Mechanistic investigations of the ethylene tetramerisation reaction

Article abstract of DOI:10.1021/ja052327b

The unprecedented selective tetramerisation of ethylene to 1-octene was recently reported. In the present study various mechanistic aspects of this novel transformation were investigated. The unusually high 1-octene selectivity in chromium-catalyzed ethylene tetramerisation reactions is caused by the unique extended metallacyclic mechanism in operation. Both 1-octene and higher 1-alkenes are formed by further ethylene insertion into a metallacycloheptane intermediate, whereas 1-hexene is formed by elimination from this species as in other reported trimerisation reactions. This is supported by deuterium labeling studies, analysis of the molar distribution of 1-alkene products, and identification of secondary co-oligomerization reaction products. In addition, the formation of two C₆ cyclic products, methylenecyclopentane and methylcyclopentane, is discussed, and a bimetallic disproportionation mechanism to account for the available data is proposed.

Full text of DOI:10.1021/ja052327b

Published on Web 07/08/2005
Mechanistic Investigations of the Ethylene Tetramerisation
Reaction
Matthew J. Overett,*, Kevin Blann, Annette Bollmann, John T. Dixon,
†
†
†
†
†
†
†
‡
Daleen Haasbroek, Esna Killian, Hulisani Maumela, David S. McGuinness, and
†
David H. Morgan
Contribution from the Sasol Technology (Pty) Ltd, R&D DiVision, 1 Klasie HaVenga Road,
Sasolburg 1947, South Africa, and Sasol Technology U.K. Ltd., Purdie Building, North Haugh,
St. Andrews, KY16 9ST, United Kingdom
Received April 11, 2005; Revised Manuscript Received May 25, 2005; E-mail: Matthew.Overett@Sasol.com
Abstract: The unprecedented selective tetramerisation of ethylene to 1-octene was recently reported. In
the present study various mechanistic aspects of this novel transformation were investigated. The unusually
high 1-octene selectivity in chromium-catalyzed ethylene tetramerisation reactions is caused by the unique
extended metallacyclic mechanism in operation. Both 1-octene and higher 1-alkenes are formed by further
ethylene insertion into a metallacycloheptane intermediate, whereas 1-hexene is formed by elimination
from this species as in other reported trimerisation reactions. This is supported by deuterium labeling studies,
analysis of the molar distribution of 1-alkene products, and identification of secondary co-oligomerization
reaction products. In addition, the formation of two C cyclic products, methylenecyclopentane and
6
methylcyclopentane, is discussed, and a bimetallic disproportionation mechanism to account for the available
data is proposed.
Background
Scheme 1. First Selective Ethylene Tetramerisation Reaction
Commercial processes for the production of linear alpha-
olefins (LAOs) typically rely on nonselective ethylene oligo-
1
merization reactions. More selective processes for the produc-
tion of the higher-value LAOs are highly desirable from an
industrial viewpoint. The selective trimerisation of ethylene to
2
Scheme 2. Catalytic Cycle for the Trimerisation of Ethylene
1-hexene is a well-known technology. However, until recently,
a corresponding selective tetramerisation reaction of ethylene
to give 1-octene was unknown. We recently reported the first
catalyst capable of this transformation with selectivities of up
3
to 70% 1-octene. The active catalyst is typically generated in
situ by the addition of an aluminoxane (e.g., methylalumin-
oxane) to a mixture of a Cr(III) source and a bidentate
diphosphinoamine ligand (Scheme 1).
The observed selectivity in ethylene trimerisation reactions
is a consequence of the unusual metallacyclic mechanism in
4
operation (Scheme 2). Oxidative coupling of two ethylene
molecules with the active catalytic metal gives a metallacyclo-
pentane intermediate. The geometrical constraints of this species
limit interaction of the â-hydrogens with the metal, thus
preventing the elimination of 1-butene. Coordination and
subsequent insertion of another ethylene is thus facilitated, and
the resultant metallacycloheptane is sufficiently flexible for
â-hydride transfer and reductive elimination of 1-hexene. Until
recently, it was thought unlikely that further ethylene insertion
to give a metallacyclononane could compete with 1-hexene
elimination or that selective elimination of 1-octene from such
†
Sasol Technology.
Sasol Technology, U.K.
‡
(
1) Alpha Olefins (02/03-4), PERP report, Nextant Chem Systems.
2) Dixon, J. T.; Green, M. J.; Hess, F. M.; Morgan, D. H. J. Organomet.
Chem. 2004, 689, 3641.
(
(
3) 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. J.; Slawin, A. M. Z.; Wasserscheid, P.; Kuhlmann, S. J. Am. Chem.
Soc. 2004, 126, 14712.
(
4) (a) Briggs, J. R. J. Chem. Soc., Chem. Commun. 1989, 674. (b) Emrich,
R.; Heinemann, O.; Jolly, P. W.; Kruger, C.; G. P. J. Verhovnik, G. P. J.
Organometallics 1997, 16, 1511. (c) Meijboom, N.; Schaveien, C.; Orpen,
A. G., Organometallics 1990, 9, 774. (d) Agapie, T.; Schofer, S. J.;
Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc, 2004, 125, 1304.
5
a species would take place.
(5) (a) Yu, Z.-X.; Houk, K. N. Angew. Chem., Int. Ed. 2003, 42, 808. (b) Blok,
A. N. J.; Budzelaar, P. H. M.; Gal, A. W. Organometallics 2003, 22, 2564.
10.1021/ja052327b CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 10723-10730
9
10723

A R T I C L E S
Overett et al.
Scheme 3. Postulated Mechanisms for the Formation of 1-Octene
Scheme 4. Postulated Mechanisms for the Formation of
Methylcyclopentane and Methylenecyclopentane
Mechanistic Proposals
We have previously shown that ethylene tetramerisation and
trimerisation reactions with Cr/diphosphine/aluminoxane cata-
lysts are closely related and that the selectivity to 1-hexene or
1
-octene may be controlled by subtle ligand modifications
6
through both steric and coordinative effects. It thus seems
reasonable that a common metallacycloheptane intermediate4
(
1, Scheme 3), proven for related Cr/diphosphine-catalyzed
4
d
trimerisation reactions, is involved in the pathways to both
-hexene and 1-octene. Two proposed mechanisms for the
1
formation of 1-octene may thus be considered (Scheme 3). In
the first possibility, a further coordination and insertion of
ethylene into the metallacycloheptane intermediate 1 (pathway
A) is followed by 1-octene elimination from the resultant
metallacyclononane species (2). This elimination, either via
4
a
formal â-hydride transfer to the metal or a concerted metal
assisted hydride transfer, is facile relative to further insertion
of ethylene to give higher metallacycles. In the second pos-
Table 1. Rate and Selectivity for an Ethylene Tetramerisation
Reaction
5a,7
Cr
mol) Al:Cr
temp press time
activity
C
4
1-hexene C
(%)
6
cyclics 1-octene C10
−
C
14 PE
sibility, â-hydride transfer from the metallacycloheptane to a
(µ
(
°
C) (bar) (min) (g/g Cr) (%)
(%)
(%)
(%) (%)
8
coordinated ethylene (or to the metal with subsequent ethylene
6
5
500 60 45 45 1.2 × 10 0.8 13.8
4.1
66.3
10.4 0.3
insertion into Cr-H) gives a hexenyl ethyl Cr species (3), and
reductive elimination to give 1-octene is facile relative to linear
chain growth of this species (pathway B).
The discovery of the selective ethylene tetramerisation
reaction is thus of profound interest from a mechanistic point
of view. An understanding of its mechanism may lead to the
improvement of 1-octene selectivity and to the design of new
tetramerisation catalysts. We report here the results of a study
that confirms the intermediacy of metallacyclic species in the
formation of 1-octene and allows us to propose a mechanism
that accounts for the primary and secondary products formed
in a typical tetramerisation reaction.
Methylcyclopentane and methylenecyclopentane are typically
formed in a ratio of 1:1 and are the third most abundant products
formed after 1-octene and 1-hexene. The mechanism of forma-
tion of these cyclic products is thus of interest, particularly as
methylenecyclopentane is not a stoichiometric oligomer of
ethylene. We propose that rearrangement of 1 to the Cr
cyclopentylmethyl hydride species 4 competes with 1-hexene
elimination and ethylene insertion (Scheme 4). This may take
place via a formal â-hydride transfer to metal, with a subsequent
cyclization of the 5-hexenyl moiety analogous to that in the
Analysis of an Ethylene Tetramerisation Product
Mixture
cyclopolymerization of 1,5-hexadiene to form poly(methylene-
An ethylene tetramerisation reaction was performed using a
9
1
,3-cyclopentane). Alternatively, the rearrangement may take
i
catalyst generated in situ from Cr(acac)3, Ph2PN( Pr)PPh2, and
place in a single concerted step. A cyclopentylmethyl intermedi-
ate species has also been invoked to account for the zirconium-
catalyzed oligomerization of ethylene to methylenecyclo-
MMAO.3 The product mixture was then analyzed by gas
chromatography flame-ionization detection (GC-FID) and GC
mass spectrometry (GC-MS) (Table 1). Although selectivities
to 1-octene of close to 70 mass % can be achieved, significant
quantities of other products are also formed, including 1-butene,
(
6) (a) Blann, K.; Bollmann, A.; Dixon, J. T.; Hess, F. M.; Killian, E.; Maumela,
H.; Morgan, D. H.; Neveling, A.; Otto, S.; Overett, M. J. Chem. Commun.
2
005, 620. (b) Overett, M. J.; Blann, K.; Bollmann, A.; Dixon, J. T.; Hess,
1-hexene, methylcyclopentane, methylenecyclopentane, 2-pro-
F. M.; Killian, E.; Maumela, H.; Morgan, D. H.; Neveling, A.; Otto, S.
Chem. Commun. 2005, 622.
penylcyclopentane, n-propylcyclopentane, C10-C36+ 1-alkenes,
C4-C20+ n-alkanes, and a complex mixture of C10, C12, and
C14 secondary oligomerization products. Detailed product
identification is given in the Supporting Information. A complete
mechanism for the ethylene tetramerisation reaction must thus
account for the formation of all these products.
(
7) (a) Janse van Rensberg, W. J.; Grov e´ , C.; Steynberg, J. P.; Stark, K. B.;
Huyser, J. J.; Steynberg, P. J. Organometallics 2004, 23, 1207. (b) Tobisch
S., Ziegler, T. Organometallics 2003, 22, 5392.
(8) Manyik, R. M.; Walker, W. E.; Wilson, T. P. J. Catal. 1977, 47, 197.
(9) (a) Marvel, C. S.; Stille J. K. J. Am. Chem. Soc. 1958, 80, 1740. (b) Resconi,
L.; Waymouth, R. M. J. Am. Chem. Soc. 1990, 112, 4953. (c) Cavallo, L.;
Guerra, G.; Corradini, P.; Resconi, L.; Waymouth, R. M. Macromolecules
1993, 26, 260.
10724 J. AM. CHEM. SOC.
9
VOL. 127, NO. 30, 2005

Ethylene Tetramerisation Reaction
A R T I C L E S
Figure 1. Observed 1-hexene isotopomer distribution against predicted distributions for the metallacycle and Cossee-Arlman mechanisms.
Figure 2. Observed 1-octene isotopomer distribution against predicted distributions for the metallacycle and Cossee-Arlman mechanisms.
pentane.10 From this intermediate, two alternatives may be
considered (Scheme 4). First, the Cr cyclopentylmethyl hydride
species undergoes two competitive reactions with similar rates,
namely, reductive elimination to give methylcyclopentane and
â-hydride elimination to give methylenecyclopentane (pathways
C and D). Second, however, the equimolar formation of
methylcyclopentane and methylenecyclopentane under a variety
of reaction conditions is suggestive of a disproportionation
process (pathway E). For example, it is known that dispropor-
tionation of cyclopentylmethyl free radicals yields these two
Deuterium-Labeling Studies
The metallacycle mechanism for ethylene trimerisation was
for the first time conclusively demonstrated by Bercaw and co-
workers using a series of experiments with deuterated and
4d
partially deuterated ethylene. For example, a 1:1 mixture of
C2H4 and C2D4 was trimerised using the Cr/(o-MeOPh)2PN-
(
Me)P(o-MeOPh)2/MAO catalyst system first described by Wass
12
and co-workers. The resultant 1-hexene isotopomer distribution
(11) (a) Sheldon, R. A.; Kochi, J. K. J. Am. Chem. Soc. 1970, 92, 4395.
(b) Jenkins, C. L.; Kochi, J. K. J. Am. Chem. Soc. 1972, 94, 843. (c) Foster,
D. F.; Glidewell, C.; Cole-Hamilton, D. J.; Povey, I. M.; Hoare, R. D.;
Pemble, M. E. J. Cryst. Growth 1994, 145, 104. (d) Foster, D. F.; Bell,
W.; Stevenson, J.; Cole-Hamilton, D. J.; Hails, J. E. J. Cryst. Growth 1994,
145, 520.
1
1
cyclic products.
(
10) Wang, M.; Shen, Y.; Qian, M.; Li, R.; He, R. J. Organomet. Chem. 2000,
5
99, 143.
J. AM. CHEM. SOC.
9
VOL. 127, NO. 30, 2005 10725

A R T I C L E S
Overett et al.
Figure 3. Observed methylcyclopentane isotopomer distribution against predicted distributions for the metal-mediated reductive elimination and
dispropotionation mechanisms
of a 1:3:3:1 ratio of C6H12, C6H8D4, C6H4D8, and C6D12 was
consistent with a metallacyclic mechanism involving no H/D
scrambling.
and 2.4, respectively), the value of x for 1-octene was
significantly lower (1.9). This implies that secondary kinetic
isotope effects (higher reactivity of standard ethylene vs
deuterated ethylene) are in operation. These kinetic isotope
effects are expected to be higher for the rate-determining step
in the reaction. Thus the greater preference for standard ethylene
incorporation in the C6 products suggests that the rate-
determining step is the oxidative coupling of the first two
ethylenes to form the metallacyclopentane intermediate and not
the subsequent ethylene insertions to give metallacycloheptane
and metallacyclononane species. This agrees with other studies
that have used the second order dependence of rate on ethylene
concentration in trimerisation reactions to deduce that formation
Similarly, in the present study, a mixture of C2H4 and C2D4
was exposed to a tetramerisation catalyst generated in situ by
i
the combination of Cr(acac)3, Ph2PN( Pr)PPh2, and an alumin-
oxane cocatalyst. The products of this reaction were then
analyzed by GC-FID and GC-MS. 1-Octene, 1-hexene, and
methylcyclopentane were obtained in sufficient amounts to
obtain quantitative isotopomer distributions in the GC-MS
analysis. The ratio of C2H4 to C2D4 incorporated in each product
may be calculated, independent of mechanism, from the
isotopomer distribution. By use of this value, and a first order
correction for the natural abundance of deuterium in hydrogen
8
of the metallacyclopentane is rate determining.
(
(
1
0.79% D) and the imperfect labeling of the deuterated ethylene
0.5% H), theoretical isotopomer distributions for 1-octene and
The formation of methylcyclopentane by reductive elimina-
tion of the Cr cyclopentylmethyl hydride species 4 (pathway
C, Scheme 4) involves no H/D scrambling, and thus the
isotopomer distribution should follow the same pattern as for
1-hexene. On the other hand, a disproportionation mechanism
of cyclic formation does involve H/D scrambling in the
disproportionation step, and this should allow distinction
between the two mechanisms by analysis of the isotopomer
-hexene produced by both the proposed metallacyclic mech-
13
anism and the Cossee-Arlman linear chain growth mechanism
14
may be calculated. The correlation of the observed isotopomer
distribution with the predicted distribution for the metallacyclic
mechanism is excellent for both these products (Figures 1 and
2
). For 1-octene formation, however, this experiment does not
distinguish between the two mechanisms in Scheme 3, as neither
mechanism involves H/D scrambling. The formation of 1-hexene
and 1-octene via the Cossee-Arlman mechanism, as expected,
is however excluded.
distributions of this C cyclic product. It can clearly be seen
that the observed distribution does not correlate well with the
predicted distribution for a mechanism, such as reductive
elimination, where no scrambling is involved (Figure 3).
6
It is worth noting that the ratio of standard ethylene to
deuterated ethylene incorporated (x) differs for the C6 and C8
products. Whereas x was determined to be the same within
experimental error for 1-hexene and methylcyclopentane (2.5
The calculation of a theoretical distribution for the dispro-
portionation mechanism is complicated by two kinetic isotope
effects involved, namely, for the â-hydride transfer to the metal
(
k1) and for the hydrogen transfer of the disproportionation step
14
itself (k2). The values of k1 (equivalent to kH/kD if irreversibility
is assumed) and k2 are unknown. However, by application of a
least-squares optimization of the theoretical distribution against
the observed distribution, with k1 and k2 as variables, a good fit
between the theoretical and observed distributions is obtained
(
(
(
12) Carter, C.; Cohen, S. A.; Cooley, N. A.; Murphy, A.; Scutt, J.; Wass, D.
F. Chem. Commun. 2002, 858.
13) (a) Cossee, P. J. Catal. 1964, 3, 80. (b) Arlman, E. J.; Cossee, P. J. Catal.
1
964, 3, 99.
14) Details of the mathematical derivations are given in the Supporting
Information.
10726 J. AM. CHEM. SOC.
9
VOL. 127, NO. 30, 2005

Ethylene Tetramerisation Reaction
A R T I C L E S
Figure 4. Molar amounts of higher 1-alkenes as a percentage of 1-octene formed.
(
(
Figure 3) and reasonable values of k1 and k2 are deduced
k1 ) 3.0; k2 ) 4.9). By comparison, kH/kD (also assuming
These results provide evidence in support of an extended
metallacycle mechanism for the formation of all 1-alkenes in
the tetramerisation reaction. The observed selectivities are
governed by the relative rates of 1-alkene elimination vs further
ethylene insertion. Clearly, a key role of the ligand in tetrameri-
sation is to stabilize the metallacycloheptane but not the
metallacyclononane intermediate against 1-alkene elimination.
irreversibility) was determined as 2.4 for the â-hydride transfer
from metallacycloheptane at room temperature in a closely
related trimerisation system.4d
Analysis of C10+ 1-Alkenes
Tetramerisation reactions typically produce a full range of
higher 1-alkenes, albeit as minor products. These oligomers may
be formed by extension of a metallacyclononane through further
insertions of ethylene or by a different, linear chain growth,
catalytic species. In the former case, the distribution of 1-alkenes
will depend on the relative lability of each extended metallacycle
vs the probability of further insertion. In the latter case, a
Schulz-Flory distribution, typical of linear chain growth
mechanisms, should be apparent.
Analysis of Secondary (C10, C12, and C14) Products
Tetramerisation reactions produce a complex mixture of
oligomers from co-trimerisation and co-tetramerisation of ethyl-
ene with 1-hexene and 1-octene. The amount of these products
formed is proportional to the productivity of the reaction, i.e.,
the ratio of 1-hexene and 1-octene to available ethylene. Further
proof that these products are not formed directly from ethylene
is obtained by spiking a tetramerisation reaction with 1-pentene.
In this case, C9 co-trimerisation and C11 co-tetramerisation
products are also formed. However, only trace quantities of
A series of ethylene tetramerisation reactions were performed
under three different temperature and pressure conditions
(45 bar ethylene, 45 °C; 45 bar, 60 °C; 80 bar, 60 °C). The
1
-nonene and 1-undecene are formed, which further demon-
relative molar amounts of all linear 1-alkenes above C8 were
carefully determined by GC-FID and scaled as a percentage of
the molar amount of 1-octene produced (Figure 4). The resulting
distribution is inconsistent with a linear chain growth mecha-
nism, where the molar amount of Cn+2 1-alkene cannot exceed
that of Cn 1-alkene. This is especially clear under conditions of
high ethylene solubility (45 bar ethylene, 45 °C; 80 bar, 60 °C
reactions), where 1-hexadecene exceeds 1-tetradecene, which
in turn equals or exceeds 1-dodecene.
strates that the substantially more abundant 1-decene, 1-dodecene,
and 1-tetradecene are not formed as secondary products.
Scheme 5. Proposed Formation of 7-Methylenetridecane and
7-Methyl-1-tridecene
The observed distribution is consistent with an extended
metallacycle mechanism, where subtle variations in conforma-
tion confer greater stability against 1-alkene elimination on some
metallacycles than others. By assuming no difference in the rate
of ethylene insertion, the relative stabilities of a Cr-Cn
metallacycle may be approximated by the 1-alkene molar ratio
of ∑Cx: Cn (for all x > n). By this measure, the Cr-C12 and
Cr-C14 metallacycles are the most stable of the metallacycles
Cr-Cn (n g 8) Similarly, the metallacyclononane(Cr-C8) is
the least stable of the large metallacycles, which is the reason
for the high selectivity to 1-octene.
Analysis of the structures of these co-oligomerization products
gives further insight into the mechanism of tetramerisation. Of
particular interest are the branched C14 products, as these can
only be formed by co-tetramerisation reactions. Of the two most
abundant C14 oligomers (ca. 70% of the secondary C14 products),
one was positively identified as 7-methylenetridecane. Upon
hydrogenation, these two oligomers were converted to a single
product, 7-methyltridecane, thereby demonstrating that they are
J. AM. CHEM. SOC.
9
VOL. 127, NO. 30, 2005 10727

A R T I C L E S
Overett et al.
isomers with the same skeletal structure.15 It may thus reason-
ably be postulated that the second isomer is 7-methyl-1-
tridecene. Formation of these co-tetramers is readily rationalized
from a substituted metallacyclononane intermediate as shown
in Scheme 5. The preference for this mode of co-oligomerization
Scheme 6. Postulated Binuclear Mechanism for the Formation of
the C6 Cyclics
(1,2-insertion of alkene adjacent to metal) has previously been
reported for trimerisation reactions, where 5-methyl-1-nonene
was found to be the most abundant co-trimer.16 The hydroge-
nated C14 region contains only two peaks apart from 7-meth-
yltridecane and tetradecane. While these were not identified,
hydrogenation of the four C14 oligomers deriving from the other
two possible modes of 1-octene incorporation into a metalla-
cyclononane (4- and 5-hexylmetallacyclononane) would be
expected to give two branched alkanes (6-ethyldodecane and
5
-propylundecane).
The predominance of 7-methylene-tridecane and 7-methyl-
-tridecene as secondary C14 products cannot plausibly be
1
explained by the mechanism involving a â-hydride transfer from
metallacycloheptane to a coordinated alkene (pathway B,
Scheme 3). While in principle these products are accessible via
such a pathway, they would be derived from very different
modes of incorporation, and 5-propyl-1-undecene and 1-tetra-
decene (as a secondary product) would be expected to be more
abundant than 7-methylene-tridecane and 7-methyl-1-tridecene,
respectively.15
GC-MS analysis was made. This indicated the absence of the
coupled product. Furthermore, tetramerisation reactions con-
ducted in the presence of an excellent H• donor such as
triphenylmethane still produce 1:1 ratios of the C6 cyclic
products, where more methylcyclopentane may have been
expected. These results strongly suggest that free radicals are
not involved.
As with the C14 oligomers, the C10 and C12 secondary products
that could be identified are expected co-oligomerization products
deriving from metallacyclic catalytic intermediates with hexene
or octene incorporation.15
The mechanism of cyclic formation thus remains unproven;
however, a recent report by Theopold and co-workers suggests
a plausible answer. Reaction of the Cr(III) compound
[nacnacCr(CH2SiMe3)2] (nacnac ) N,N′-bis(2,6-dimethylphe-
nyl)-2,4-pentanediiminato) with H2 gives the Cr(II) alkyl hydride
Formation of the Cyclic Products
19
species [(nacnacCr)2(µ-CH2SiMe3)(µ-H)]. This transformation
proceeds by hydrogenolysis of one alkyl group followed by a
binuclear reductive elimination to give SiMe4 and the above
binuclear Cr(II) complex, which is remarkably stable against
alkane elimination. A similar reaction occurred in attempts to
prepare [nacnacCrEt2]. The isolated species was found to be
The 1:1 ratio of methylcyclopentane and methylenecyclo-
pentane formed under a variety of reaction conditions is
plausibly explained by a disproportionation process but not
easily rationalized by competitive reductive elimination and
â-hydride elimination reactions from a common intermediate.
In addition, the deuterium labeling results are more consistent
with a disproportionation mechanism.
[(nacnacCr)2(µ-H)2]. This was reportedly formed by successive
â-hydride elimination of the diethyl complex, binuclear reductive
The formation of methylcyclopentane and methylenecyclo-
pentane from cyclopentylmethyl or 5-hexen-1-yl free radicals
is well established.11 With this in mind, homolytic Cr-alkyl
cleavage of 4 (Scheme 4) may be envisioned as the origin of
cyclopentylmethyl radicals. Of relevance in this regard is
chemistry reported by Theopold and co-workers in which
reaction of the Cr(III) compound [Cp*Cr(THF)(bz)2] with
elimination of ethane, presumably yielding [(nacnacCr) (µ-Et)-
(µ-H)], and a second â-hydride elimination to give the Cr(II)
2
hydride dimer. The analogy of this chemistry to the proposed
Cr cyclopentylmethyl hydride species 4 (Scheme 4) is direct
and allows the proposal of a bimetallic disproportionation
mechanism (Scheme 6) to account for the equimolar formation
of the C6 cyclic products, the results of the deuterium labeling
studies, and the absence of free radical byproducts in tetrameri-
sation product mixtures.
2
,2′-bipyridyl gave the unexpected Cr(II) product [Cp*Cr(bipy)-
(
bz)].17 A homolytic reduction yielding a benzyl radical byprod-
uct was hypothesized. However, in addition to disproportion-
ation, free cyclopentylmethyl radicals also couple to give
It should be noted that this proposed pathway leaves the
n+1
catalytic chromium in a formal Cr oxidation state, assuming
1
,2-dicyclopentylethane and abstract H• from other available
n
n+2
a Cr /Cr
-octene. The active catalyst thus needs to be “regenerated” in
catalytic cycle for the formation of 1-hexene and
11a
sources, such as the solvent. To test for the intermediacy of
1
free radicals, comparison of tetramerisation product mixtures
n
n+2
some process which restores the Cr /Cr
oxidation state and
1
8
with an authentic sample of 1,2-dicyclopentylethane by
the overall hydrogen balance of the catalytic reaction. One
possibility is the ethylene mediated disproportionation of the
(
15) See Supporting Information for chromatograms of the C14 region (un-
hydrogenated and hydrogenated) and further mechanistic analysis of the
observed secondary products.
n+1
n+2
n
Cr hydride dimer into Cr dihydride and Cr bis(ethylene)
mononuclear species; however this remains just one of several
possibilities.
(
(
(
16) Deckers, P. J. W.; Hessen, B.; Teuben, J. H. Organometallics 2002, 21,
5
122.
17) Bhandari, G.; Kim, Y.; McFarland, J. M.; Rheingold, A. L.; Theopold, K.
H. Organometallics 1995, 14, 738.
18) Zelinsky, N. D.; Michlina, S. E.; Eventowa, M. S. Chem. Ber. 1933, 66,
(19) MacAdams, L. A.; Buffone, G. P.; Incarvito, C. D.; Golen, J. A.; Rheingold,
1
422.
A. L.; Theopold, K. H. Chem. Commun. 2003, 1164.
10728 J. AM. CHEM. SOC.
9
VOL. 127, NO. 30, 2005

Ethylene Tetramerisation Reaction
A R T I C L E S
Table 2. Effect of Temperature and Pressure on Formation of Higher 1-Alkenes
temp
C)
press
(bar)
1-hexene
(mass %)
1-octene
(mass %)
1-decene
1-C12
1-C14
1-C16
1-C18
1-C20
(
°
(molar % of 1-octene)
(molar % of 1-octene)
(molar % of 1-octene)
(molar % of 1-octene)
(molar % of 1-octene)
(molar % of 1-octene)
4
6
6
5
0
0
45
45
80
11.5
13.7
10.7
68.8
66.4
69.9
1.04
1.19
1.22
0.36
0.48
0.41
0.42
0.32
0.43
0.72
0.48
0.69
0.65
0.26
0.62
0.46
0.15
0.50
Table 3. Ethylene Tetramerisation in the Presence of 1-Pentene
60 °C and pressurized to 45 bar with ethylene. A gas-entraining stirrer
at 1000 rpm was used to obtain thorough mixing. The reaction mixture
was maintained at 60 °C, and ethylene was fed on demand. After
activity
1-hexene
(%)
C
6
cyclics 1-octene
C
9
C
10
C
11
C
12
C
14
P.E.
(%)
(g/g Cr)
(%)
(%)
(%)
(%)
(%)
(%)
(%)
4
5 min, the ethylene supply was terminated and the reaction mixture
5
1
.2 × 10
15.8
4.3
67.3
1.4 1.7 1.9 3.1 2.1 0.2
was cooled to 10 °C and depressurized. The liquid products were
analyzed by GC-FID to determine selectivities and the solids were
recovered by filtration. The mass of the product formed was 312 g.
After filtration through a plug of silica, the light products and solvent
were removed in vacuo into a trap at -196 °C. The two fractions
thus obtained (concentrated in light products and heavy products
respectively) were analyzed by GC-MS. Full-product identification
is given in the Supporting Information. The heavy fraction was
A complete correspondence of the mechanism of cyclic
formation with the chemistry reported by Theopold and co-
workers would imply that the active tetramerisation catalyst
adopts a Cr /Cr catalytic cycle with Cr intermediates involved
in the formation of the cyclics. However, elucidation of the exact
structures and oxidation states of the active catalytic species is
not in the ambit of this study.
I
III
II
hydrogenated (10% Pd on C, 50 bar H
GC-MS.
2
) and again analyzed by
Deuterium-Labeling Study. A U-shaped 1/
in. metal pipe contain-
2
Conclusions
ing a few glass beads was fitted with ball valves at each end, dried in
the oven, and cooled under vacuum. Deuterated ethylene was condensed
from a lecture bottle into the tube at -196 °C. The tube was sealed
and warmed to room temperature, and the pressure (∼15 bar) was
measured using a gauge fitted to the one end. Standard ethylene was
A variety of evidence has been obtained which supports an
extended metallacyclic mechanism for the formation of all the
-alkenes in ethylene tetramerisation reactions. The unique
feature of the tetramerisation catalyst is the enhanced stability
relative to trimerisation catalysts) against 1-hexene elimination
1
then added up to 35 bar. The catalyst mixture [Cr(acac)
3
(7.0 mg,
(10.7 mg, 25 µmol), and MMAO-12
9.0 mmol Al) in 1,3-diisopropylbenzene (15 mL)] was forced in with
high-pressure N . The temperature of the reaction was maintained at
(
i
2
(
2 2
0 µmol), Ph PN( Pr)PPh
from the metallacycloheptane intermediate. This stability allows
competitive reactions leading to the formation of 1-octene
2
(
ethylene insertion) and cyclic products (rearrangement and
40 °C by immersing the tube in warm water, and mass transfer of
ethylene into the reaction mixture was ensured by shaking the tube
regularly. After 1 h, the mixture was cooled to 0 °C and slowly
depressurized. Ethanol (0.5 mL) was added, and the volatile products
were distilled at 70 °C under vacuum into a trap at -196 °C. The clear
liquid product was analyzed by GC-FID and GC-MS. Details of the
resulting isotopomer distributions are given in the Supporting Informa-
tion.
disproportionation). The high selectivity to 1-octene is thus a
consequence of both the relative stability of the metallacyclo-
heptane intermediate and the instability of the metallacy-
clononane species. Clearly, these differences with respect to
ethylene trimerisation systems must be caused by subtle ligand-
mediated electronic and steric effects. Molecular modeling
studies into the fundamental origins of the observed selectivities
in tetramerisation reactions are underway.
Analysis of Higher 1-Alkenes. Runs were conducted under three
different temperature and pressure conditions. A stirred mixture of
i
Experimental Section
Cr(acac)
3
(3.49 mg, 10 µmol) and Ph
2
PN( Pr)PPh
2
(6.4 mg, 15 µmol)
in methylcyclohexane (10 mL) was added to an autoclave containing
methylcyclohexane (90 mL) and MMAO-3A (5.0 mmol Al). The reactor
was pressurized with ethylene and maintained at the required temper-
ature and pressure throughout the reaction. A gas-entraining stirrer at
General Considerations. Methylcyclohexane, 1,3-diisopropyl-
benzene, and 1-pentene were purified by passage through a column of
i
activated alumina. 1,2-Dicyclopentylethane and Ph
prepared by literature procedures.
2
PN( Pr)PPh
2
were
1
8,20
MMAO-3A (in heptanes,
1
000 rpm was used to obtain thorough mixing. When the reactor was
methylaluminoxane with ca. 30% replacement of methyl groups by
isobutyl groups) and MMAO-12 (in toluene, c.a. 5% replacement of
methyl groups by n-octyl groups) were obtained from Akzo-Nobel.
full, the ethylene supply was terminated and the reaction mixture was
cooled to 10 °C and depressurized. The liquid products were analyzed
by GC-FID, and the solids were recovered by filtration. The selectivities
are shown in Table 2.
Ethylene-d
conducted on instruments equipped with a PONA 50 m × 0.5 µm ×
.2 mm inside diameter (i.d.) column and either electron impact
4
was obtained from Sigma Aldrich. GC analyses were
Ethylene Tetramerisation in the Presence of 1-Pentene. MMAO-
0
3
A (5.0 mmol Al) was added to a stirred mixture of Cr(acac)
3
ionization mass spectrometry (GC-MS) or FID. Identification of the
components was done with a combination of the Wiley275 library,
boiling point information, interpretation of the mass spectra with first
i
2 2
(3.50 mg, 10.0 µmol) and Ph PN( Pr)PPh (6.4 mg, 15.0 µmol) in
methylcyclohexane (5 mL). This solution was immediately added to
an autoclave containing methylcyclohexane (115 mL) and 1-pentene
principles, and elution order.
(
80 mL) at 60 °C and pressurized to 45 bar with ethylene. A gas-
i
Ethylene Tetramerisation Reaction with Cr(acac)
PPh
mixture of Cr(acac)
.5 µmol) in methylcyclohexane (5 mL). This solution was immediately
added to an autoclave containing methylcyclohexane (195 mL) at
3
/Ph
2
PN( Pr)-
entraining stirrer at 1000 rpm was used to obtain thorough mixing.
The reaction mixture was maintained at 60 °C, and ethylene was fed
on demand. After 45 min, the ethylene supply was terminated and the
reaction mixture was cooled to 10 °C and depressurized. The liquid
products were analyzed by GC-FID and GC-MS, and the solids were
recovered by filtration. The mass of the product formed was 150 g,
and the selectivities are shown in Table 3.
2
/MMAO-3A. MMAO-3A (2.5 mmol Al) was added to a stirred
i
3
2 2
(1.75 mg, 5.0 µmol) and Ph PN( Pr)PPh (3.2 mg,
7
(
20) Balakrishna, M. S.; Prakasha, T. K.; Krishnamurthy, S. S. J. Organomet.
Chem. 1990, 390, 203.
J. AM. CHEM. SOC.
9
VOL. 127, NO. 30, 2005 10729

A R T I C L E S
Overett et al.
Acknowledgment. The authors thank Sasol Technology (Pty)
Ltd for permission to publish this work and acknowledge
Professor David Cole-Hamilton for fruitful discussions.
products and mathematical derivations of predicted isotopomer
distributions in the deuterium labeling studies for various
mechanisms. This material is available free of charge via the
Internet at http://pubs.acs.org.
Supporting Information Available: Full product analysis of
a typical ethylene tetramerisation reaction, including secondary
JA052327B
10730 J. AM. CHEM. SOC.
9
VOL. 127, NO. 30, 2005