantalacyclopen-
tane(V), a key intermediate for the oligomerization reac-
[
6,7]
tion.
However, to date, there is no direct evidence for the
formation of this intermediate. Therefore, we investigated the
V
conversion of Ta precursor into the active species using
a dynamic bed reactor coupled with a GC detector in situ to
monitor this reduction step closely.
À1
Contacting 2 with ethylene (12 bar, 0.42 mLmin , ca.
0
.18 mol of ethylene/mol·Ta·min) at 708C in a flow bed
reactor gave with time a mixture of methane, propylene,
ethane, and butane resulting from a reductive elimination
step (Figure 2). Within 150 min, dimerization of ethylene
Scheme 2. Proposed mechanism pathways for the formation of
butane, ethane, methane, propylene, and active Ta surface species.
III
mediate 3 (path a) and 4 (path b) respectively, and which also
leads to 6.
Additionally, analysis of the quantity of the four gases
formed during activation (up to 1050 min) showed that
butane/ethane/propylene were generated in a 1/1.11/1.33
ratio by three competitive pathways, and the amount of
products (52.1 mmol of CH4, 38.6 mmol of propylene,
3
2.0 mmol of C H , and 28.9 mmol of butane, accordingly, the
2 6
methane generated from deactivation pathway and ethylene
disproportionation pathway has a ratio around 1:3) with
V
respect to Ta (110.9 mmol) shows that about 90% of Ta
Figure 2. Initial products observed of the reaction of 2 with ethylene in
a dynamic reactor.
species have been converted into active species 6 (Supporting
Information, Figure S17 and S18). This species is converted,
in the presence of ethylene, into the key putative tantalacycle
intermediates, which finally yields 1-alkenes, that is, 1-butene
[
6b]
started and gave mainly 1-butene and traces of 1-hexene in
flow conditions. After 450 min the activation process was
nearly finished, and the generation of these four gases
stopped. Moreover, the catalyst produces 1-butene as the
major product during an extended period. After 5000 min,
and 1-hexene in gas versus condensed-phase conditions.
In conclusion, a supported catalyst for the ethylene
oligomerization has been designed by a molecular approach;
it operates efficiently without co-catalyst delivering 1-hexene
in good selectivity in a liquid phase and 1-butene in a gas
1
10 mol of butene per mol Ta atom were produced (Support-
phase. The use of a well-defined catalyst precursor
V
ing Information, Figure S16). All four gases observed in the
activation process are consistent with the formation of an
Ta
(ꢀSiO)Ta Cl Me and monitoring the initiation products
2
2
V
show clearly that ethylene initially converts the Ta catalyst
precursor into the formally reduced Ta , which can then
(
III)
III
intermediate species through competitive pathways
(
Scheme 2). The process probably starts with the coordination
generate the key metallacycle intermediates, and thus the
of ethylene; it can insert into the TaÀMe bond to generate 4,
which by a b-hydrogen elimination would produce first 5 and
then 6 via a subsequent reductive elimination of methane
selective production of a-olefins.
[
6]
(
path c). We also observed ethane and butane, which are
Received: August 5, 2012
presumably formed by reductive elimination from the inter-
Published online: October 22, 2012
1
1888 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 11886 –11889

Angewandte
Chemie
Keywords: ethylene · oligomerization ·
organometallic compounds · surface reactions · tantalum
b) M. Yanagawa, S. Oda, K. Mashima (Sumitomo Chemical Co),
.
JP 2006/051489 (WO Patent 2006/006433), 2006.
8] a) J.-M. Basset, R. Psaro, D. Roberto, R. Ugo, Modern Surface
Organometallic Chemistry, Wiley-VCH, Weinheim, 2009; b) C.
Copꢀret, Chem. Rev. 2010, 110, 656 – 680; c) S. L. Wegener, T. J.
Marks, P. C. Stair, Acc. Chem. Res. 2012, 45, 206 – 214, and
references therein.
[
[
[
1] a) D. S. McGuinness, Chem. Rev. 2011, 111, 2321 – 2341; b) T.
Agapie, Coord. Chem. Rev. 2011, 255, 861 – 880; c) P. W. N. M.
van Leeuwen, N. D. Clꢀment, M. J.-L. Tschan, Coord. Chem.
Rev. 2011, 255, 1499 – 1517; d) J. Skupinska, Chem. Rev. 1991, 91,
9] a) P. R. Sharp, D. Astruc, R. R. Schrock, J. Organomet. Chem.
1
979, 182, 477 – 510; b) R. R. Schrock, G. W. Parshall, Chem.
Rev. 1976, 76, 243 – 368; c) R. R. Schrock, J. Organomet. Chem.
976, 122, 209 – 225; d) R. R. Schrock, P. R. Sharp, J. Am. Chem.
6
13 – 648; e) H. Wittcoff, B. G. Reuben, J. S. Plotkin, Industrial
nd
Organic Chemicals, 2 ed., Wiley-Interscience, New York, 2004.
2] Y. Shaikh, K. Albahily, M. Sutcliffe, V. Fomitcheva, S. Gambar-
otta, I. Korobkov, R. Duchateau, Angew. Chem. 2012, 124, 1395 –
1
[
[
Soc. 1978, 100, 2389 – 2399; e) R. R. Schrock, L. W. Messerle,
C. D. Wood, L. J. Guggenberger, J. Am. Chem. Soc. 1978, 100,
1
398; Angew. Chem. Int. Ed. 2012, 51, 1366 – 1369.
3
1
1
6
793 – 3800; f) R. R. Schrock, J. D. Fellmann, J. Am. Chem. Soc.
978, 100, 3359 – 3370; g) R. R. Schrock, Acc. Chem. Res. 1979,
2, 98 – 104; h) R. R. Schrock, J. Am. Chem. Soc. 1974, 96, 6796 –
797.
3] a) Y. Chauvin, D. Commereuc, Y. Glaize (IFP), European
Patent No. 0200654, 1989; b) D. Commereuc, J. Gaillard, G.
Leger (IFP), French Patent No. 2540488, 1984; c) K. Ziegler, H.
Martin, U.S. Patent 2943125, 1960.
[
10] a) R. Anwander, Chem. Mater. 2001, 13, 4419 – 4438; b) C.
Copꢀret, M. Chabanas, R. Petroff Saint-Arroman, J.-M. Basset,
Angew. Chem. 2003, 115, 164 – 191; Angew. Chem. Int. Ed. 2003,
[
4] a) W. K. Reagen, B. K. Conroy (Phillips Petroleum), EP 416403,
1990; b) J. T. Dixon, M. J. Green, F. M. Hess, D. H. Morgan, J.
Organomet. Chem. 2004, 689, 3641 – 3668; c) J. R. Briggs (Union
Carbide), U.S. Patent 4,668,838, 1987; d) M. Tamura, K. Iwa-
naga, Y. Ito (Sumitomo Chemical), EP 699 648, 1995; e) A.
Carter, S. A. Cohen, N. A. Cooley, A. Murphy, J. Scutt, D. F.
Wass, Chem. Commun. 2002, 858 – 859.
42, 156 – 181.
[
11] The structure of 1 was found to be similar as that previously
reported by Sattler et al. The complex has crystallographic D3h
symmetry about an axis that passes through Cl1 and Ta1. As
a consequence of this symmetry, the Cl-Ta-Cl, C-Ta-C, and Cl-
Ta-C angles are exactly 180, 120 and 908, respectively. See: A.
Sattler, S. Ruccolo, G. Parkin, Dalton Trans. 2011, 40, 7777 –
[
5] a) K. Blann, A. Bollmann, J. T. Dixon, A. Neveling, D. H.
Morgan, H. Maumela, E. Killian, F. Hess, S. Otto, L. Pepler, H.
Mahomed, M. Overett (Sasol Technology), WO Patent
7782.
04056479A1, 2004; b) A. H. Tullo, Chem. Eng. News 2005,
[
12] H. Geen, J. J. Titman, J. Gottwald, H. W. Spiess, Chem. Phys.
Lett. 1994, 227, 79 – 86.
13] a) M. Taoufik, A. de Mallmann, E. Prouzet, G. Saggio, J.
Thivolle-Cazat, J.-M. Basset, Organometallics 2001, 20, 5518 –
83(43), 30 – 33; c) A. Bollmann, K. Blann, J. T. Dixon, F. M.
Hess, E. Killian, H. Maumela, D. S. McGuinness, D. H. Morgan,
A. Neveling, S. Otto, M. Overett, A. M. Z. Slawin, P. Wassersc-
heid, S. Kuhlmann, J. Am. Chem. Soc. 2004, 126, 14712 – 14713;
d) D. S. McGuinness, A. J. Rucklidge, R. P. Tooze, A. M. Z.
Slawin, Organometallics 2007, 26, 2561 – 2569.
[
5
2
521; b) D. Gajan, C. Copꢀret, New J. Chem. 2011, 35, 2403 –
408.
[
14] Methyl transfer from metal to silicon has already been observed
when grafting silica dehydroxylated at high temperature (7008C)
and having strained siloxane bridges (metal – methyl, Groups 4
and 5). See: H. Ahn, T. J. Marks, J. Am. Chem. Soc. 2002, 124,
[
6] a) C. Andes, S. B. Harkins, S. Murtuza, K. Oyler, A. Sen, J. Am.
Chem. Soc. 2001, 123, 7423 – 7424; b) Z.-X. Yu, K. N. Houk,
Angew. Chem. 2003, 115, 832 – 835; Angew. Chem. Int. Ed. 2003,
42, 808 – 811.
7103 – 7110.
[
7] a) R. Arteaga-Mꢁller, H. Tsurugi, T. Saito, M. Yanagawa, S.
Oda, K. Mashima, J. Am. Chem. Soc. 2009, 131, 5370 – 5371;
Angew. Chem. Int. Ed. 2012, 51, 11886 –11889
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11889