A selective chromium catalyst system for the trimerization of ethene and its coordination chemistry

Article abstract of DOI:10.1002/ejic.201000044

In this paper a novel ligarid of the type [PNPNH] is presented for the application in a now homogeneous highly selective ethene trimerization system for the formation of 1-hexene, which consists of the chromium source CrCl ₃(thf)₃, the ligand Ph₂PN(iPr)P(Ph)N(iPr)H (1), and Et₃Al as an activator in toluene. The excellent characteristics of this new system, e.g. very high selectivity to C₆ with highest purity of the C₆ fraction (>99% 1-hoxone), activity on a constant: level on a long timescale, use of small amounts of Et₃Al as a cheap activator, and only very low production of PE, make it to a hot candidate for industrial application. Its organometallic background gives an indication of the nature of the active catalyst species.

Full text of DOI:10.1002/ejic.201000044

SHORT COMMUNICATION
DOI: 10.1002/ejic.201000044
A Selective Chromium Catalyst System for the Trimerization of Ethene and Its
Coordination Chemistry
Stephan Peitz,[a] Normen Peulecke,[a] Bhaskar R. Aluri,[a] Sven Hansen,[a]
Bernd H. Müller,[a] Anke Spannenberg,[a] Uwe Rosenthal,*[a] Mohammed H. Al-Hazmi,[b]
Fuad M. Mosa,*[b] Anina Wöhl,[c] and Wolfgang Müller*[c]
Keywords: Aminophosphorus ligands / Amido ligands / Chromium / Homogeneous catalysis / Oligomerization / Selectivity
In this paper a novel ligand of the type [PNPNH] is presented
for the application in a new homogeneous highly selective
ethene trimerization system for the formation of 1-hexene,
which consists of the chromium source CrCl3(thf)3, the ligand
Ph2PN(iPr)P(Ph)N(iPr)H (1), and Et3Al as an activator in tolu-
ene. The excellent characteristics of this new system, e.g.
very high selectivity to C6 with highest purity of the C6 frac-
tion (Ͼ99% 1-hexene), activity on a constant level on a long
timescale, use of small amounts of Et3Al as a cheap activator,
and only very low production of PE, make it to a hot candi-
date for industrial application. Its organometallic background
gives an indication of the nature of the active catalyst spe-
cies.
In this paper we describe the synthesis of the novel ligand
1 of the type [PNPNH], its complexation (Scheme 1) and
metalation as well as its application as a new homogeneous
selective trimerization system to 1-hexene consisting of the
chromium source CrCl3(thf)3, the ligand Ph2PN(iPr)P(Ph)-
N(iPr)H (1) and Et3Al as an activator in toluene.[5]
Introduction
Linear α-olefins (LAOs) are useful and versatile interme-
diates for many industrially important substances like co-
monomers for HDPE or LLDPE, alcohols, aldehydes, car-
boxylic acids or sulfonates. Conventional full-range pro-
ducers of LAOs, which obtain a wide product distribution,
have to meet a formidable challenge to match the market
demand.[1] Conventional technologies, such as the SHOP
(Shell Higher Olefin Process) or the α-Sablin process, are
based on an ethene insertion/β-elimination mechanism by
chain growth, resulting in a Schulz–Flory distribution.
Thus, the selective production of the economically most via-
ble LAOs, i.e. comonomer-grade 1-hexene and more re-
cently 1-octene, appears highly desirable.[2] For selective eth-
ene trimerization, instead of the chain-growth pathway, a
metallacycle mechanism is generally accepted. In this pro-
cess, two ethene units coordinate at the catalytically active
site and form a metallacyclopentane by oxidative coupling.
Insertions of further ethene units result in larger metalla-
cycles. Through β-elimination/reductive elimination, LAOs
are formed.[3,4]
Scheme 1. Formation of complex 2 from ligand 1.
Results and Discussion
For easier access to kinetic parameters and mechanistic
observations, the average productivity of this new trimeri-
zation system was held constant at around 5200 g/(gCr·h)
(determined after 2 h), but with some smaller modifications
an increase up to 27000 g/(gCr·h) and more was achieved.[5c]
The results of a typical ethene oligomerization experi-
ment[6] concerning selectivity are given in Table 1. In vari-
ous experiments it could be shown that the activity, the
achieved selectivity to C6, and the purity of the C6 fraction
(Ͼ99 wt.-% of 1-hexene) of this new trimerization catalyst
system are very promising. The high purity within the C6
fraction has the advantage that it can be used without fur-
ther purification in co-polymerization. Additionally, due to
[a] Leibniz-Institut für Katalyse an der Universität Rostock e.V.
Albert-Einstein-Str. 29 A, 18059 Rostock, Germany
E-mail: Uwe.rosenthal@catalysis.de
[b] Saudi Basic Industries Corporation
P. O. Box 42503, Riyadh 11551, Saudi Arabia
[c] Linde AG, Linde Engineering Division
Dr.-Carl-von-Linde-Str. 6–14, 82049 Pullach, Germany
E-mail: Wolfgang.mueller@linde-le.com
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201000044.
Eur. J. Inorg. Chem. 2010, 1167–1171
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1167

U. Rosenthal, F. M. Mosa, W. Müller et al.
SHORT COMMUNICATION
the high catalytic stability, the productivity stays at a con-
stant level for a long time in comparison to other sys-
tems.[2i–2l,2q–2s] Also, the use of Et3Al as activator is advan-
tageous, due to lower costs and its more precise specifica-
tion compared to MAO (methylalumoxane). Only low ra-
tios of co-catalyst/Cr are necessary to activate the new sys-
tem, whereas for many other catalyst systems molar ratios
of Al/Cr up to 850 are employed.[2j–2l] Moreover, a very low
amount of PE (polyethylene) is produced during the cata-
lytic trimerization reaction. Given the advantages, this
novel trimerization catalyst system competes very well with
comparable systems, especially in view of a potential imple-
mentation in a technical process. In particular, the moderate
average activities may allow for a relatively simple design
on a technical scale, due to easier control of the heat release
of the exothermal reaction.
Figure 1. Molecular structure of 2 with thermal ellipsoids set at
30% probability. The asymmetric unit contains (besides solvent
molecules) two molecules of 2; only one molecule is depicted. Hy-
drogen atoms, except H1A, have been omitted for clarity. Selected
bond lengths [Å] and angles [°]: N1–P1 1.644(4), N2–P2 1.694(4),
N2–P1 1.717(4), Cr1–P1 2.3975(14), Cr1–P2 2.5161(14); N1–P1–
N2 110.9(2), P2–N2–P1 104.9(2), P1–Cr1–P2 66.73(4).[7]
Table 1. Selectivity of the trimerization system CrCl3(thf)3, 1, Et3Al
in toluene at 65 °C and 30 bar ethene pressure.[a]
Oligomer[b] C4 1-C4 C6 1-C6 C8 1-C8 C10 1-C10 Polymer
One has to consider that during the reaction metalation
of the N–H group by aluminum (or chromium after alky-
lation) forming an amide [PNPN]– should occur in the
herein presented new catalytic ethene trimerization system,
consisting of CrCl3(thf)3, the ligand 1 and Et3Al. Since
compounds of the structure [PNPNR2] are totally inactive
in catalysis, it seems that the N–H group is essential for
wt.-%
4.7 88.2 90.7 99.3 1.3 84.1 3.4 10.9
0.09
[a] Standard conditions: loading 0.1 mmol of CrCl3(thf)3,
0.175 mmol of Ph2PN(iPr)P(Ph)N(iPr)H, 7.0 mmol of Et3Al,
100 mL of toluene as solvent, 2 h of reaction time, 55 g ethene up-
take. [b] Selectivity for 1-alkene (1-Cn) as percentage of total Cn
fraction.
To obtain an idea of the nature of the active site and its trimerization activity.
required structural characteristics for C6 selectivity, dif-
To come to a closer look at the real catalytic system,
ferent model compounds were isolated from relevant mix- we investigated the reaction of 1 with Et3Al and Me3Al,
tures of the single components of the system and investi- respectively. By NMR investigations, we found three dif-
gated by X-ray structure analysis.
ferent steps of interaction of 1 with Et3Al.
For the synthesis of the desired [PNPNH] ligand two dif-
The first step is the formation of an adduct [Ph2PN(iPr)-
ferent synthetic methods were elaborated. According to P(Ph)N(iPr)H][AlEt3] (3a), the second step is the alumi-
these procedures, Ph2PN(iPr)P(Ph)N(iPr)H (1) was synthe- nation of the N–H function by liberation of ethane to the
sized and characterized by X-ray analysis (see Supporting aluminum amide, which additionally coordinates Et3Al to
Information).
give the complex [Ph2PN(iPr)P(AlEt3)(Ph)N(iPr)][AlEt2]
The coordination chemistry of this ligand was investi- (4a), and the third step is the rearrangement of 4a to the
gated in detail for the (monoligand)chromium trichloride cyclic compound 5a by transamidation at higher tempera-
complex [Ph2PN(iPr)P(Ph)N(iPr)H]CrCl3(thf) (2), which ture (Scheme 2).
was prepared according to Scheme 1.
Compound 5a was characterized by X-ray analysis, and
Interestingly, the intact undeprotonated N–H function of its molecular structure is shown in Figure 2.
the [PNPNH] ligand does not coordinate to the metal atom.
By using Me3Al instead of Et3Al, we were not only able
Instead, the ligand is coordinated through both phos- to crystallize the rearrangement product 5b but also the alu-
phorus atoms to the metal center (Figure 1). This coordina- minated ligand 4b as an intermediate (Figure 3). From this
tion mode is already known from the Sasol-type [PNP] li- molecular structure it becomes obvious that a molecule of
gands.[2i]
Me3Al coordinates to the central phosphorus atom.
Scheme 2. Stepwise reaction of 1 with R3Al to afford adduct [Ph2PN(iPr)P(Ph)N(iPr)H][AlR3] (3), alumination to form [Ph2PN(iPr)-
P(AlR3)(Ph)N(iPr)][AlR2] (4) and rearrangement by transamidation to give [N(iPr)P(Ph)–(Ph)2PN(iPr)][AlR2] (5) (a: R = Et; b: R = Me).
1168
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 1167–1171

Chromium Catalyst for the Trimerization of Ethene
NPPN ligand backbone. It can be described in terms of
resonance structures of a phosphane–phosphonium com-
plex or a phosphane-phosphazene cation.[8] According to
these investigations, compounds 5a and 5b can be interpre-
ted analogously (Scheme 2).
Catalytic tests with complexes 4a and 4b in the presence
of CrCl3(thf)3 and Et3Al at 65 °C show comparable activi-
ties to the in situ system concerning the slope of the ethene-
uptake curve after the initiation of the reaction, indicating
the aluminated ligand to be a precursor or a part of the
catalytic system. A trimerization experiment with the pure
rearranged ligand 5a, CrCl3(thf)3, and Et3Al complex
showed no activity at all. A catalytic run with the in situ
system at 90 °C indicates 5a to be responsible for the decline
in ethene consumption (for more details see Supporting In-
formation). This is why we regard 5a as the main species
causing derogation of catalysis at higher temperatures.
To arrive at a chromium complex with the amide-bonded
ligand, we tried CrCl3(thf)3, the preferred chromium source
in catalysis, to react with the aluminum amides 4a and 4b to
obtain a dimetallic complex that could be active in catalysis.
Unfortunately, all efforts in this direction failed so far.
Reaction of 1 with n-butyllithium and Li[CpCrCl3] in-
stead of CrCl3(thf)3 as chromium source causes elimination
of butane and 2 equiv. of LiCl and leads to the formation
of CpCrCl[N(iPr)P(Ph)N(iPr)PPh2] (6) (Scheme 3), which
bears a chromium–amide bond (Figure 4).
Figure 2. Molecular structure of 5a with thermal ellipsoids set at
30% probability. All hydrogen atoms have been omitted for clarity.
Selected bond lengths [Å] and angles [°]: P1–P2 2.2262(7), N1–P1
1.668(2), P2–N2 1.614(2), N1–Al1 1.886(2), N2–Al1 1.928(2); N1–
Al1–N2 95.90(8).
Scheme 3. Formation of 6.
Figure 3. Molecular structures of 4b and 5b with thermal ellipsoids
set at 30% probability. The asymmetric unit of 4b contains two
molecules; only one is depicted. All hydrogen atoms have been
omitted for clarity. Selected bond lengths [Å] and angles [°]: 4b:
N1–P1 1.653(2), P1–N2 1.740(2), N2–P2 1.684(2), N1–Al1
1.879(2), P2–Al1 2.5093(10), P1–Al2 2.5831(10); N2–P1–N1
105.08(10), P2–N2–P1 113.71(10), P2–Al1–N1 85.36(6); 5b: P1–P2
2.2178(4), N1–P1 1.6645(10), P2–N2 1.6191(10), N1–Al1
1.8796(11), N2–Al1 1.9303(10); N1–Al1–N2 95.79(4).
Figure 4. Molecular structure of 6 with thermal ellipsoids set at
30% probability. All hydrogen atoms have been omitted for clarity.
Selected bond lengths [Å] and angles [°]: N1–P1 1.6671(13), P1–N2
1.7558(14), N2–P2 1.6786(14), N1–Cr1 1.9650(13), P2–Cr1
2.5081(5); N2–P1–N1 102.74(7), P2–N2–P1 118.08(7), P2–Cr1–N1
82.68(4).
Complexes 5a and 5b bear an N(iPr)P(Ph)–P(Ph)2N(iPr)
ligand N,NЈ-chelated to the metal center. A quite similar
This complex can serve as a model compound for the
complex Me2Al[N(dipp)P(Ph)–P(Me)PhN(dipp)] {dipp = interaction of the metallated ligand with the chromium
2,6-(iPr)2C6H3} was published before, having a comparable atom. It was also tested in the selective oligomerization of
Eur. J. Inorg. Chem. 2010, 1167–1171
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1169

U. Rosenthal, F. M. Mosa, W. Müller et al.
SHORT COMMUNICATION
for trimerization reactions with important complexes and crystallo-
graphic information.
ethene with Et3Al as activator, but it generates 1-butene as
the main product. With MAO as a co-catalyst only PE is
produced. This different selectivity compared to the origi-
nal system is explained by the influence of the Cp ligand.
The lithiated ligand was also tested in trimerization and
shows nearly identical ethene consumption compared to the
other systems, again being proof for the involvement of a
deprotonated ligand species in catalysis (see Supporting In-
formation).
[1]
a) E. Greiner, R. Gubler, Y. Inoguchi, “Linear Alpha Olefins”,
in CEH Marketing Research Report, Chemical Economics
Handbook, SRI International, Menlo Park, SF, May 2004; b)
M. Morgan, “Alpha Olefins”, in PERP Report 02/03-4, Nexant
Chem. Systems, January 2004; c) P. Quinlan, World Petrochemi-
cal Conference, Houston, USA, March 2005; d) E. Burridge,
ICIS Chemical Business 2008.
As a result of all these experiments, we are sure that the
catalytically active species is a complex, in which the depro-
tonated [PNPNH] ligand causes C6 selectivity. With regard
to the fact that an organoaluminum and a chloride com-
pound are present in the system, bridged dimetallic com-
pounds (Scheme 4) must be considered, which fits very well
with the results recently published by Gambarotta et
al.[9a,9b] and Hor et al.[9c] for other systems.
[2]
a) W. K. Reagan (Phillips Petroleum Company), EP 0417477,
1991; b) W. K. Reagan, J. W. Freeman, B. K. Conroy, T. M.
Pettijohn, E. A. Benham (Phillips Petroleum Company), EP
0608447, 1994; c) R. D. Knudsen, J. W. Freeman, M. E. Lashier
(Phillips Petroleum Company), US 5563312, 1996; d) J. W.
Freeman, J. L. Buster, R. D. Knudsen (Phillips Petroleum
Company), US 5856257, 1999; e) W. K. Reagan, J. W. Freeman,
B. K. Conroy, T. M. Pettijohn, E. A. Benham (Phillips Petro-
leum Company), US 5451645, 1995; f) J. T. Dixon, M. J. Green,
F. M. Hess, D. H. Morgan, J. Organomet. Chem. 2004, 689,
3641 and references cited therein; g) A. Carter, S. A. Cohen,
N. A. Cooley, A. Murphy, J. Scutt, D. F. Wass, Chem. Commun.
2002, 858; h) D. F. Wass (BP Chemicals Ltd), WO 02/04119,
2002; i) A. Bollmann, K. Blann, J. T. Dixon, F. M. Hess, E.
Killian, H. Maumela, D. S. McGuinness, D. H. Morgan, A.
Nevelling, S. Otto, M. Overett, A. M. Z. Slawin, P. Wasser-
scheid, S. Kuhlmann, J. Am. Chem. Soc. 2004, 126, 14712; j)
D. S. McGuinness, P. Wasserscheid, W. Keim, C. Hu, U. En-
glert, J. T. Dixon, C. Grove, Chem. Commun. 2003, 334; k) D. S.
McGuinness, P. Wasserscheid, W. Keim, D. Morgan, J. T. Di-
xon, A. Bollmann, H. Maumela, F. Hess, U. Englert, J. Am.
Chem. Soc. 2003, 125, 5272; l) D. S. McGuinness, P. Wasser-
scheid, D. H. Morgan, J. T. Dixon, Organometallics 2005, 24,
552; m) P. R. Elowe, C. McCann, P. G. Pringle, S. K. Spitz-
messer, J. E. Bercaw, Organometallics 2006, 25, 5255; n) J.
Zhang, P. Braunstein, T. S. A. Hor, Organometallics 2008, 27,
4277; o) D. F. Wass, Dalton Trans. 2007, 816 and references
cited therein; p) K. Blann, A. Bollmann, H. de Bod, J. T. Di-
xon, E. Killian, P. Nongodlwana, M. C. Maumela, H. Mau-
mela, A. E. McConnell, D. H. Morgan, M. J. Overett, M.
Prétorius, S. Kuhlmann, P. Wasserscheid, J. Catal. 2007, 249,
244; q) H. Mahomed, A. Bollmann, J. T. Dixon, V. Gokul, L.
Griesel, C. Grove, F. Hess, H. Maumela, L. Pepler, Appl. Catal.
A: Gen. 2003, 255, 355; r) D. H. Morgan, S. L. Schwikkard,
J. T. Dixon, J. J. Nair, R. Hunter, Adv. Synth. Catal. 2003, 345,
939; s) T. Monoi, Y. Sasaki, J. Mol. Catal. A 2002, 187, 135.
a) R. M. Manyik, W. E. Walker, T. P. Wilson (Union Carbide
Corporation), US 3300458, 1967; b) R. M. Manyik, W. E.
Walker, T. P. Wilson, J. Catal. 1977, 47, 197; c) J. R. Briggs, J.
Chem. Soc., Chem. Commun. 1989, 674.
Scheme 4. Dinuclear model for the catalytic active species.
Conclusions
We have reported the development of a new system for
the chromium-catalyzed selective trimerization of ethene,
accompanied by its organometallic background, which
gives an indication of the nature of the active catalyst spe-
cies. The excellent characteristics of this new trimerization
system, e.g. high selectivity to C6 with highest purity of the
C6 fraction (Ͼ99% of -hexene), activity on a constant level
on a long timescale, use of small amounts of Et3Al as a
cheap activator and only very low production of PE, make
it a hot candidate for industrial application with many ad-
vantages compared to other systems.
[3]
[4]
[5]
a) T. Agapie, S. J. Schofer, J. A. Labinger, J. E. Bercaw, J. Am.
Chem. Soc. 2004, 126, 1304; b) T. Agapie, J. A. Labinger, J. E.
Bercaw, J. Am. Chem. Soc. 2007, 129, 14281.
Experimental Section
a) A. Wöhl, B. H. Müller, M. Hapke, N. Peulecke, U. Rosen-
thal, W. Müller, F. Winkler, A. Wellenhofer, H. Bölt, P. M.
Fritz, M. Al-Hazmi, V. Aliyev. F. Mosa (Linde AG/Sabic), WO
09/006979, 2009; b) A. Wöhl, B. H. Müller, M. Hapke, N. Peu-
lecke, U. Rosenthal, W. Müller, F. Winkler, A. Wellenhofer, H.
Bölt, P. M. Fritz, M. Al-Hazmi, V. Aliyev. F. Mosa (Linde AG/
Sabic), WO 09/068157, 2009; c) unpublished results, submitted
as patent: A. Wöhl, B. H. Müller, N. Peulecke, S. Peitz, B. R.
Aluri, U. Rosenthal, W. Müller, A. Meiswinkel, H. Bölt, M.
Al-Hazmi, M. Al-Masned, K. Al-Eidan, F. Mosa (Linde AG/
Sabic), 2009.
A. Wöhl, W. Müller, N. Peulecke, B. H. Müller, S. Peitz, D.
Heller, U. Rosenthal, J. Mol. Catal. A 2009, 297, 1.
Further characterization of all compounds and detailed dis-
cussion of their molecular structures will follow in a separate
article.
All air- and moisture-sensitive compounds were handled under ar-
gon by using standard Schlenk techniques or in a glove box. Trieth-
ylamine (99% purity), isopropylamine (99% purity), n-butyllithium
solution (2.5  in heptanes) were obtained from Acros Organics
and used as received. Chlorodiphenylphosphane (98% purity),
CrCl3(thf)3 (97% purity), Et3Al (1.9 mol/L in toluene) and Me3Al
(1.0  Me3Al in toluene) were obtained from Sigma–Aldrich and
used without further purification. Toluene (Ͼ99.9% purity), thf
and n-hexane were obtained from Acros Organics, dried with so-
dium/benzophenone and then distilled under argon. Argon 5.0 and
ethene 3.0 were purchased from Linde Gas and used as received.
[6]
[7]
Supporting Information (see footnote on the first page of this arti-
cle): Experimental details for the preparation of all new compounds
(including detailed analytical data); detailed procedures and results
1170
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 1167–1171

Chromium Catalyst for the Trimerization of Ethene
[8] T. Chivers, M. C. Copsey, M. Parvez, Chem. Commun. 2004,
2818.
[9] a) I. Vidyaratne, G. B. Nikiforov, S. I. Gorelsky, S.
Gambarotta, R. Duchateau, I. Korobkov, Angew. Chem. Int.
Ed. 2009, 48, 6552; b) A. Jabri, C. B. Mason, Y. Sim, S.
Gambarotta, T. J. Burchell, R. Duchateau, Angew. Chem. Int.
Ed. 2008, 47, 9717; c) J. Zhang, A. Li, T. S. A. Hor, Organome-
tallics 2009, 28, 2935.
Received: January 18, 2010
Published Online: February 9, 2010
Eur. J. Inorg. Chem. 2010, 1167–1171
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1171