Ethylene trimerization with Cr-PNP and Cr-SNS complexes: Effect of ligand structure, metal oxidation state, and role of activator on catalysis

Article abstract of DOI:10.1021/om0601091

Selected PNP and SNS ethylene trimerization ligands have been coordinated to Cr^II and Cr^III, and further reactions of these complexes have been studied. The ligands are easily deprotonated to afford monoanionic tridentate ligands. Al

Full text of DOI:10.1021/om0601091

Organometallics 2006, 25, 3605-3610
3605
Ethylene Trimerization with Cr-PNP and Cr-SNS Complexes:
Effect of Ligand Structure, Metal Oxidation State, and Role of
Activator on Catalysis
David S. McGuinness,*,† David B. Brown,† Robert P. Tooze,† Fiona M. Hess,‡
John T. Dixon,‡ and Alexandra M. Z. Slawin
Sasol Technology (UK) Ltd, Purdie Building, North Haugh, St. Andrews, KY16 9ST, U.K., Sasol
Technology (Pty) Ltd, R&D DiVision, 1 Klasie HaVenga Road, Sasolburg, 1947, South Africa, and School
of Chemistry, UniVersity of St. Andrews, Purdie Building, St. Andrews, KY16 9ST, U.K.
ReceiVed February 3, 2006
Selected PNP and SNS ethylene trimerization ligands have been coordinated to CrII and CrIII, and
further reactions of these complexes have been studied. The ligands are easily deprotonated to afford
monoanionic tridentate ligands. All prepared complexes gave ethylene trimerization catalysis with varying
degrees of activity upon activation with both MAO and AlR3/B(C6F5)3. The results of this study show
that the role of MAO during activation is one of deprotonation, Cr reduction, and cation generation. A
CrII f CrIV cationic mechanism is suggested.
ethylene trimerization when activated with MAO.10,11 The effects
of central N-donor substitution, chelate ring size, and also
changes to the tridentate donor set have been explored.12 The
most active systems arise with ligands of the type (RSCH2CH2)2-
NH, in which the S-donor is substituted with an n-alkyl group.
It was also found that a secondary amine donor was essential
for high activity, raising the possibility that the ligand is
deprotonated during catalyst activation. However, the precise
role of the cocatalyst, the mode of ligand binding, and the
oxidation state of the metal during catalysis has hitherto not
been studied.
The mechanism of ethylene trimerization is generally sup-
posed to follow a metallacyclic route,1 involving oxidative
addition of two ethylene molecules to the metal followed by
insertion of another to yield a metallacycloheptane species,
although conclusive experimental evidence for this is limited.
Jolly and co-workers13 have isolated a chromacycloheptane
species that liberates 1-hexene upon thermolysis, while Bercaw
and co-workers, and our research group, used deuterium-labeled
ethylene to probe the mechanism.6,14 Additionally, Gibson and
co-workers have recently provided evidence for large ring
1. Introduction
While the trimerization of ethylene to 1-hexene has been
known for many years,1 this reaction has gained increased
attention recently2 due to the importance of 1-hexene in the
production of linear low-density polyethylene (LLDPE). The
trimerization route largely avoids the production of undesired
olefins that conventional (full-range) olefin oligomerization
processes produce. Perhaps the best known of the ethylene
trimerization catalysts is the Chevron Phillips system that is
based on pyrrolyl-Cr complexes, which was commercialized
in Qatar during 2003.3 However, a number of other highly active
and selective systems based on Cr,4-7 and also Ti,8 have recently
been disclosed both in the open literature and in corresponding
patents. Additionally, an important advance is the selective
tetramerization of ethylene to 1-octene, which we have recently
reported for the first time.9 1-Octene is also an important
comonomer used for the production of LLDPE.
We have previously reported on the use of CrIII-PNP and
-SNS complexes as highly selective and active systems for
* To whom correspondence should be addressed. E-mail:
davemcguinness@yahoo.com.
(9) 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. 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. (Sasol Technology) WO 04056479,
2004.
(10) McGuinness, D. S.; Wasserscheid, P.; Keim, W.; Hu, C.; Englert,
U.; Dixon, J. T.; Grove, C. Chem. Commun. 2003, 334. Dixon, J. T.; Grove,
J. J. C.; Wasserscheid, P.; McGuinness, D. S.; Hess, F. M.; Maumela, H.;
Morgan, D. H.; Bollmann, A. (Sasol Technology) WO 03053891A1, 2003.
(11) 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. Dixon, J. T.; Wasserscheid, P.; McGuinness, D. S.; Hess,
F. M.; Maumela, H.; Morgan, D. H.; Bollmann, A. (Sasol Technology)
WO 03053890A1, 2003.
† Sasol Technology UK.
‡ Sasol Technology (Pty) Ltd.
University of St. Andrews.
(1) Manyik, R. M.; Walker, W. E.; Wilson, T. P. J. Catal. 1977, 47,
197. Briggs, J. R. Chem. Commun. 1989, 674.
(2) Dixon, J. T.; Green, M. J.; Hess, F. M.; Morgan, D. H. J. Organomet.
Chem. 2004, 689, 3641.
(3) Lashier, M. E. (Phillips Petroleum) EP 0780353A1, 1997.
(4) Wu, F.-J. (Amoco) US 5811618, 1998.
(5) Carter, A.; Cohen, S. A.; Cooley, N. A.; Murphy, A.; Scutt, J.; Wass,
D. F. Chem. Commun. 2002, 858.
(6) Agapie, T.; Schofer, S. J.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem.
Soc. 2004, 126, 1304.
(7) 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.
(12) McGuinness, D. S.; Wasserscheid, P.; Morgan, D. H.; Dixon, J. T.
Organometallics 2005, 24, 552.
(13) Emrich, R.; Heinemann, O.; Jolly, P. W.; Kru¨ger, C.; Verhovnik,
G. P. J. Organometallics 1997, 16, 1511.
(8) Deckers, P. J. W.; Hessen, B.; Teuben, J. H. Angew. Chem., Int. Ed.
2001, 40, 2516. Deckers, P. J. W.; Hessen, B. (Stichting Dutch Polymer
Institute) WO 02066405A1, 2002.
(14) 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.
10.1021/om0601091 CCC: $33.50 © 2006 American Chemical Society
Publication on Web 06/23/2006

3606 Organometallics, Vol. 25, No. 15, 2006
McGuinness et al.
metallacycle intermediates in Schulz-Flory oligomerization.15
These results provide good evidence for a mechanism based
upon metallacycle formation. More extensive mechanistic
studies have been carried out by theoretical methods on Cr,16
Ta,17 and Ti18-20 based systems. These studies also support a
metallacyclic mechanism. With regards to the formal oxidation
state of the metal in Cr-catalyzed trimerization, there is evidence
for both CrI/CrIII 6,21 and CrII/CrIV 16,22,23 mechanisms, and it
seems possible that the oxidation state of Cr in the catalytic
cycle is ligand dependent.
Table 1. Ethylene Trimerization with Cr-PNP and
Cr-SNS Complexes/MAOa
equiv PE C6
MAO (wt %) (wt %)
selec 1-C6
(wt %)b
TOF
c
entry
catalyst
(h-1
)
1
2
3
4
5
6
7
CrCl3(HL1)
CrCl3(HL2)
CrCl3(HL3)
CrCl2(HL1)
CrCl2(HL2)
[CrClL1]2
300
150
300
300
150
300
300
1.0
1.1
1.7
1.9
4.4
98.7
98.5
96.1
72.2
94.9
79.8
45.2
99.7
99.8
99.7
98.5
99.8
98.5
98.8
24 625
27 625
97 880
26 060
20 020
14 465
4785
2.2
51.0
[CrClL2]2
Herein we report further studies on the PNP and SNS
trimerization systems in which the nature of the ligand (neutral
or anionic), effect of Cr oxidation state, and role of the cocatalyst
are investigated. A number of new CrII and CrIII complexes have
been prepared and their reactivity has been studied. Evidence
for a cationic CrII/CrIV mechanism in which the ligand is
monoanonic is presented.
a Conditions: 20 µmol of Cr, 100 mL of toluene, 80 °C, 30 min.
b Selectivity fraction. c TOF
to 1-hexene within C6
mol(ethylene)‚mol(Cr)-1‚h-1
)
.
Scheme 1. Synthesis of CrCl2(HL1) and CrCl2(HL2)
2. Results and Discussion
2.1. Ligands. While PNP and SNS ligands with a low steric
demand give rise to the most active catalysts,10 the aim of the
present study was not to prepare highly active catalysts but rather
to probe the mechanistic questions posed above. As the synthesis
of complexes with lower coordination number was envisaged
(see below), we chose to employ ligands with some degree of
steric bulk, but not so much that catalysis is completely
suppressed. Ligands HL1 and HL2 were thus used in these
studies, although highly active trimerization catalysts based on
HL3 and HL4 have also been examined for comparative
purposes.
2.2. Effect of Chromium Oxidation State. Trimerization
precatalysts of the PNP and SNS ligands studied here have
previously been based exclusively on CrIII, due mainly to the
ready availability of the complex precursor, CrCl3(thf)3. It is
possible that this oxidation state is not maintained during
activation with MAO, reduction to a lower oxidation state being
possible. To study the effect of using precatalysts containing
Cr in a lower oxidation state, the complexes CrCl2(HL1) and
CrCl2(HL2) were prepared in good yield (>70%) from CrCl2-
(thf)2 and the respective ligand (Scheme 1). The magnetic
moments of blue CrCl2(HL1) (µeff ) 5.14 µB) and pale blue-
green CrCl2(HL2) (µeff ) 4.43 µB) indicate both complexes are
high-spin d4. Both complexes are highly air and moisture
sensitive. Additionally, CrCl2(HL1) was found to react exother-
mically with CH2Cl2 to yield CrCl3(HL1), presumably via a one-
electron oxidation with radical generation,24 although this was
not investigated further.
Crystals of CrCl2(HL1) suitable for an X-ray analysis were
grown from hot MeCN, and the structure is shown in Figure 1.
A square-pyramidal structure is displayed, with the ligand
coordinated in a meridonal arrangement. The same mode of
coordination was observed in CrIII SNS and PNP complexes,10,11
and with the exception of an extra chloride ligand in the sixth
coordination site of CrCl3(HL1), the chromium(II) complex
exhibits comparable bond distances and angles (Figure 1). One
interesting feature is the NH hydrogen bonding consisting of
an intermolecular N-H‚‚‚Cl bond (N-H‚‚‚Cl(2′) 2.34(3) Å,
N‚‚‚Cl(2′) 3.17(4) Å, N-H‚‚‚Cl(2′) 141(4)°) and an intra-
molecular N-H to apical chloride H-bond (H‚‚‚Cl 2.71(4) Å,
N‚‚‚Cl 3.220(4) Å, N-H‚‚‚Cl 113(3)°).
The preparation of CrCl3(HL1), and subsequent ethylene
trimerization upon activation with MAO, has been reported
previously.10 CrCl3(HL2) was prepared in the same manner, and
the results of ethylene trimerization reactions using both
complexes are shown in Table 1. Entries 1 and 2 show that
both complexes give comparable (moderate) activities, with
good selectivity toward C6. The introduction of bulky substit-
uents has attenuated the activity of the SNS catalyst system, as
expected. For comparison, a run conducted under comparable
conditions with CrCl3(HL3) is given in Table 1 (entry 3). It
can be seen that less bulky ethyl-S substitution gives rise to a
much higher activity.
(15) Tomov, A. K.; Chirinos, J. J.; Jone, D. J.; Long, R. J.; Gibson, V.
C. J. Am. Chem. Soc. 2005, 127, 10166.
(16) Janse van Rensburg, W.; Grove, C.; Steynberg, J. P.; Stark, K. B.;
Huyser, J. J.; Steynberg, P. J. Organometallics 2004, 23, 1207.
(17) Yu, Z.-X.; Houk, K. N. Angew. Chem., Int. Ed. 2003, 42, 808.
(18) Blok, A. N. J.; Budzelaar, P. H. M.; Gal, A. W. Organometallics
2003, 22, 2564.
(19) de Bruin, T. J. M.; Magna, L.; Raybaud, P.; Toulhoat, H.
Organometallics 2003, 22, 3404.
(20) Tobisch, S.; Ziegler, T. Organometallics 2003, 22, 5392.
(21) Ko¨hn, R. D.; Smith, D.; Mahon, M. F.; Prinz, M.; Mihan, S.; Kociok-
Ko¨hn, G. J. Organomet. Chem. 2003, 683, 200.
Ethylene trimerization results obtained with these CrII com-
plexes are shown in Table 1, entries 4 and 5. The activities
obtained with both are broadly similar to those obtained with
the CrIII congeners, although the selectivities to C6 are somewhat
lower for CrCl2(HL1), this being due to the formation of some
higher oligomers (C8-C20). Despite this difference, the results
show that CrII precatalysts do give rise to the active species
(22) Morgan, D. H.; Schwikkard, S. L.; Dixon, J. T.; Nair, J. J.; Hunter,
R. AdV. Synth. Catal. 2003, 345, 1.
(23) Jabri, A.; Crewdson, P.; Gambarotta, S.; Korobkov, I.; Duchateau,
R. Organometallics 2006, 25, 715.
(24) Fryzuk, M. D.; Leznoff, D. B.; Rettig, S. J.; Young, V. G. J. Chem.
Soc., Dalton Trans. 1999, 147. Sugawara, K.-i.; Hikichi, S.; Akita, M. J.
Chem. Soc., Dalton Trans. 2002, 4514.

Ethylene Trimerization with Cr-PNP and Cr-SNS Complexes
Organometallics, Vol. 25, No. 15, 2006 3607
Figure 1. Molecular structure of CrCl2(HL1). Selected bond
distances (Å) and angles (deg): Cr(1)-P(1) 2.4789(15); Cr(1)-
P(2) 2.4684(14); Cr(1)-N(4) 2.168(4); Cr(1)-Cl(1) 2.3199(13);
Cr(1)-Cl(2) 2.4642(14); P(1)-Cr(1)-N(4) 80.83(10); P(2)-Cr-
(1)-N(4) 81.02(10); P(1)-Cr(1)-P(2) 152.19(5).
Figure 3. Infrared absorbance spectrum (KBr disk) in the ν(NH)
region of CrCl2(HL1) (s) and [CrL1(µ-Cl)]2 (- - -).
L1 and L2. Removal of HCl from CrCl2(HL1) with DABCO
proceeded in ca. 50% yield to afford green-yellow [CrL1(µ-
Cl)]2 (reaction 1). Although a structural determination has not
been carried out, a chloride-bridged dimeric structure is proposed
on the basis of the structure of {Cr[N(SiMe2CH2PPh2)2](µ-Cl)}2
obtained by Fryzuk and the low magnetic moment of 3.79 µB
per Cr, indicative of antiferromagnetic coupling. The complex
can also be prepared in similar yield from the reaction of NaL1
and CrCl2(thf)2. In each case a toluene-insoluble purple product
also formed, presumably a Cr complex formed by a side reaction
(see below). The IR spectra of CrCl2(HL1) and [CrL1(µ-Cl)]2
are compared in Figure 3, and again the disappearance of the
ν(NH) band at 3132 cm-1 supports deprotonation.
Figure 2. Infrared absorbance spectrum (KBr disk) in the ν(NH)
region of CrCl3(HL4) before (s) and after (- - -) treatment with
LiCH2SiMe3.
upon reaction with MAO. Assuming that MAO does not oxidize
the CrII precatalyst, this suggests that, during activation, the CrIII
complexes are reduced to an oxidation state of II or lower.
2.3. Ligand Deprotonation. It has been shown that a
secondary amine donor is essential to activity; substitution of
an alkyl group for the proton results in greatly attenuated activity
and selectivity.12 One explanation for this is that the ligand is
deprotonated during activation, yielding a monoanionic tridenate
species. This has been investigated by reacting CrCl3(HL4) with
LiCH2SiMe3, a more well-defined alkylating agent than MAO.
LiCH2SiMe3 was chosen in order that an alkylated Cr complex
might be isolated. Although we were unable to characterize the
product from this reaction, IR spectroscopy clearly shows the
disappearance of the strong N-H band at 3182 cm-1 (Figure
2). The same result was obtained with deprotonating agents
DBU and DABCO (1,4-diazabicyclo(2.2.2)octane). On the basis
of these results, it seems likely that MAO, or AlMe3 contained
therein, would deprotonate the SNS and PNP ligands during
activation.
Similarly, reaction of LiL2 with CrCl2(thf)2 afforded [Cr-
ClL2]2 as dark blue-purple crystals in 59% yield. The magnetic
moment of 2.82 µB is consistent with either a low-spin d4 metal
center or a dinuclear structure with spin coupling. An X-ray
analysis of the complex reveals an unusual dinuclear structure
in which the amide donor bridges two Cr centers (Figure 4),
with each Cr in a geometry that is midway between square-
pyramidal and trigonal-bipyramidal. The Cr-Cr distance of
2.9381(11) Å is indicative of some degree of interaction between
the metal centers. The Cr-N distances (2.076(3), 2.071(2) Å)
are consistent with Cr-N single bonds. The ligand coordinates
in a facial manner, as opposed to the meridonal coordination
observed for HL1 and HL2. The possibility of this mode of
coordination during catalysis has been speculated upon previ-
ously10 (both facial and meridonal modes of coordination have
been observed for -N(SiMe2CH2PPh2)2).25 Evidently, the steric
bulk at S is insufficient to prevent dinuclear complex formation.
However, it is also interesting to note a large elongation of the
Cr-S distances (2.6854(12), 2.5655(10) Å) in this complex
relative to those in CrCl3(HL3) (2.4508(7), 2.4556(7) Å). It may
be that steric congestion is relieved by elongation of the Cr-S
bonds, thus allowing the dinuclear structure to form. The
The coordination chemistry of the anionic ligand -N(SiMe2-
CH2PPh2)2 with CrII has been extensively studied by Fryzuk
and co-workers in a number of reports.25 These ligands are
closely related to the SNS and PNP trimerization ligands, so
we decided to investigate similar chemistry with deprotonated
(25) Fryzuk, M. D.; Leznoff, D. B.; Rettig, S. J.; Thompson, R. C. Inorg.
Chem. 1994, 33, 5528. Fryzuk, M. D.; Leznoff, D. B.; Rettig, S. J.
Organometallics 1995, 14, 5193. Fryzuk, M. D.; Leznoff, D. B.; Rettig, S.
J. Organometallics 1997, 16, 5116.

3608 Organometallics, Vol. 25, No. 15, 2006
McGuinness et al.
is favored for the SNS and PNP trimerization systems. The use
of perfluoroaryl boranes, along with alkyl metal complexes or
alkylating agents, has been shown to be a highly effective
alternative to the use of MAO, in many instances allowing the
characterization of well-defined cationic active species.26,27 For
this reason, the likelihood of a cationic active species has been
investigated further by studying AlR3/B(C6F5)3 activation of the
Cr-SNS and Cr-PNP systems. The addition of 1 equiv of
B(C6F5)3 to the alkylated (using AlMe3 or AlEt3) Cr complex
was employed. Such a mode of activation is successful with
these systems, as shown in Table 2.
The results with [CrL1(µ-Cl)]2 show that the amount of
trialkylaluminum employed has a large effect on activity. With
only 5 equiv of AlMe3 a poor activity is obtained (entry 1),
while 10 equiv gives rise to a high activity (notably, higher
than when MAO is employed, see Table 1, entry 6). This
suggests that a minimum loading of alkyl aluminum is required
to fully alkylate the precatalyst and possibly to also act as a
poison scavenger. It is notable that this method of activation
can give rise to a very high oVerall selectivity (99.3 wt %
1-hexene in entry 2), the implication being that a single active
site is formed. Entry 3, in which B(C6F5)3 was omitted, shows
that the Lewis acid is an essential component, as without it the
activity is an order of magnitude less. AlEt3 as the alkylating
agent is as effective as AlMe3 (entry 4), although more
polyethylene is formed in this case. Employing more AlEt3
(entry 5) does not lead to a significant increase in activity. This
method of activation with the SNS complex [CrClL2]2 gave a
disappointing selectivity (entry 6). As was found for MAO
activation of this complex, a high proportion of polymer was
formed.
The precatalysts in which the ligands are protonated are also
active after AlEt3/B(C6F5)3 activation. Interestingly, in this case
a higher amount of trialkyl aluminum is required to reach full
activity. For instance, whereas [CrL1(µ-Cl)]2 is fully active with
10 equiv of AlEt3, CrCl2(HL1) requires 20 equiv (cf. entries 7
and 8). This trend seems to continue for the CrIII precursor,
CrCl3(HL1), which requires 30 equiv of AlEt3 to reach full
activity (entries 10 and 11). This seems to make sense, as extra
alkyl aluminum should be required to fully alkylate the di- and
trichlorides, deprotonate the nitrogen, and, in the case of CrIII,
reduce the metal center. However, it was also found that adding
too much AlEt3 leads to deactivation and poor selectivity (entry
12).
Figure 4. Molecular structure of [CrClL2]2. Selected bond distances
(Å) and angles (deg): Cr(1)-S(1) 2.5655(10); Cr(1)-S(7) 2.6854-
(12); Cr(1)-N(4) 2.076(3); N(4)-Cr(1A) 2.071(2); Cr(1)-Cr(1A)
2.9381(11); Cr(1)-Cl(1) 2.3584(11); S(1)-Cr(1)-S(7) 98.30(4);
S(1)-Cr(1)-N(4) 81.55(7); S(7)-Cr(1)-N(4) 80.16(8); N(4)-Cr-
(1)-Cl(1) 176.94(7); S(1)-Cr(1)-N(4A) 151.73(8); S(7)-Cr(1)-
N(4A) 106.70(8); Cr(1)-N(4)-Cr(1A) 90.21(10).
catalytic relevance of such a structure remains uncertain. It is
formed in the absence of other ligating compounds; however
with ligands such as ethylene or sources of alkyl groups present,
it may not form under catalytic conditions.
Both deprotonated CrII complexes are active for ethylene
trimerization after treatment with MAO (Table 1, entries 6 and
7). The activity of [CrL1(µ-Cl)]2 is around half of that of CrCl2-
(HL1), while the activity of [CrClL2]2 is greatly attenuated,
relative to the precursor CrCl2(HL2), and a high proportion of
polymer was formed. The poor performance of [CrClL2]2 may
be due to the bimetallic nature of the precatalyst and bridging
coordination of the SNS ligand. Despite this drop in activity,
this result shows that the anionic SNS and PNP ligands can
give rise to an active species and lends further support to such
a mode of coordination during catalysis, resulting from cocata-
lyst-mediated deprotonation during activation.
The reaction of LiL2 with CrCl3(thf)3 was also carried out;
however in this case we were unable to isolate the desired
complex, CrCl2L2. Instead, purple crystals of [Cr(THF)2Cl2(µ-
Cl2)Li(THF)2] were obtained from THF solution, as identified
by an X-ray structural analysis (see Supporting Information).
While the fate of the ligand was not investigated, evidently the
LiCl formed during the reaction has complexed to Cr.
2.4. Activation with AlR3/B(C6F5)3. Given the above
observations, and the conclusion that the PNP and SNS ligands
are deprotonated and monoanionic in the active species, two
possible mechanisms for trimerization can be envisaged. A cycle
made up of the formal oxidation states CrI f CrIII would involve
neutral species, whereas the couple CrII f CrIV would neces-
sitate formally cationic active species (Scheme 2).
The use of MAO as a cocatalyst in olefin oligomerization
and polymerization chemistry is normally thought to implicate
a cationic active metal center.26 Therefore, a cationic mechanism
Similar results are achieved with HL2 complexes (entries 9,
13, and 14), although in general the activities and selectivities
are lower than those obtained with the HL1 precatalysts. The
method of activation also seems to have an effect on the results
obtained. Whereas the normal method employed was to add
the trialkyl aluminum to the Cr complex, followed by B(C6F5)3
and addition to the reactor, when AlEt3 was first introduced
into the reactor, followed by a Cr/B(C6F5)3 mixture, a reduced
activity and high polymer content resulted (CrCl3(HL2), entry
15).
Finally, AlEt3/B(C6F5)3 activation was also tested along with
the more active system CrCl3(HL3). Entries 16-18 show that,
in contrast to when MAO is used, poor results were obtained
when this complex was employed. Curiously, in this case when
no prealkylation step was carried out (the AlEt3 was added to
the reactor prior to Cr/B(C6F5)3), a much improved activity and
selectivity resulted (entry 19). Even so, the activity of this
catalyst with AlEt3/B(C6F5)3 activation was well below that
obtained with MAO. It was again found for this catalyst that
(26) Chen, E. Y.-X.; Marks, T. J. Chem. ReV. 2000, 100, 1391, and
references therein.
(27) Krossing, I.; Raabe, I. Angew. Chem., Int. Ed. 2004, 43, 2066.

Ethylene Trimerization with Cr-PNP and Cr-SNS Complexes
Organometallics, Vol. 25, No. 15, 2006 3609
Scheme 2. Possible Mechanistic Cycles with Cr-SNS and Cr-PNP Complexes:
Neutral (CrI f CrIII) and Cationic (CrII f CrIV)
Table 2. Ethylene Trimerization with Cr-PNP and
It is interesting to note that while ligands with a low steric
a
bulk, such as L3 and L4, provide the most active systems when
MAO is used to activate the complexes, the use of the most
sterically encumbered ligand tested in this study (L1) seems to
be most suited to AlEt3/B(C6F5)3 activation. It is well known
from olefin polymerization chemistry that cation-anion interac-
tions play a significant role in catalysis, and it seems also for
trimerization that the combination of activating agent and ligand
must be tailored to obtain the best results. The use of AlEt3/
B(C6F5)3 as an activating system is thus of more than just
mechanistic interest. The use of MAO, or other AlMe3-derived
activators, represents a major cost-driver in the potential large-
scale use of such a trimerization technology. The ability to utilize
AlEt3 (which is considerably cheaper) as a component of the
activating system could therefore provide significant economic
benefits, with respect to catalyst cost.
Cr-SNS Complexes/AlR3/B(C6F5)3
AlR3
(equiv)
PE
C6
selec 1-C6
(wt %)b
TOF
(h-1
entry
catalyst
(wt %) (wt %)
)
1
[CrClL1]2
[CrClL1]2
[CrClL1]2
[CrClL1]2
[CrClL1]2
[CrClL2]2
AlMe3 (5)
AlMe3 (10)
AlMe3 (10)
AlEt3 (10)
AlEt3 (20)
AlMe3 (5)
6.3
0.4
1.7
2.8
2.9
85.9
4.6
0.9
4.6
2.5
93.6
99.6
98.3
97.1
96.8
14.1
95.3
98.9
94.9
97.2
96.5
76.2
89.3
90.8
40.0
15.3
58.4
86.1
97.8
93.4
99.6
99.7
99.8
99.7
99.7
ca. 100
99.2
99.6
99.7
99.6
99.5
99.5
99.7
99.4
98.7
95.5
98.8
89.3
99.5
95.5
955
21 610
2105
21 850
24 310
515
2
3c
4
5
6
7
8
CrCl2(HL1) AlEt3 (10)
CrCl2(HL1) AlEt3 (20)
CrCl2(HL2) AlEt3 (20)
CrCl3(HL1) AlEt3 (20)
CrCl3(HL1) AlEt3 (30)
CrCl3(HL1) AlEt3 (40)
CrCl3(HL2) AlEt3 (20)
CrCl3(HL2) AlEt3 (30)
2300
41 810
16 125
21 160
25 810
8270
10 205
10 580
5325
9
10
11
12
13
14
2.8
22.8
10.1
8.2
59.6
84.7
40.7
7.7
15d CrCl3(HL2) AlEt3 (30)
Our structural studies on these ligands and complexes seem
to point to a high degree of flexibility in the way the ligand
binds to the metal center. Both meridonal and facial type
coordination have been observed, and the ability of the SNS
ligand to bind with widely varying Cr-S distances is also noted.
Such behavior has been postulated for other trimerization
ligands. The pendant arene group in Ti trimerization systems
is thought to moderate its coordination strength throughout the
catalytic cycle,18-20 while the pyrrolyl ligand of the Phillips
catalyst does likewise by undergoing haptotropic shifts between
η1 and η5 coordination.16
16
17
18
CrCl3(HL3) AlEt3 (20)
CrCl3(HL3) AlEt3 (30)
CrCl3(HL3) AlEt3 (100)
4630
7535
1160
19d CrCl3(HL3) AlEt3 (30)
20d CrCl3(HL3) AlEt3 (50)
1.3
4.0
29 390
1960
a Conditions: 20 µmol of Cr, 20 µmol of B(C6F5)3, 100 mL of toluene,
80 °C, 30 min. b Selectivity to 1-hexene within C6 fraction. c No B(C6F5)3
added in run 3. d AlEt3 added to reactor followed by Cr/B(C6F5)3 solution,
in runs 15, 19, and 20.
too much AlEt3 leads to catalyst deactivation (entry 20). This
effect may be related to the know propensity of perfluorophenyl
borate anions and AlR3 to undergo alkyl/aryl exchange, leading
to catalyst deactivation.28
4. Experimental Section
General Comments. All manipulations were carried out using
standard Schlenk techniques or in a nitrogen glovebox, using
solvents purified and dried by standard procedures. The synthesis
of CrIII PNP and SNS complexes has been reported previously.10,11
X-ray Crystallography. Data were collected on a Bruker
SMART CCD diffractometer and 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).
CrCl2(HL1). C28H29NP2Cl2Cr, M ) 564.36, blue plate, crystal
size 0.1 × 0.1 × 0.01 mm, orthorhombic, Pbca, a ) 13.181(4) Å,
b ) 17.030(5) Å, c ) 24.820(8) Å, V ) 5571(3) Å3, Z ) 8, Dcalcd
) 1.346 Mg m-3; Mo KR radiation (λ ) 0.71073 Å), µ ) 0.735
mm-1, T ) 125(2) K; 36 107 data (5046 unique, Rint ) 0.1378).
3. Conclusion
The coordination chemistry of selected PNP and SNS ligands
with CrIII and CrII has been studied, and the resulting complexes
were evaluated for ethylene trimerization. The performance of
monometallic CrII precatalysts is comparable with their CrIII
counterparts when MAO activation is used and somewhat better
when AlR3/B(C6F5)3 is employed. These ligands are easily
deprotonated, leading to monoanionic PNP and SNS complexes,
which are also active trimerization precatalysts. The complexes
can be activated with low amounts of AlR3 followed by
treatment with the alkyl-abstracting agent B(C6F5)3, suggesting
that a formally cationic active species is responsible for
trimerization with these ligands. Although further mechanistic
studies are necessary, these facts lead us to suggest a CrII-
CrIV mechanistic cycle.29 The role of MAO with CrCl3(HL)
catalysts is thus to deprotonate the ligand, reduce the metal
center, and facilitate cation generation.
2 2
Rw ) {∑[w(Fo2 - Fc2)2]/ ∑[w(Fo ) ]}1/2 ) 0.1185, conventional R
(29) We have conducted preliminary magnetic susceptibility studies using
the Evans solution NMR method that further support a Cr(II) species upon
activation and demonstrate that a similar minimum amount of alkylating
agent is required to effect this change. When CrCl3(HL3) (µeff ) 4.05 µB)
is treated with 10 equiv of MAO, the magnetic moment remains unchanged
(µeff ) 4.08 µB), while treatment with 25 equiv of MAO leads to a magnetic
moment of 2.95 µB, consistent with a low-spin CrII center (or CrIV).
(28) Bochmann, M.; Sarsfield, M. J. Organometallics 1998, 17, 5908.

3610 Organometallics, Vol. 25, No. 15, 2006
McGuinness et al.
2
2
) 0.0600 for F values of reflections with Fo > 2σFo ) [3244
observed reflections], S ) 1.031 for 312 parameters. Residual
which were suitable for X-ray analysis. Yield: 2.05 g (97%). µeff
) 5.14 µB.
electron density extremes were 0.775 and -0.458 e Å-3
.
CrCl2(HL2). A solution of (tBuSCH2CH2)2NH (0.995 g, 3.99
mmol) in 20 mL of THF was added to a suspension of CrCl2(thf)2
(1.00 g, 3.73 mmol) in 20 mL of THF and stirred overnight. The
solvent was removed under vacuum and the residue taken up in 60
mL of hot MeCN and filtered. Cooling to -18 °C gave light blue
microneedles. After drying under vacuum a pale blue-green powder
remained. Yield: 0.979 g (70%). Anal. Calcd for C12H27S2NCrCl2
[CrClL2]2. C24H52N2S4Cl2Cr2, M ) 671.82, blue prism, crystal
size 0.13 × 0.1 × 0.1 mm, triclinic, P1h, a ) 8.291(2) Å, b ) 9.907-
(3) Å, c ) 11.420(3) Å, R ) 88.901(5)°, â ) 70.838(5)°, γ )
68.979(5) ° V ) 822.0(4) Å3, Z ) 1, Dcalcd ) 1.357 Mg m-3; Mo
KR radiation (λ ) 0.71073 Å), µ ) 1.094 mm-1, T ) 125(2) K,
5276 data (2925 unique, Rint ) 0.025). Rw ) {∑[w(Fo - Fc2)2]/
2
2 2
∑[w(Fo ) ]}1/2 ) 0.0670, conventional R ) 0.0395 for F values of
(found): C, 38.70 (38.56); H, 7.31 (7.39); N, 3.76 (3.74). µeff
4.43 µB.
)
reflections with Fo2 > 2σFo ) [2108 observed reflections], S ) 0.936
2
[CrL1(µ-Cl)]2. A solution of CrCl2(HL1) (0.288 g, 0.510 mmol),
prepared as above, in 15 mL of THF was treated with a THF
solution of 1,4-diazabicyclo(2.2.2)octane (DABCO, 0.0294 g, 0.262
mmol). After stirring overnight the solution was filtered and the
solvent removed under vacuum. The residue was taken up in
toluene/THF (10 mL/10 mL) and filtered again. Removal of the
solvent afforded a green-yellow solid. Yield: 0.146 g (54%). Anal.
Calcd for C28H28P2NCrCl (found): C, 63.70 (63.89); H, 5.34 (5.41);
N, 2.65 (2.73). µeff ) 3.79 µB/Cr (5.36 µB per dimer).
for 155 parameters. Residual electron density extremes were 0.395
and -0.378 e Å-3
.
Cr(THF)2Cl2(µ-Cl)2Li(THF)2. C16H32LiO4Cl4Cr, M ) 489.16,
purple prism, crystal size 0.05 × 0.1 × 0.1 mm, monoclinic, C2/c,
a ) 36.954(4) Å, b ) 10.2303(12) Å, c ) 19.885(2) Å, â )
117.807(2)°, V ) 6649.3(13) Å3, Z ) 12, Dcalcd ) 1.466 Mg m-3
;
Mo KR radiation (λ ) 0.71073 Å), µ ) 1.016 mm-1, T ) 125(2)
2
K, 18 443 data (6041 unique, Rint ) 0.047). Rw ) {∑[w(Fo
-
Fc2)2]/∑[w(Fo ) ]}1/2 ) 0.1089, conventional R ) 0.0441 for F
values of reflections with Fo2 > 2σFo2 [3926 observed reflections],
S ) 0.936 for 353 parameters. Residual electron density extremes
2 2
[CrL2Cl]2. A solution of (tBuSCH2CH2)2NH (0.485 g, 1.94
mmol) in 10 mL of THF was treated with 1.3 mL of 1.6 M BuLi
(2.1 mmol) and stirred for 1 h. The solution was then added to a
suspension of CrCl2(thf)2 (0.495 g, 1.85 mmol) in 20 mL of THF.
After stirring for 2 h the solvent was removed under vacuum and
the residue taken up in 30 mL of toluene and filtered. The toluene
was removed and the product washed with ether (3 × 5 mL) to
give a dark purple powder. Yield: 0.366 g (59%). Anal. Calcd for
C12H26S2NCrCl (found): C, 42.91 (42.85); H, 7.80 (7.67); N, 4.17
(4.00). µeff ) 2.82 µB/Cr (3.98 µB per dimer).
were 0.371and -0.429 e Å-3
.
Trimerization. Ethylene trimerization was carried out in a 300
mL stainless steel reactor with mechanical stirring. The oven-dried
vessel was purged with N2 followed by ethylene and charged with
80 mL of toluene. After heating to 80 °C, MAO (AlR3) was injected
followed by a toluene (20 mL) solution of the catalyst. Alternatively,
a solution of the catalyst that had been pretreated with AlR3
followed by B(C6F5)3 was added. After addition of Cr, the reactor
was immediately charged with 40 bar of ethylene and maintained
at this pressure for the duration of the reaction. After 30 min the
reactor was cooled in an ice bath, the excess ethylene bled, and an
internal standard added (nonane, 1000 µL). After quenching with
MeOH followed by 10% HCl, the organic phase was analyzed by
GC, while solids were filtered, washed, dried, and weighed.
CrCl2(HL1). A solution of (Ph2PCH2CH2)2NH (1.748 g, 3.96
mmol) in 20 mL of THF was added to a suspension of CrCl2(thf)2
(1.00 g, 3.74 mmol) in 20 mL of THF. The solution was stirred
for 2 h and the THF removed under vacuum. The product was
recrystallized from 40 mL of hot acetonitrile to give blue crystals,
Acknowledgment. The authors thank the members of the
Sasol Technology Olefin Transformations group for helpful
discussions and suggestions.
Supporting Information Available: Crystallographic data, in
CIF format, for structures reported. Preparation and crystal structure
of LiCrCl2(µ-Cl)2(THF)2 (PDF). This material is available free of
charge via the Internet at http://pubs.acs.org.
OM0601091