Ethylene tetramerization with cationic chromium(I) complexes

Article abstract of DOI:10.1021/om0701975

Various [Cr(CO)₄(P?P)]⁺ complexes have been synthesized and tested for the tetramerization of ethylene to give principally 1-octene. The effect of the anion has been investigated, and it has been found that for successful catalysis a

Full text of DOI:10.1021/om0701975

2782
Organometallics 2007, 26, 2782-2787
Ethylene Tetramerization with Cationic Chromium(I) Complexes
Adam J. Rucklidge,†,§ David S. McGuinness,†, Robert P. Tooze,† Alexandra M. Z. Slawin,‡
Je´re´mie D. A. Pelletier,† Martin J. Hanton,*,† and Paul B. Webb†
Sasol Technology (UK) Ltd., Purdie Building, North Haugh, St. Andrews, KY16 9ST, U.K., and School of
Chemistry, UniVersity of St. Andrews, Purdie Building, St. Andrews, KY16 9ST, U.K.
ReceiVed March 2, 2007
Various [Cr(CO)4(P∩P)]+ complexes have been synthesized and tested for the tetramerization of ethylene
to give principally 1-octene. The effect of the anion has been investigated, and it has been found that for
successful catalysis an extremely weakly coordinating anion is required. This study shows that when
[Al(OC(CF3)3)4]- is employed as the anion, along with triethylaluminum and PNPiPr as the coordinated
diphosphine, the system is active for the tetramerization of ethylene. This is the first example of CrI
being used as a catalyst precursor for selective tetramerization and gives evidence toward the catalytic
cycle acting through a CrI f CrIII cationic mechanism.
Scheme 1. Cationic Metallocycle Mechanism Proposed for
the Cr/PNP Ethylene Tetramerization System
Introduction
In recent years there has been a considerable amount of work
carried out on both the tri- and tetramerization of ethylene to
produce 1-hexene and 1-octene, respectively.1 For the most part,
this has been due to increasing demand for these olefins as
comonomers in the polymer industry. A variety of highly active
Cr-based systems have been developed for both trimerization2-4
and tetramerization,5 as well as a number of trimerization
catalysts based upon other early transition metals.6,7 Sasol has
developed a highly active tetramerization system with the PNP
family of ligands and methylaluminoxane (MAO) as a cocata-
lyst. It has been shown by our research group, through the use
of deuterium labeling, that the catalysis proceeds via a metal-
locycle mechanism.8 It is generally believed that the cocatalyst
MAO acts upon the Cr precursor to generate a cationic species
as the catalytically active complex. Thus, in the highly active
Cr/PNPR (PNP ) Ph2PN(R)PPh2) system it is expected that the
generated species must be in oxidation state +1, as the PNP is
believed to remain neutral after the system has been activated.
This would lead to a cationic CrI f CrIII mechanism as shown
in Scheme 1. Bercaw and co-workers have also provided
evidence for a metallocyclic CrI f CrIII mechanism for related
Cr-PNP trimerization catalysts.9
* Corresponding author: E-mail: martin.hanton@uk.sasol.com.
† Sasol Technology (UK) Ltd.
§ Current address: Scientific and Medical Products Ltd., Shirley House,
12 Gatley Rd., Cheadle, Cheshire, SK8 1PY.
In the tetramerization process a high MAO to Cr ratio is
required, which is undesirable due to the high cost of MAO.
Considerable research has taken place to find alternative
cocatalysts that display similar catalytic productivity and
selectivity to MAO. We recently reported the use of [Ph3C]-
[B(C6F5)4] in combination with triethylaluminum, AlEt3, as a
replacement for MAO.10 While successful in terms of generating
active tetramerization systems, with liquid fraction selectivities
similar to that obtained with MAO, this alternative activator
system generated catalysts with very short lifetimes. This short
catalyst lifetime is due to very fast exchange reactions between
the borate anion and excess trialkylaluminum, which leads to
rapid anion degradation, a more full discussion of this topic
having been disclosed previously.10
Current address: School of Chemistry, University of Tasmania, Private
Bag 75, Hobart 7001, Australia.
‡ University of St. Andrews.
(1) Dixon, J. T.; Green, M. J.; Hess, F. M.; Morgan, D. H. J. Organomet.
Chem. 2004, 689, 3641.
(2) McGuinness, D. S.; Wasserscheid, P.; Keim, W.; Morgan, D.; Dixon,
J. T.; Bollmann, A.; Maumela, H.; Hess, F.; Englert, U. J. Am. Chem. Soc.
2003, 125, 5272.
(3) Carter, A.; Cohen, S. A.; Cooley, N. A.; Murphy Scutt, A. J.; Wass,
D. F. Chem. Commun. 2002, 858.
(4) Blann, K.; Bollmann, A.; Dixon, J. T.; Hess, F. M.; Killian, E.;
Maumela, H.; Morgan, D. H.; Neveling, A.; Otto, S.; Overett, M. Chem.
Commun. 2005, 620. Blann, K.; Bollmann, A.; Dixon, J. T.; Neveling, A.;
Morgan, D. H.; Maumela, H.; Killian, E.; Hess, F. M.; Otto, S.; Pepler, L.;
Mahomed, H. A.; Overett, M. J.; Green, M. J. (Sasol Technology) WO
04056478A1, 2004.
(5) Bollmann, A.; Blann, K.; Dixon, J. T.; Hess, F. M.; Killian, E.;
Maumela, H.; McGuinness, D. S.; Morgan, D. H.; Neveling, A.; Otto, S.;
Overett, M.; Slawin, A. M. Z.; Wasserscheid, P.; Kuhlmann, S. J. Am. Chem.
Soc. 2004, 126, 14712.
In the search for alternative anions that might be more
resistant to degradation of this kind, we studied the alternative
activator system trityl tetrakis[perfluoro-tert-butoxy]aluminate
(6) Deckers, P. J. W.; Hessen, B.; Teuben, J. H. Angew. Chem., Int. Ed.
2001, 40, 2516.
(7) Andes, C.; Harkins, S. B.; Murtuza, S.; Oyler, K.; Sen, A. J. Am.
Chem. Soc. 2001, 123, 7423.
(8) Overett, M. J.; Blann, K.; Bollmann, A.; Dixon, J. T.; Haasbroek,
D.; Killian, E.; Maumela, H.; McGuinness, D. S.; Morgan, D. H. J. Am.
Chem. Soc. 2005, 127, 10723.
(9) Agapie, T.; Schofer, S. J.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem.
Soc. 2004, 126, 1304.
(10) McGuinness, D. S.; Overett, M.; Tooze, R. P.; Blann, K.; Dixon, J.
T.; Slawin, A. M. Z. Organometallics 2007, 26, 1108.
10.1021/om0701975 CCC: $37.00 © 2007 American Chemical Society
Publication on Web 04/05/2007

Ethylene Tetramerization with CrI Precursors
Organometallics, Vol. 26, No. 10, 2007 2783
Scheme 2. Synthesis of [Cr(CO)4(PNPiPr)][1]
([Ph3C][Al(OC(CF3)3)4], [Ph3C][1]) in combination with AlEt3.
The results of these studies have been reported recently and
include an NMR study illustrating the stability of anion [1] in
the presence of excess trialkylaluminum.11 It is believed that
the presence of Et3Al is required for alkylation of the CrIII
precursor, whereupon alkyl abstraction and reductive elimination
can occur to generate a cationic CrI active species. While such
catalyst generation from an inexpensive CrIII precatalyst is most
convenient from a process perspective, it affords little informa-
tion regarding the oxidation state of the metal during catalysis.
An alternative approach would be to start from a CrI species,
as such a catalyst, if active, would provide valuable insight into
the likely oxidation states of the active species. Herein, we report
the synthesis and isolation of cationic CrI aluminates and their
subsequent testing under tetramerization conditions. We show
for the first time that CrI complexes can lead to effective
catalysts for this transformation and provide further evidence
for a CrI f CrIII catalytic cycle.12,13
shows a singlet peak at -76.0 ppm with a half-height width
(∆1/2) of 11.5 Hz. It has been shown by Ko¨hn et al. that the
broadening of the 19F NMR signal of a fluorinated anion
indicates an interaction with the paramagnetic cation,18 ion-
pairing being deemed present when ∆1/2 is in the region 100-
300 Hz or greater. Hence, it can clearly be seen that 1 is weakly
coordinating in [Cr(CO)4(PNPiPr)][1] (cf. ∆1/2 of Li[1], 3 Hz).
Crystals of [Cr(CO)4(PNPiPr)][1] suitable for X-ray analysis
were grown from toluene/petroleum ether, 40:60 (see Figure
1). Due to a high degree of rotational isomerism of the CF3
groups, the refinement of the anion reveals some significant
disorder. However, the refined structure obtained for the Cr
cation is of good quality and shows a similar coordination
geometry to the neutral Cr0 complex. Notably four separate
structure determinations were obtained from different crystals
of this material, and all displayed similar disorder within the
anion fragment. The shortest cation-anion distances are 2.792
Å intramolecularly and 2.839 Å intermolecularly, corresponding
to interactions between a fluorine and a carbonyl oxygen (F13-
O34 and F33-O32, respectively).19 The disparity between the
two Cr-P bonds lengths of 0.4 Å is of interest, notably that
the Cr-P(2) bond, which is the longer of the two, resides trans
to the carbonyl group, showing the intramolecular interaction
with the anion.
Results and Discussion
Synthesis of Cationic CrI Aluminate Complexes. The
majority of low-valent Cr complexes in the literature have
carbonyls present as a stabilizing ligand. Using a published
method Cr(CO)4PNPiPr was prepared in near quantitative yield.14
Crystals of Cr(CO)4PNPiPr suitable for X-ray analysis were
grown by slow cooling of a hot DCM/MeOH solution.12,15 To
perform the one-electron oxidation of Cr0 to CrI, it has been
shown by Connelly and co-workers that the reaction of AgBF4
with Cr(CO)4(dppm) (dppm ) bis(diphenylphosphino)methane)
at room temperature in DCM gives [Cr(CO)4(dppm)][BF4].16
The synthesis of the required Ag(CH2Cl2)[1] has been reported
by Krossing17 and was prepared as described from the lithium
salt. Using a method analogous to that of Connelly, [Cr(CO)4-
(PNPiPr)][1] was prepared (Scheme 2) in near quantitative yield
as a deep blue solid with a magnetic moment of µeff ) 1.83 µB,
which confirms that the complex is low-spin d5, as expected
for a CrI carbonyl. The IR spectrum shows three carbonyl
stretches at 2086, 2032, and 1963 cm-1, which is a shift toward
higher wavenumbers compared to the corresponding neutral Cr0
complex (2005, 1920, and 1888 cm-1), as expected upon moving
from a neutral to a charged species. The 19F NMR (CD2Cl2)
In order to probe the effect of the N atom within the ligand
backbone upon catalysis, the oxygen and carbon analogues were
examined. The preparation of [Cr(CO)4(POP)][1] (POP )
Ph2P-O-PPh2) proceeded smoothly from the known precursor
[Cr(CO)4(POP)]. Ph2P-O-PPh2 is not isolable as a discrete
compound, existing in the isomeric form Ph2P-P(dO)Ph2;
however, it has been shown by Wong et al.20 that when heating
the diphosphine monoxide with Cr(CO)6, rearrangement takes
place upon coordination to give the desired [Cr(CO)4(POP)].
In contrast, the action of Ag(CH2Cl2)[1] upon [Cr(CO)4-
(dppm)] (dppm ) bis(diphenylphosphino)methane) did not yield
the desired [Cr(CO)4(dppm)][1], but gave a new species isolated
as a pink-purple-colored solid. This material, which eluded
positive identification, exhibited two broad signals in the 31P
NMR spectrum [δ 15.4 and 11.1] and a range of unassignable
1
signals in the H NMR spectrum. This unexpected reactivity
could be the result of the acidic methylene protons of dppm,21
or alternatively that nucleophilic attack upon phosphorus leading
to cleavage of the P-C backbone bond is not without
precedent.22 Given the failure of the dppm complex to oxidize
cleanly, the dippm (dippm ) bis(diisopropylphosphino)methane)
analogue was examined in the belief that the methylene protons
(11) (a) McGuinness, D. S.; Rucklidge, A. J. (Sasol Technology UK)
IPN PCT/IB2006/0053494, 2006. (b) McGuinness, D. S.; Rucklidge, A.
J.; Tooze, R. P.; Slawin, A. M. Z. Organometallics 2007, in press.
(12) Bowen, L. E.; Haddow, M. F.; Orpen, A. G.; Wass, D. F. J. Chem.
Soc., Dalton Trans. 2007, 1160.
(13) During the submission process of this article an excellent report
appeared from Wass, who has followed a similar pathway to ours (see ref
12). However, the crucial difference is in the selection of the anion. The
choice of [B(C6F5)4]- by Wass gave only poor catalysis for the reasons
discussed herein.
(14) Grim, O. G.; Briggs, W. L.; Barth, R. C.; Tolman, C. A.; Jesson, J.
P. Inorg. Chem. 1974, 13, 1095.
(15) This structure was reported in a publication (ref 12) that appeared
during the preparation of this article. The structure determined here is
included in the Supporting Information.
(16) Ashford, P. K.; Baker, P.K.; Connelly, N. G.; Kelly, R. L.; Woodley,
V. A. J. Chem. Soc., Dalton Trans. 1982, 477.
(18) Ko¨hn, R. D.; Smith, D.; Mahon, M. F.; Prinz, M.; Mihan, S.; Kociok-
Ko¨hn, G. J. Organomet. Chem. 2003, 683, 200.
(19) It should be noted that the fluorines involved in these closest contacts
are two of the less disordered within the anion, giving a reasonable degree
of certainty to these values.
(20) Wong, E. H.; Ravenelle, R. M.; Gabe, E. J.; Lee, R. M.; Prasad, L.
J. Organomet. Chem. 1982, 233, 321.
(21) Overett, M. J.; Blann, K.; Bollmann, A.; de Villiers, R.; Dixon, J.;
Killian, E.; Maumela, C.; Maumela, H.; McGuinness, D.; Morgan, D. H.;
Rucklidge, A.; Slawin, A. M. Z. AdV. Synth. Catal. 2006, in press.
(22) Ruiz, J.; Garc´ıa-Granda, S.; Rosario D´ıaz, M.; Quesada, R. J. Chem.
Soc., Dalton Trans. 2006, 4371.
(17) Krossing, I. Chem.-Eur. J. 2001, 7, 490.

2784 Organometallics, Vol. 26, No. 10, 2007
Rucklidge et al.
progressed, the three carbonyl bands of [Cr(CO)4(PNPiPr)][1]
diminished with concomitant formation of three new carbonyl
stretching frequencies at 2005, 1920, and 1888 cm-1. Surpris-
ingly, this suggests formation of the Cr0 species, [Cr(CO)4-
(PNPiPr)], and from initial inspection of the IR data (see Figure
2), it also appears as though this conversion is quantitative.
However, through determination of absorption coefficients from
standard solutions of both CrI and Cr0 species, it is found that
only 33% of the initial CrI inventory can be accounted for by
formation of [Cr(CO)4(PNPiPr)]. This suggests that the remaining
Cr metal no longer possessed coordinated CO ligands.
Figure 1. Molecular structure of [Cr(CO)4(PNPiPr)][1]. Selected
bond distances (Å) and angles (deg): Cr(1)-P(1) 2.263(7); Cr-
(1)-P(2) 2.660(8); Cr(1)-C(31) 1.90(3); Cr(1)-C(32) 2.01(3); Cr-
(1)-C(33) 1.87(4); Cr(1)-C(34) 1.87(3); P(1)-Cr(1)-P(2) 62.2(2);
P(1)-N(1)-P(2) 95.8(9).
Ethylene Tetramerization. When a CrIII precursor is used
for the tetramerization of ethylene, both an alkylating and alkyl
abstracting agent is required. This can be either a single
substance such as MAO or two different compounds each
playing a role, for example Et3Al (alkylating) and the trityl
cation (alkyl abstracting). However, in the CrI precursor case
neither of these criteria are required, provided a route to remove
the carbonyl ligands is utilized instead. Tetramerization was
attempted by either photolyzing the solution or reacting it with
Et3Al to remove the carbonyl ligands, the results being shown
in Table 1As expected, in the absence of a carbonyl-abstracting
agent, [Cr(CO)4(PNPiPr)][1] is inactive for ethylene tetramer-
ization (entry 1). To determine whether an active catalyst could
be formed with only activation of the complex by UV photolysis,
[Cr(CO)4(PNPiPr][1] (20 µmol) was dissolved in toluene (10
cm3) and photolyzed under ethylene (8 bar). The ethylene was
used to stabilize any low-coordinate species that are formed in
the photolysis process. After 10 min the solution was transferred
under positive pressure of ethylene to an autoclave containing
methylcyclohexane (MCH) at 60 °C. The reactor was pressur-
ized with ethylene (40 bar) and the reactor stirred vigorously
for 1 h. Upon cooling and venting of the reactor, the liquid
fraction was analyzed and no oligomers were found to be present
(entry 2). Hence, this procedure was repeated in the presence
of 400 equiv of AlEt3 (entry 3). A 25 min induction period was
observed to occur in the reactor (as indicated by gas uptake
data), after which an active catalyst was successfully generated,
and the autoclave was stirred for a further 50 min. It should be
noted in this regard that the role of AlEt3 may also be to act as
an additional poison scavenger (aside from CO), and hence an
amount may be required to protect the active species, as is often
the case in ethylene polymerization chemistry. The failure of
irradiation alone to activate the catalyst may also be due to a
lack of scavenger to permanently remove the carbonyl ligands
from Cr.
may be less acidic and that the phosphorus center would be
more electron rich and hence less prone to nucleophilic attack.21
The synthesis of the precursor complex [Cr(CO)4(dippm)] was
facile; however, the subsequent oxidation to the desired [Cr-
(CO)4(dippm)][1] did not proceed, despite repeated attempts,
instead appearing to decompose. Hence, in order to access a
ligand backbone with a carbon bridge, the two-carbon dppe
analogue [Cr(CO)4(dppe)][1] (dppe ) bis(diphenylphosphino)-
ethane) was prepared from the known [Cr(CO)4(dppe)], the
oxidation proceeding smoothly on this occasion.
To determine the effect of the anion upon catalysis, other
CrI-PNP complexes with different weakly coordinating anions
were prepared, namely, [Cr(CO)4(PNPiPr)][BF4] and [Cr(CO)4-
(PNPiPr)][PF6] from [Cr(CO)4(PNPiPr)] and the corresponding
silver salt. The measured ∆1/2 of the 19F peaks for the PF6 and
the BF4 complex were found to be 140 and 625 Hz, respectively,
and are clearly indicative of stronger interactions between the
anion and cation than was found in [Cr(CO)4(PNPiPr)][1].
Carbonyl Ligand Removal. In order that the CrI cations may
become active catalysts, it was envisaged that the carbonyl
ligands must be removed from the coordination sphere of Cr,
such that free coordination sites exist for ethylene to coordinate
and in turn undergo oxidative coupling to form the theorized
metallocyclopentane. Three methods were attempted to remove
the carbonyl ligands; (i) by UV photolysis,23 (ii) via reaction
with Et3Al, and (iii) via reaction with Me3NO,24 all of which
are known procedures for carbonyl abstraction. Surprisingly, it
was found that refluxing a mixture of Me3NO and [Cr(CO)4-
(PNPiPr)][1] in acetonitrile under nitrogen gave no reaction, as
indicated by the unchanging IR spectrum. In contrast, a DCM
solution of [Cr(CO)4(PNPiPr)][1] following addition of 100 equiv
of AlEt3 (1.0 M in hexanes) showed a complete loss of signal,
within 1 min, in the region of the IR spectrum associated with
metal-bound carbonyl groups. A concomitant color change from
royal blue to brown was observed, with all of the material
remaining in solution for at least 15 min, indicating the loss of
carbonyl signals, was not an artifact of carbonyl species
precipitating from solution.
In contrast, when 50-400 equiv of AlEt3 was employed
without the use of photolysis, [Cr(CO)4(PNPiPr][1] was found
to be immediately active for the trimerization and tetramerization
of ethylene (entries 4-7). The C6 and C8 selectivities obtained
closely match those obtained with a CrIII/MAO catalyst system
under the same conditions, along with the formation of small
amounts of methylcyclopentane and methylenecyclopentane.8
From this, we conclude that similar active species are present
in both cases. The productivity obtained is dependent upon the
amount of AlEt3 employed relative to Cr with an optimum
activity around 200 equiv (entry 6). A very low activity was
observed with only 50 equiv of AlEt3, and so no further
exploration to lower amounts was undertaken.
A DCM solution of [Cr(CO)4(PNPiPr)][1] was placed in an
IR solution cell containing a UV transparent window perpen-
dicular to the IR windows for simultaneous photolysis and data
acquisition of the sample. The solution was subjected to UV
radiation with a power of 3 W cm-2 for 30 s periods and a
quiescent period of 90 s between irradiations. A series of IR
spectra were taken during the quiescent period. As time
It is notable that the use of photolysis seems to reduce the
productivity of the catalyst (cf. entries 3 and 7). This may be
related to the above observation (Vide supra) that irradiation
leads to some degradation (∼33% of Cr) of the Cr cation back
to Cr(CO)4(PNPiPr).
(23) Wrighton, M. Chem. ReV. 1974, 74, 401. Welch, J. A.; Peters, K.
S.; Valda, V. J. Phys. Chem. 1982, 86, 1941.
(24) Chen, H.; Johnson, B. F. G.; Lewis, J.; Braga, D.; Grepioni, F.;
Parisini, E. J. Chem. Soc., Dalton Trans. 1991, 215.

Ethylene Tetramerization with CrI Precursors
Organometallics, Vol. 26, No. 10, 2007 2785
Figure 2. IR spectra of [Cr(CO)4(PNP)][1] undergoing UV photolysis, showing the evolution with time, as the initial species disappears
and new signals corresponding to the neutral [Cr(CO)4(PNP)] grow in.
Table 1. Ethylene Tetramerization with [Cr(CO)4(P-P)][1] Complexes Activated with Photolysis of Trialkyl Aluminum
photolysis
(min)
C6 (wt %)
C8 (wt %)
productivity
(g/g Cr)
activity
(g/g Cr/h)
entry
catalyst
AlR3 (equiv)
PE (wt %)
(1 - C6)b
(1 - C8)b
1
2
3
4
5
6
7
8
9
[Cr(CO)4(PNPiPr)][1]
[Cr(CO)4(PNPiPr)][1]
[Cr(CO)4(PNPiPr)][1]
[Cr(CO)4(PNPiPr)][1]
[Cr(CO)4(PNPiPr)][1]
[Cr(CO)4(PNPiPr)][1]
[Cr(CO)4(PNPiPr)][1]
[Cr(CO)4(PNPiPr)][1]
[Cr(CO)4(PNPiPr)][1]
0
10
10
0
0
0
0
0
0
0
0
0
0
0
0
AlEt3 (400)
AlEt3 (50)
AlEt3 (100)
AlEt3 (200)
AlEt3 (400)
iBu3Al (400)
iBu3Al (100)
AlEt3 (300)
AlEt3 (400)
AlEt3 (400)
AlEt3 (100)
AlEt3 (400)
AlEt3 (400)
AlEt3 (400)
AlEt3 (100)
1.2
1.8
0.2
1.1
0.23
2.4
0.8
21.4 (76.9)
24.8 (83.8)
20.4 (77.1)
21.4 (78.8)
21.4 (78.1)
21.2 (79.0)
21.7 (76.5)
71.6 (98.7)
67.2 (97.1)
72.8 (98.8)
70.7 (99.0)
72.4 (99.0)
70.2 (100)
72.7 (99.0)
85 000
10 600
83 900
139 800
126 300
28 400
67 200
85 000
10 600
83 900
139 800
126 300
28 400
67 200
10
11
12
13
14
15
16
[Cr(CO)4(PNPiPr)][PF6]
[Cr(CO)4(PNPiPr)][BF4]
0
0
0
0
0
0
0
0
0
0
0
1600
0
Cr(CO)4(PNPiPr
)
59.4
0
0
10.4
6.3
15.5 (65.1)
20.6 (35.4)
800d
0
[Cr(CO)4(POP)][1]
[Cr(CO)4(POP)][1]c
[Cr(CO)4(dppe)][1]
[Cr(CO)4(dppe)][1]c
0
0
24.2 (51.8)
24.3 (58.9)
46.7 (100)
59.9 (97.6)
6000
19 300
6000
19 300
a Conditions: Cr (5 µmol), 10 mL of toluene, 90 mL of methylcyclohexane, 60 °C, ethylene pressure (40 bar). All runs (except entry 12) 60 min.
c
d
bSelectivity to R-olefin within carbon length fraction. Cr (20 µmol). 30 min run.
A series of patents from Tosoh Corporation describe the use
of triisobutyl aluminum to activate low-valent Cr precursors
toward trimerization in conjunction with UV irradiation to
remove carbonyl ligands.25 Utilization of this alternative alu-
minum activator in our system (without UV photolysis) did
indeed generate an active catalyst (entries 8 and 9); however, it
was inferior to AlEt3, so no further examination in the presence
of UV photolysis was made.
being observed in either case. This is rationalized as an effect
of the relatively strongly coordinating nature of these anions
(cf. [1]) to the cation (Vide supra), which is believed to have a
negative effect upon the catalysis.
As a CrI precursor is theorized as the initiating species for
tetramerization, it was of interest to screen the neutral parent
carbonyl complex Cr0(CO)4(PNPiPr) for catalytic activity (entry
12). This precatalyst source did display some limited activity
toward tetramerization; however, the catalyst was both poorly
active and primarily a polymerization system.
In order to probe the effect of the anion upon catalysis, the
complexes with the PF6 and BF4 anions in place of the
aluminate, [1], were screened (entries 10 and 11), no activity
In order to better understand the implications of the N atom
in the ligand backbone toward catalysis, the analogous com-
pounds [Cr(CO)4(POP)][1] and [Cr(CO)4(dppe)][1] were tested
(entries 13-16). However the POP analogue did not generate
oligomers of any kind, indicating that this ligand does not
support an active tetramerization catalyst. Meanwhile, the dppe
analogue was moderately active and selective toward tetramer-
ization, notably with similar liquid fraction selectivity to the
reported CrIII/dppe/MAO system,5 suggesting that a similar Cr-
active species is present in both cases. This suggests that the
nature of the bridging atom is crucial to the catalysis and that
(25) (a) Oguri, M.; Mimura, H.; Aoyama, T.; Okada, H.; Koie, Y. (Tosoh
Corporation) JP11092407, 1999. (b) Oguri, M.; Mimura, H.; Aoyama,
T.; Okada, H.; Koie, Y. (Tosoh Corporation) JP11092408, 1999. (c)
Oguri, M.; Mimura, H.; Okada, H. (Tosoh Corporation) JP2000176291,
2000. (d) Oguri, M.; Mimura, H.; Okada, H.; Yamamoto, T.; Murakita,
H. (Tosoh Corporation) US6337297, 2002. (e) Oguri, M.; Mimura,
H.; Okada, H. (Tosoh Corporation) JP2000202299, 2000. (f) Oguri, M.;
Mimura, H.; Okada, H.; Yamamoto, T.; Osamu, U. (Tosoh Corporation)
JP2001002724, 2001. (g) Oguri, M.; Mimura, H.; Okada, H.; Yamamoto,
T.; Osamu, U. (Tosoh Corporation) JP2001002725, 2001. (h) Okada, H.;
Yamamoto, T.; Osamu, U.; Murakita, S. (Tosoh Corporation) JP2001149788,
2001.

2786 Organometallics, Vol. 26, No. 10, 2007
Rucklidge et al.
with a concomitant color change to dark blue. The mixture was
allowed to stir for 1 h, after which it was filtered and the solvent
was removed in Vacuo. The resultant blue solid was washed with
petrol 40:60 (2 × 20 cm3) and dried in Vacuo (0.54 g, 77%). Anal.
Calcd for C31H27BCrF4NO4P2 (found): C, 54.89 (54.98); H, 4.01
(4.08); N, 2.06 (2.07). µeff ) 1.63 µB. 19F NMR (CD2Cl2, 295 K):
simple modification of the bridge can lead to dramatic changes
in catalytic performance.
Summary and Conclusions
A variety of Cr0 and CrI carbonyl complexes of PNP and
related ligands have been prepared and fully characterized. The
evaluation of these complexes for the ethylene tetramerization
reaction reveals that a CrI cation along with a weakly coordinat-
ing anion is required to produce an active system and that a
trialkylaluminum reagent is necessary to abstract the carbonyl
ligands from Cr and possibly to act as a poison scavenger. Aside
from the postulated oxidative coupling believed to occur in the
catalytic cycle, it is unlikely that under the conditions employed
oxidation of the Cr to a higher oxidation state active species
would take place. This conclusion, together with the near lack
of activity displayed by a Cr0 complex, lends further support
for a cycle involving formally CrI and CrIII active species. We
have also demonstrated that MAO is not an indispensable
cocatalyst in the tetramerization of ethylene, and this work opens
the way for the further development of catalyst systems based
on less expensive cocatalysts. Further work in this regard is
ongoing and will be reported in due course.
δ -149.7 (br s, ν1/2 ) 625 Hz). IR (CH2Cl2): νCO ) 2085 cm-1
,
νCO ) 2032 cm-1, νCO ) 1960 cm-1
.
[Cr(CO)4(PNPiPr)][PF6]. A Schlenk flask was charged with Cr-
(CO)4(PNPiPr) (0.59 g, 1.00 mmol), and dichloromethane (15 cm3)
was added. AgPF6 (0.25 g, 1.00 mmol) was added to the solution,
with a concomitant color change to dark blue. The mixture was
allowed to stir for 1 h, after which it was filtered and the solvent
was removed in Vacuo. The resultant blue solid was washed with
petrol 40:60 (2 × 20 cm3) and dried in Vacuo (0.62 g, 84%). Anal.
Calcd for C31H27CrF6NO4P3 (found): C, 50.56 (50.65); H, 3.70
(3.71); N, 1.90 (1.89). µeff ) 1.78 µB. 19F NMR (CD2Cl2, 295 K):
δ -69.6 (d, 1JP-F ) 709 Hz, ν1/2 ) 140 Hz). IR (CH2Cl2): νCO
)
2085 cm-1, νCO ) 2032 cm-1, νCO ) 1962 cm-1
.
[Cr(CO)4(POP)][1]. A Schlenk flask was charged with Cr(CO)4-
(POP) (0.095 g, 0.17 mmol), and dichloromethane (10 cm3) was
added. Ag[1] (0.200 g, 0.17 mmol) was added to the solution, an
immediate darkening in color being observed. The mixture was
allowed to stir overnight at RT, and a second portion of Ag[1] (0.17
mmol) was added, followed by stirring for a further 18 h. The
solution was filtered, the solid was washed with petrol 40:60 (1 ×
20 cm3), and the combined fractions were reduced in Vacuo to leave
a green powder (0.191 g, 74%). Anal. Calcd for C44H20AlCrF36O9P2
(found): C, 34.83 (34.60); H, 1.33 (1.43). 19F NMR (CD2Cl2, 295
K): δ -76.1 (br s, ν1/2 ) 12.7 Hz).
Experimental Section
General Comments. All manipulations were carried out using
standard Schlenk techniques or in a nitrogen glovebox, using
solvents purified and dried via standard procedures. Ag[Al{OC-
(CF3)3}4],17 Ph2PP(dO)Ph2,26 Cr(CO)4(POP),20 Cr(CO)4(dppm),27
Cr(CO)4(dippm),27 Cr(CO)4(dppe),27 and Cr(CO)4(PNPiPr)28 were
prepared using literature procedures or slight modifications
thereof.
[Cr(CO)4(dppe)][1]. A Schlenk flask was charged with Cr(CO)4-
(dppe) (0.075 g, 0.13 mmol), and dichloromethane (10 cm3) was
added. Ag[1] (0.143 g, 0.13 mmol) was added to the solution, an
immediate darkening in color being observed. The mixture was
allowed to stir overnight at RT, and a second portion of Ag[1] (0.13
mmol) was added, followed by stirring for a further 18 h. The
solution was filtered, the solid was washed with petrol 40:60 (1 ×
20 cm3), and the combined fractions were reduced in Vacuo to leave
a purple powder (0.186 g, 94%). Anal. Calcd for C46H24AlCrF36O8P2
(found): C, 36.12 (36.01); H, 1.58 (1.54). 19F NMR (CD2Cl2, 295
Representative Procedure for Neutral Chromium Tetracar-
bonyl Species: Cr(CO)4(PNPiPr). A Schlenk flask was charged
with Cr(CO)6 (4.00 g, 18.2 mmol) and PNPiPr (7.77 g, 18.2 mmol),
and anhydrous diglyme (50 cm3) was added. The mixture was
heated to 160 °C for 2 h, after which time the resultant solution
was allowed to cool. The solution was concentrated in Vacuo, and
MeOH (30 cm3) was added. The solution was placed in the freezer
and the resultant yellow powder collected by filtration, washed with
MeOH (2 × 20 cm3), and dried in Vacuo (8.94 g, 83%). A portion
was recrystallized from hot DCM/MeOH, which gave yellow blocks
suitable for X-ray analysis. Anal. Calcd for C31H27CrNO4P2
(found): C, 62.95 (62.93); H, 4.60 (4.56); N, 2.37 (2.31).
[Cr(CO)4(PNPiPr)][1]. A Schlenk flask was charged with Cr-
(CO)4(PNPiPr) (0.075 g, 0.13 mmol), and dichloromethane (15 cm3)
was added. Ag[1] (0.140 g, 0.13 mmol) was added to the solution,
and an immediate color change to dark blue was observed. The
mixture was allowed to stir for 1 h, after which a second portion
of Ag[1] (0.13 mmol) was added. After stirring overnight the
solution was filtered and the solvent was removed in Vacuo. The
resultant blue solid was washed with petrol 40:60 (2 × 20 cm3)
and dried in Vacuo (0.150 g, 74%). Anal. Calcd for C47H27AlCrF36-
NO8P2 (found): C, 36.22 (36.19); H, 1.75 (1.73); N, 0.90 (0.87).
K): δ -76.1 (br s, ν1/2 ) 5.6 Hz). IR (CH2Cl2): νCO ) 2084 cm-1
νCO ) 2031 cm-1, νCO ) 1970 cm-1
,
.
Tetramerization. Ethylene tetramerization was carried out in a
300 cm3 stainless steel reactor with mechanical stirring. The oven-
dried vessel was purged with N2 followed by ethylene, charged
with methylcyclohexane (90 cm3), and heated to 60 °C. The catalyst
solution was prepared by dissolving the required amount of catalyst
in toluene (10 cm3) and adding the required amount of trialkyla-
luminum (AlR3). The solution was injected into the prepared
autoclave, and the reactor was immediately charged with 40 bar of
ethylene and maintained at this pressure for the duration of the
reaction. After 1 h the reactor was cooled in an ice bath, the excess
ethylene was bled, and an internal standard was added (nonane,
1000 µL). After quenching with MeOH followed by 10% HCl, the
organic phase was analyzed by GC, and the while solids were
filtered, washed, dried, and weighed.
µeff ) 1.83 µB. 19F NMR (CD2Cl2, 295 K): δ -76.0 (s, ν1/2
11.5 Hz). IR (CH2Cl2): νCO ) 2086 cm-1, νCO ) 2032 cm-1, νCO
)
After the catalyst was photolyzed prior to the reaction, the catalyst
solution (with or without AlR3) was transferred to a glass Buchi
reactor, which was subsequently pressurized with ethylene (8 bar).
The solution was then photolyzed with UV light from a Hg-
discharge lamp with the output being directed with a light guide.
) 1963 cm-1
.
[Cr(CO)4(PNPiPr)][BF4]. A Schlenk flask was charged with Cr-
(CO)4(PNPiPr) (0.61 g, 1.03 mmol), and dichloromethane (15 cm3)
was added. AgBF4 (0.20 g, 1.03 mmol) was added to the solution,
The power of the lamp at the end of the light guide was 3 W cm-2
.
The solution was photolyzed with stirring for 10 min, after which
the ethylene was slowly vented and the solution transferred to the
prepared autoclave by cannula.
IR Spectroscopy. Infrared spectra were recorded on a Nicolet
Nexus instrument fitted with a MCT-High D* detector. Absorption
(26) McKechnie, J.; Payne, D. S.; Sim, W. J. Chem Soc. 1965, 3500.
(27) (a) Grim, S. O.; Briggs, W. L.; Barth, R. C.; Tolman, C. A.; Jesson,
J. P. Inorg. Chem. 1974, 13, 1095. (b) VanAtta, S. L.; Duclos, B. A.; Green,
D. B. Organometallics 2000, 19, 2397-2399.
(28) Balakrishna, M. S.; Prakasha, T. K.; Krishnamurthy, S. S.; Siri-
wardane, U.; Hosmane, N. S. J. Organomet. Chem. 1990, 390, 203.

Ethylene Tetramerization with CrI Precursors
Organometallics, Vol. 26, No. 10, 2007 2787
2
2
coefficients were determined from IR spectra of standard solutions
in dichloromethane using a solution cell with a path length of 1
mm. The absorbance as a function of concentration was determined
at 1888 cm-1 for [Cr(CO)4(PNPiPr)] and 2085 cm-1 for [Cr(CO)4-
(PNPiPr)][1] with a linear baseline defined by the limits 2080 and
1788 cm-1 in each case. The photolysis experiment was carried
out in a variable path length solution cell with a third UV-
transparent window perpendicular to the IR beam for irradiation
of the solution. The solution was photolyzed using an EXFO
Acticure 4000 spot curing system fitted with a liquid light guide.
reflections with Fo > 2σFo , S ) 1.127 for 704 parameters.
Residual electron density extremes were 0.478 and -0.496 e Å-3
.
[Cr(CO)4(PNPiPr)][1]. Very poor crystal quality precluded a
good analysis; several different crystals were examined, using both
Mo and Cu radiation. The best Cu data set was selected and is
reported here. C31H27CrNO4P2, C16F36AlO4, M ) 1558.62, blue
platelet, crystal size 0.1 × 0.1 × 0.01 mm, monoclinic, P2(1)/c, a
) 14.7195(13) Å, b ) 25.217(2) Å, c ) 16.4329(14) Å, â )
103.915(3)°, V ) 5920.6(9) Å3, Z ) 4, Dcalcd ) 1.749 Mg m-3; Cu
KR radiation (λ ) 1.54178 Å), µ ) 3.787 mm-1, T ) 173(2) K;
28 807 data (4661 unique, Rint ) 0.1554), conventional R ) 0.2615
The solution was irradiated with a power of 3 W cm-2
.
2
2
X-ray Crystallography. Data were collected using a Rigaku
MM007 Mo high-brilliance rotating anode/confocal optics with a
Rigaku Mercury detector, for [Cr(CO)4(PNPiPr)], whereas [Cr(CO)4-
(PNPiPr)][1] used a Rigaku Saturn92 detector on a Rigaku MM007
Cu high-brilliance rotating anode/confocal optics, and the data were
corrected for absorption. The structures were solved by direct
methods and refined by full-matrix least-squares on F2 values of
all data (G. M. Sheldrick, SHELXTL, Bruker AXS Madison WI,
2001, version 6.1).
Cr(CO)4(PNPiPr). C31H27CrNO4P2, M ) 591.48, yellow, crystal
size 0.3 × 0.2 × 0.1 mm, triclinic, P1h, a ) 10.799(2) Å, b )
15.442(3) Å, c ) 17.983(4) Å, R ) 90.504(9)°, â ) 106.818(7)°,
γ ) 95.228(7)°, V ) 2856.8(9) Å3, Z ) 4 (two crystallographically
independent molecules), Dcalcd ) 1.375 Mg m-3; Mo KR radiation
(λ ) 0.71073 Å), µ ) 0.735 mm-1, T ) 93(2) K; 15 687 data
(9302 unique, Rint ) 0.0400), conventional R ) 0.0665 for 6658
for 3816 reflections with Fo > 2σFo , S ) 3.020 for 286
parameters. All carbon and fluorine atoms were refined isotropically,
with the CF3 groups having fixed carbon thermal parameters and
the fluorines riding thermal parameters in idealized geometries.
Residual electron density extremes were 2.268 and -1.956 e Å-3
.
Acknowledgment. The authors thank the members of the
Sasol Technology Olefin Transformations group for helpful
discussions and suggestions, and Sasol Technology UK Ltd.
and Sasol Technology (Pty) Ltd. for permission to publish.
Supporting Information Available: Crystallographic data, in
CIF format, for structures reported. This material is available free
of charge via the Internet at http://pubs.acs.org.
OM0701975