Insertion and isomerisation of internal olefins at alkylaluminium hydride: Catalysis with zirconocene dichloride

Article abstract of DOI:10.1039/c5dt03257a

The insertion of internal olefins (hydroalumination) and chain walking isomerisation at di-n-octylaluminium hydride [Al(Oct)₂H], promoted by zirconocene dichloride [Cp₂ZrCl₂] has been studied. The reaction between [Cp₂ZrCl₂] and [Al(Oct)₂H] in non-polar solvents leads to clusters containing bridging hydride ligands between Zr and Al. This system promotes hydroalumination of 1-octene but is largely ineffective for internal octenes (2-, 3-, 4-octene). In tetrahydrofuran the Zr-Al hydride clusters formed are more reactive and catalyse insertion and isomerisation of internal olefins to primary metal-alkyls, although this is accompanied by catalyst deactivation. Elimination and removal of 1-octene from the system post insertion/isomerisation was attempted, but it was found that the presence of the Zr catalyst leads to back-isomerisation to internal octenes, along with further decomposition with n-octane formation. Some possible pathways of catalyst decomposition, involving reduction of Zr and alkane elimination, have been studied theoretically.

Full text of DOI:10.1039/c5dt03257a

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Insertion and isomerisation of internal olefins
at alkylaluminium hydride: catalysis with
zirconocene dichloride†
Cite this: Dalton Trans., 2015, 44,
20098
Nandita M. Weliange,a David S. McGuinness,*a Michael G. Gardinera and Jim Patelb
The insertion of internal olefins (hydroalumination) and chain walking isomerisation at di-n-octylaluminium
hydride [Al(Oct)2H], promoted by zirconocene dichloride [Cp2ZrCl2] has been studied. The reaction between
[Cp2ZrCl2] and [Al(Oct)2H] in non-polar solvents leads to clusters containing bridging hydride ligands
between Zr and Al. This system promotes hydroalumination of 1-octene but is largely ineffective for internal
octenes (2-, 3-, 4-octene). In tetrahydrofuran the Zr–Al hydride clusters formed are more reactive and cata-
lyse insertion and isomerisation of internal olefins to primary metal–alkyls, although this is accompanied by
catalyst deactivation. Elimination and removal of 1-octene from the system post insertion/isomerisation was
attempted, but it was found that the presence of the Zr catalyst leads to back-isomerisation to internal
octenes, along with further decomposition with n-octane formation. Some possible pathways of catalyst
decomposition, involving reduction of Zr and alkane elimination, have been studied theoretically.
Received 24th August 2015,
Accepted 27th October 2015
DOI: 10.1039/c5dt03257a
www.rsc.org/dalton
Introduction
Linear α-olefins (LAOs) are an important feedstock of the
modern chemical industry.1 The typical route to produce LAOs
involves metal catalysed ethylene oligomerisation,2–6 and
recent research efforts have focussed on the development of
catalysts for selective production of co-monomer range olefins
(particularly 1-hexene and 1-octene).7–11
An attractive potential alternative to ethylene oligomerisa-
tion is the conversion of n-alkanes into LAOs. Established
industrial processes for alkane dehydrogenation to mono-
alkenes do exist, however generally lead to mixtures of pre-
dominately internal olefins.12 The conversion of internal olefins
into α-olefins, which is a contrathermodynamic process, is
therefore required in order to develop an overall n-alkane to
LAO conversion process. For this reason, we have recently13,14
been investigating the individual steps of the cycle shown in
Scheme 1 as a potential method for internal to α-olefin conver-
sion, which involves three steps. The first is insertion of an
internal olefin into a metal–hydride bond to yield a sec-alkyl-
metal compound. In the second step, which is actually a series
of elimination and re-insertion steps (chain walking), the
sec-alkyl group undergoes isomerisation to form the thermo-
Scheme 1 Proposed group 13-hydride based cycle for contrathermo-
dynamic isomerisation of olefins.
dynamically preferred primary alkyl. The final step involves
β-hydride elimination of an α-olefin and, if done under kinetic
control (rapid removal of the liberated olefin from the system),
may lead to α-olefin selectively.
For alkylboranes, the first two steps are reasonably
facile,15,16 but we were unable to effect elimination and iso-
lation of α-olefins on a preparative scale,13 despite reports to
the contrary.17,18 On the other hand, we showed that α-olefins
can be removed from tri-n-alkylaluminium with very high
selectivity (>97%), but in this case insertion and isomerisation
aSchool of Physical Sciences – Chemistry, Private Bag 75, Hobart 7001, Australia.
E-mail: david.mcguinness@utas.edu.au
bCSIRO Energy, 71 Normanby Rd, Clayton North 3169, Australia
†Electronic supplementary information (ESI) available: Additional supporting
NMR spectra and optimised geometries and energies of all stationary points. See
DOI: 10.1039/c5dt03257a
20098 | Dalton Trans., 2015, 44, 20098–20107
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
may be taken to ca. 80% conversion without decompo-
sition.14 As such, the aluminium hydride reagent employed
in this study has
a nominal composition of 70–80%
[Al(Oct)2H] and 20–30% [Al(Oct)3]. At this composition, and
in non-coordinating solvents (toluene, benzene), the actual
speciation of aluminium corresponds to complexes 1, 2
and 3 as the major species, with 3 the most abundant. If this
[Al(Oct)2H] is treated with 1-octene in the absence of a catalyst,
insertion proceeds to yield [Al(Oct)3] with a half-life of
approximately 30 minutes at 75 °C. Internal octenes (cis and
trans 2-, 3- and 4-octene) insert only partially, as described in
the introduction.14
Scheme 2 Metal-catalysed hydroalumination.
of internal olefins was unsuccessful.14 The reaction of
[Al(n-Alk)2H] with internal olefins reaches an equilibrium posi-
tion (ca. 50% insertion), and isomerisation to the more stable
primary alkyl is too slow to drive the reaction to completion.
We have therefore started investigating the use of transition
metal catalysts to promote the insertion and isomerisation
steps at aluminium hydrides. Metal catalysed hydroalumina-
tion of alkenes and alkynes (Scheme 2) has been widely
studied19–21 and can be considered as an insertion into a tran-
sition metal hydride (rapid) followed by transalkylation with
aluminium, although this is probably an over simplification in
many cases. While the majority of catalysed hydroalumina-
tions involve terminal olefins, a number of these catalysts are
also capable of olefin isomerisation,22–24 and some also
promote secondary to primary alkyl conversion on their own.
The classic example is Schwartz’s reagent, [Cp2ZrHCl].25,26
This complex reacts with internal olefins under mild con-
ditions to yield [Cp2ZrCl(n-alkyl)] complexes. In situ ana-
logues, prepared by reacting [Cp2ZrCl2] with alkylaluminium
reagents (normally in near stoichiometric ratios), are effective
catalysts for hydroalumination and carboalumination of
olefins and alkynes (but again, are normally employed for
α-olefin conversions).27,28 For this reason, we were interested
in the reactivity of internal olefins with the [Cp2ZrCl2]/alkyl-
aluminium system, ultimately with catalytic loadings of
zirconium, and report here our studies in this regard.
Furthermore, the effect such an isomerisation catalyst might
have on the selectivity of the elimination step was unknown,
and is investigated also.
When three equivalents of [Al(Oct)2H] are reacted with
1
[Cp2ZrCl2] in toluene-d8, the H NMR Cp signal at 5.89 ppm is
replaced with a signal at 5.65 ppm along with broad Zr–H reso-
nances at −1.06 (1H) and −2.20 ppm (2H). These signals are
closely analogous to those observed when [Cp2ZrCl2] is treated
with [AliBu2H],33 and the work of Bercaw and Brintzinger
suggests the time averaged structure 4 (Reaction (1), R = iBu).35
Integration of the signals observed in our work is likewise
indicative of structure 4 (R = n-octyl). This complex is stable in
solution for days. Unfortunately all attempts to crystallise it
were unsuccessful. The 3 : 1 Al : Zr stoichiometry is also sup-
ported by the observation that less than three equivalents of
[Al(Oct)2H] lead to formation of 4 only to the extent dictated by
this stoichiometry. Thus,
a 1 : 1 reaction of Al–H with
[Cp2ZrCl2] leads to three signals for the [Cp2Zr] moiety, of
which 30% is the signal for 4. With a 2 : 1 Al–H : Zr ratio, 66%
of the [Cp2Zr] signals correspond to 4 (1H NMR spectra at
different Al–H : Zr ratios are shown in the ESI, Fig. S1†). In
these sub-stoichiometric cases, the other Cp signals corres-
pond to unreacted [Cp2ZrCl2] and [Cp2ZrCl(Oct)] (5). The
latter complex gives rise to a cyclopentadienyl signal at
5.79 ppm along with a characteristic multiplet integrating for 2
protons at 1.08 ppm, corresponding to the Zr–CH2 resonance
(60.7 ppm in the 13C NMR spectrum).25,38 Complex 5 presum-
ably results from chloro/alkyl exchange between [Cp2ZrCl2]
and [Al(Oct)3] which is present in our system; others have
reported that AlR3 (R = Me, Et) reacts likewise to form mixtures
of [Cp2ZrClR] and Cp2ZrCl2.39,40
The interaction of [Cp2ZrCl2] with [AliBu3] and [AliBu2H],
and the hydroalumination of α-olefins with this system, has
been studied in depth by Parfenova, Dzhemilev and Khalilov
(PDK),27–32 and also Schwartz.33,34 Bercaw and Brintzinger
have further clarified the reactivity of zirconocenes with alkyl-
aluminium hydrides.35–37 We have drawn on the studies of
these groups to aid in understanding of some of the results
observed herein.
Results and discussion
ð1Þ
The reactions shown in Scheme 1 have been studied with tri-
n-octylaluminium [Al(Oct)3] and di-n-octylaluminium hydride
[Al(Oct)2H], and the elimination from/insertion of octenes to
these. [Al(Oct)2H] is prepared by thermal elimination (T =
The hydride signals of 4, which are broad at room temp-
170 °C, vacuum) of 1-octene from [Al(Oct)3], and the reaction erature, split at low temperature such that the expected35
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 20098–20107 | 20099

View Article Online
Paper
Dalton Transactions
coupling (J = 7 Hz) between the downfield triplet and tures are suggested only as possibilities based on the spectral
upfield doublet is partially resolved at −50 °C (Fig. 1). In similarities with previous work.
addition, a number of extra Zr–hydride signals are observed
at low temperature (−1.15, −1.92 and −2.38 ppm at −50 °C).
These indicate one or a number of additional Zr–hydride
complexes are present in addition to 4, which undergo
exchange/equilibrium processes with one another.‡ The PDK
group observed a signal at −1.94 ppm which was attributed
i
to complex 6 (R = Bu), and proposed two equal intensity
signals at −0.65 to −1.35 and −2.56 to −2.80 ppm are due
to complexes such as 7 and 8.28,29,33,41 These complexes
were observed in systems where 4 (R = iBu) was also
observed.
Olefin insertion
The existence of several species is also evidenced by mul-
tiple 1H NMR signals in the Cp region at low temperature
We first investigated the reaction of 1-octene with the
[Cp2ZrCl2]/[Al(Oct)2H] system (in toluene), which allows for
more straightforward characterisation of the products
obtained and also provides for comparisons with subsequent
internal olefin addition. When a 1 : 1 solution of [Cp2ZrCl2]
and [Al(Oct)2H] is reacted with one equivalent of 1-octene at
room temperature, insertion occurs rapidly and is essentially
complete in the time take to acquire a 1H NMR spectrum.
Insertion leads to the formation of [Cp2ZrCl(Oct)] (5) as the
major product (Reaction (2)). Along with this, a small amount
of [Cp2ZrCl2] (5%) remains along with 7% of complex 4 (by
integration of the Cp2Zr signals). Trace olefinic signals for
1-octene are also observed. The residual 1-octene and hydride
4 do not seem to result from incomplete progress of the reac-
tion, as the composition is unchanged after a further hour.
Rather, this appears to be the equilibrium position of the reac-
tion. This was confirmed by reacting [Cp2ZrCl2] with [Al(Oct)3],
which led to a similar composition of Cp2ZrCl2, 4, 5 and free
1-octene.
(Fig. S3†). Unfortunately these signals are broad and overlap,
so accurate integration of individual species and assessment
of possible Zr–H : Cp ratios (which might support formu-
lations such as 6–8) was not possible. The theoretical 1H
NMR shielding values for these structures, along with several
other possibilities, were calculated and this is discussed in
the ESI.† While the calculated 1H NMR shifts for complex 4
are in good agreement with experiment, the agreement
between experiment and theory for 6–8 is not compelling.
Thus, while the n-octyl substituted analogues of 6, 7 and 8
seem possible in our system, we cannot make this assign-
ment with a high level of confidence; these solution struc-
ð2Þ
The insertion reaction is significantly slower when an
excess of hydride is present in the system (catalytic zircono-
cene). Thus, a 1 : 6 solution of [Cp2ZrCl2] : [Al(Oct)2H] treated
with half an equivalent of 1-octene (relative to Al–hydride)
leads to only 10% insertion of the octene at room temperature
1
after 1 hour (as judged by H NMR monitoring). Heating the
solution to 60 °C leads to more complete insertion of 1-octene
(90% in 45 minutes), but it is clear that excess aluminium
hydride severely retards insertion. Similar behaviour has been
reported by the PDK group.27–31 They found that clusters such
as 4 are much less active for hydroalumination than less
encumbered Zr–hydrides (the most active reagent is thought to
be [Cp2ZrHCl]). At high ratios of Al–H/Zr complex 4 is the prin-
cipal species present, perhaps at the expense of more active
Zr–hydrides. This is also analogous to previous studies of cata-
lyzed carboalumination and chain growth, which show that
olefin insertion into the M–C bond is hindered by formation
Fig. 1 VT 1H NMR spectroscopy of the Zr–hydride region of
[Cp2ZrCl2] + 3 [Al(Oct)2H] in toluene-d8.
‡The presence of a single broadened Al–CH2 signal at room temperature, which
shows splitting at −50 °C (albeit broader still), shows that not only Al–Zr clusters
are in equilibrium, but also [Al(Oct)3] which is present in the system is involved
in alkyl exchange (see Fig. S2†).
20100 | Dalton Trans., 2015, 44, 20098–20107
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
of alkyl bridged M/Al dimers, which must break apart prior to 4.05 ppm.14 Hence, it seems likely that the Al–H signals at
olefin coordination and insertion.42–45
2.9–3.8 ppm result from one or more Al–Zr clusters. The com-
Unfortunately, this system was found to be even less active plexity of this system prevented us from identifying the species
for internal olefin insertion. When a thermodynamic mix of present, and attempts at crystallisation were unsuccessful.
octenes (1.5% 1-octene, 10.5% trans-4-, 33.4% cis/trans-3- and Only the presence of Zr–H and Al–H bonding can be con-
cis-4-, 40.5% trans-2- and 14.1% cis-2-) was reacted, no discern- firmed, most likely resulting from species in Zr–Al clusters.
ible insertion takes place at room temperature even with an
While it appears Zr/Al clusters are formed in THF, they
excess of [Cp2ZrCl2] present. Heating the solutions (40–80 °C) appear to be reactive with respect to olefin insertion. Thus,
leads to variable degrees of olefin conversion but this is despite the limited solubility of a 1 : 1 mixture of [Cp2ZrCl2]
accompanied by decomposition. The solutions turn dark and [Al(Oct)2H] in THF, insertion of 1-octene proceeds rapidly
purple in colour and numerous signals appear in the cyclo- (ca. 15 min to completion) according to Reaction (2). Complex
pentadienyl region as well as multiple resonances in the 5 is formed cleanly (along with a small amount of Cp2ZrCl2),
1
Zr–hydride region. The colour change observed is indicative of although it is only partially soluble in THF (the H NMR spec-
reduction of zirconium. A more detailed exploration of this trum is shown in the ESI, Fig. S8†).
behaviour is described below. These decomposed samples
Under similar conditions the mixture of internal octenes
proved intractable. Interestingly, with excess [Al(Oct)2H] requires 24 h to reach a high degree of insertion (87% at
1
present (Al–H/Zr = 6), decomposition of the solution was not 40 °C). Interestingly, the H NMR spectrum is identical to that
observed at elevated temperatures (60 °C, 24 h). This appears resulting from 1-octene insertion, with no evidence for the sec-
to be because zirconium is locked up in the stable cluster 4. alkyl groups expected from insertion of internal octenes. This
An undesirable consequence of this stability is very limited shows that isomerisation to the terminal position has occurred
octene insertion; ca. 16% over 24 h.
(Scheme 3). This was confirmed by quenching the solution
Given that the formation of stable Zr–Al hydride cluster(s) with D2O, which leads only to 1-d1-octane (13C NMR analysis,
seems at least partially responsible for low insertion activity,32 Fig. S9†). Hence, in THF the [Cp2ZrCl2]/[Al(Oct)2H] system
we investigated the effect of THF solvation on the reactivity of appears to lead to an in situ analogue of Schwartz’s reagent,
the system. The rationale here being that formation of more Cp2ZrHCl, with similar reactivity.
active mononuclear metal hydrides (with coordinated THF)
Encouraged by this result, we attempted internal olefin
may occur. As discussed below, this turned out not to be the insertion/isomerisation with excess [Al(Oct)2H]. As is the case
case, and Zr/Al clusters of some form appear to persist in THF. in toluene, the reaction is inhibited by excess Al–hydride, but
In the ESI† the reaction free energy of THF coordinating to in this case reaches higher conversions.
A mixture of
and breaking apart complex 4 is estimated, and it is found [Cp2ZrCl2], [Al(Oct)2H] and internal octenes (1 : 4 : 4, 25 mol%
that this reaction is not favourable. Nonetheless, THF does Zr) leads to 74% insertion of the olefins after 48 h at 40 °C.
appear to have an effect on the rate of olefin insertion. When a Continuing the reaction for a further 48 h did not improve the
solution of [Cp2ZrCl2] in THF is treated with 1 equivalent of extent of insertion, and it is clear that decomposition or
[Al(Oct)2H], the formation of a colourless precipitate is observed. further reactions of the zirconium reagent occur. The solution
Exchange of the zirconium chloro ligand with alkyl or hydride changes colour to dark purple and, along with the formation
groups is evidenced by the formation of [Al(Oct)2Cl(THF)] of 5, numerous unidentified additional cyclopentadienyl
(Al–CH2 at −0.05 ppm, along with a signal at −0.18 ppm corres- signals are observed in the 1H NMR spectrum (Fig. S10†).
ponding to [Al(Oct)3(THF)]). The area of the signals for the Additionally, the formation of n-octane appears to be associ-
Cp2Zr moiety relative to the alkylaluminium signals shows that ated with this reaction, and is detected in variable amounts by
only ca. 30% of the zirconocene remains in solution. The GC-MS analysis. The principal aluminium species present
major Zr complex is [Cp2ZrCl2], but this is accompanied by 5 after insertion is [Al(Oct)3(THF)], along with a smaller amount
additional Cp2Zr signals (Fig. S4†). There are also multiple of [Al(Oct)2Cl(THF)] due to halide exchange with Zr (Fig. S10†).
signals in the Zr–hydride region from −0.7 to −4.0 ppm; three Once again, there is no evidence for sec-octyl–metal species in
signals at room temperature resolving to 7 signals at −50 °C
(Fig. S5†). When [Cp2ZrCl2] is treated with 3 equivalents of
[Al(Oct)2H] in THF no visible precipitate is formed to begin with,
although over a number of days a small amount of precipitate
was observed. The 1H NMR spectrum of the clear solution,
while different to a 1 : 1 mixture, still indicates multiple
species. As shown in Fig. S6 and S7,† there are two main
signals for Cp2Zr along with a third smaller one, and multiple
Zr–H and Al–H resonances (the latter at 2.9–3.8 ppm, not brid-
ging to Zr), dependent again on temperature. It is known from
our previous work that homonuclear Al–hydride clusters such
as 2 and 3 do not persist in the presence of THF, instead
forming [Al(Oct)2H(THF)] with an Al–H resonance at Scheme 3 Insertion and isomerisation of internal octenes.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 20098–20107 | 20101

View Article Online
Paper
Dalton Transactions
the NMR spectra, and only 1-d1-octane is formed through ments, during olefin elimination, or both). This suggests that
quenching with D2O. Hence, all octene which has undergone decomposition occurs from a Zr–octyl species. The decompo-
insertion has isomerised to terminal product. It is concluded sition of zirconocene alkyl-hydrides or dialkyls, leading to
that the Zr species formed in situ is, at least initially, effective alkane formation and reduced zirconocene species, is quite
as a catalyst for insertion/isomerisation (Reaction (3)), but cata- well known. For example, dialkyl zirconocenes are known to be
lyst deactivation occurs along with olefin insertion.
a source of “Cp2Zr(II)”, which may be trapped by addition of
other ligands (Scheme 4a).46–50 Reductive elimination of
alkanes from [Cp2ZrHR] is also observed (Scheme 4b), again
leading to Zr(II) in the first instance.51,52 A number of groups
have reported alkane formation from Zr/Al alkyl/hydride clus-
ters, such as complex 8 (R = Me,41 Et29) or analogues of 4.35
While we have not observed [Cp2ZrH(Oct)] or [Cp2Zr(Oct)2] in
our system, the formation of transient small amounts of these
or related species seems quite possible, and may be respon-
sible for the decomposition observed.
The development of strong colour changes accompanying
decomposition is also common,35,53–55 as we have observed in
this study (red in toluene, purple in thf). Deep red or crimson
coloration appears to be characteristic of Zr(III) formation in
these reactions.48,51 Bercaw has characterised a variety of para-
magnetic Zr(III) complexes originating from reduction of con-
strained geometry zirconocenes.37 Perhaps more relevant to
the current work however are early reports from Schwartz that
[Cp2Zr(II)(PR3)2] complexes can form diamagnetic Zr(III) dimers
such as 9 with evolution of hydrogen (Reaction (4)), in which
the cyclopentadienyl ligand is non-innocent in the transform-
ation.51,56 Despite the complexity of the 1H NMR spectra
observed after decomposition (Fig. S11†), the CpH signals
remain sharp and we have not seen any evidence for para-
magnetism, thus it appears likely that all species formed are
ð3Þ
Olefin elimination
We also investigated olefin elimination in the presence of the
Zr catalyst. Selective formation of α-olefin relies on the slow
rate of isomerisation at aluminium, coupled with the rapid
removal of olefin from the system. The latter requirement is
illustrated by the fact that partial isomerisation to internal
olefins is observed if product removal is inefficient.14 With the
Zr catalyst in the system, whose role is isomerisation in the
first steps, the selective removal of α-olefin may be affected.
Indeed, this is the case at the high Zr loadings required to
promote insertion. As illustrated in Table 1 (entry 2), when
elimination of octene from [Al(Oct)3]/[Al(Oct)2Cl] is carried out
in the presence of Zr (90 °C, vacuum, Al/Zr = 3.1), the mixture
of octenes collected is substantially isomerised to internal
octenes. For comparison, elimination from [Al(Oct)3] without
Zr present yields up to 98% 1-octene (entry 3).14 The major
internal octene isomers formed are cis- and trans-2-octene, in
approximately equal amounts. These two isomers are the first
products of chain walking from the primary position.
Increased loadings of [Cp2ZrCl2] lead to more extensive iso-
merisation (less 1-octene) and higher percentages of deeper
internal olefins (3- and 4-octenes). The selectivity of elimin-
ation is not improved by conducting it at higher temperatures.
In all cases, elimination causes the reaction mixture to turn
deep red in colour, and the formation of n-octane is again
observed in varying amounts (up to ca. 1 equivalent relative to
octene, Table 1, entry 2).
As decomposition of the Zr reagent seems to be prevalent
during insertion and elimination, some comment on the poss-
ible decomposition pathways is appropriate. In the presence of
excess Al–hydride, Zr appears to be quite stable, seemingly due
to the formation of stable hydride bridged clusters such
as 4. On the other hand, once insertion has occurred and
M–hydride has been replaced by M–octyl, the decomposition
reaction is observed (either during prolonged insertion experi-
Scheme 4 Alkane elimination routes from Zr(IV) complexes.
Table 1 Octene elimination from [Al(Oct)3]/[Al(Oct)2Cl] in the presence of zirconocene catalysta
Entry
1-Octene
trans-4
cis-2
trans-2
Otherb
1. Reactant compositionc
2. Eliminated compositiond
3. [Al(Oct)3]e
1.5
37.8
97.7
10.5
3.3
14.1
24.4
40.5
20.0
33.4
14.4
2.3
a Conditions: Al/Zr = 3.1, 90 °C, vacuum. b cis- and trans-3 and cis-4-octene. c Composition of internal octene mixture prior to insertion and
isomerisation. d 1.3 equivalents of n-octane (relative to octenes) also collected. e Elimination from [Al(Oct)3] in the absence of Zr catalyst (170 °C),
from ref. 14.
20102 | Dalton Trans., 2015, 44, 20098–20107
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
1
diamagnetic. Schwartz reported H NMR signals for the η5,η1- to be unfavourable (ΔGreact = 88 kJ mol−1, red trace in Fig. 2).
cyclopentadienyl ligands in 9 between 6.05–3.97 ppm. While Simple dimerisation reactions of the [Cp2Zr(II)] fragment with
undoubtedly more complex in our reactions, the observation Zr(IV) species were found to be downhill, with the most favour-
of signals between ca. 6.5–3.5 ppm (in toluene, Fig. S11†) able reaction being that with [Cp2ZrPrCl] (a model for the
might suggest multiple species containing this same binding experimentally observed complex 5). This leads to a closed
mode of cyclopentadienyl. Certainly it is difficult to envisage shell singlet (Zr(III)⋯Zr(III)) with bridging propyl and chloro
the cyclopentadienyl ligand in a normal η5-binding mode ligands (complex 10, the triplet and antiferromagnetically
leading to signals over this broad range.
coupled species (open shell singlet) were also examined and
are predicted to be higher in energy, see the ESI† for a full dis-
cussion). The stabilisation afforded by dimer formation is
sufficient to make reductive elimination favourable, for
ð4Þ
instance the overall free energy of Scheme 5(a) is ΔGreact
=
−33 kJ mol−1. A similar dimerisation of the metallacyclo-
propane complex in Fig. 2, with release of propene, is calcu-
lated to be moderately endergonic (Scheme 5(b), ΔGreact
=
The reduction of zirconium(IV) alkyls according to
Scheme 4, followed by pathways to Zr(III) formation, has been
probed theoretically (employing a propyl group as a model for
the n-alkyl chain, see Theoretical Methods section for details).
The calculated free energy profiles corresponding to pathways
(a) and (b) in Scheme 4 are shown in Fig. 2 (free energy esti-
mates have been calculated at 298 K, and hence may be some-
what different under experimental conditions of elevated
temperature). Both routes have accessible activation barriers,
with the β-hydride transfer barrier (blue trace) a little lower
and leading to what is best described as a Zr(IV) metallacyclo-
propane complex (we could not find a local minimum corres-
ponding to the π-bound propene extreme of bonding).
Removal of propene from this species to give [Cp2Zr] and
propene is highly endergonic (ΔG = 134 kJ mol−1), but as dis-
cussed below, there are a number of further reactions of the
[Cp2Zr] fragment which could overcome this. Likewise, the
reductive elimination of propane from [Cp2ZrHPr] is calculated
13 kJ mol−1). The formation of such diamagnetic Zr(III) dimers
may be one possible route of decomposition.
The formation of η5,η1-cyclopentadienyl complexes is
shown in Fig. 3 and appears to correspond closely to that pro-
posed by Schwartz,56 with some additional subtleties which
only become apparent from theoretical studies. The reaction
starts with dimerisation of two [Cp2Zr(II)] fragments and
appears to involve a number of spin state crossings (singlet to
triplet and vice versa) along the path. As is evident in Fig. 3,
and discussed in the ESI,† the relative spin state energies of
several complexes are quite close, so there is some uncertainty
in the ground states of these. The free energy surface shown in
Fig. 3 reflects that calculated with the M06 density functional.
The effect of different functionals, and the applicability of DFT
for such problems, is discussed in the ESI.†
Oxidative addition of the first cyclopentadienyl C–H bond
can occur in the singlet or triplet state, with the latter provid-
ing the easiest pathway (11 → 12 via TS11–12). Oxidative
addition of the second C–H bond (12 → 13) was only located
in the singlet spin state, and attempts to optimise the di-
hydride 13 as a triplet led back to complex 12 in all instances.
The binuclear Zr(IV)–Zr(IV) dihydride 13 is the most stable
species along the free energy surface. Reductive elimination of
hydrogen from singlet binuclear complexes was found not to
Fig. 2 Relative free energies (M06/BS2//M06/BS1, kJ mol−1, 298 K) for
species involved in the reductive elimination of propane from
[Cp2ZrPrH] (red) and [Cp2ZrPr2] (blue).
Scheme 5 Calculated energetics of alkane elimination and dimer
formation.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 20098–20107 | 20103

View Article Online
Paper
Dalton Transactions
Fig. 3 Relative free energies (M06/BS2//M06/BS1, kJ mol−1, 298 K) for the reaction [Cp2ZrPrH] → 1/2 [CpZr(μ-η1,η5-C5H4]2 (17) + 1/2 H2. Energies
shown are per mole of Zr (per 1/2 mole of dimers). Stationary points shown in blue are in the singlet spin state, structures in red triplets, and those
in black are calculated in both spin states. The approximate MECP energy levels indicated are derived from M06/BS1 electronic energies.
be favourable, with the easiest route to a dihydrogen complex while Negishi has observed reduction from heterobinuclear
being a spin state crossing from complex 14 to triplet 15, (Zr–Al) complexes.58 Further studies of reduction mechanisms
where the transition from a dihydride to a dihydrogen from alkylzirconium(IV) cyclopentadienyl complexes are in progress.
complex corresponds to the minimum energy crossing point
(MECP).57 From 15, the barrier to loss of dihydrogen is very
low, leading to the η5,η1-cyclopentadienyl dimer 17, which is
calculated to be most stable as a closed shell singlet.
Summary and conclusion
The reaction surfaces shown in Fig. 2 and 3 are probably an Herein we have studied the use of zirconocene dichloride to
over simplification of the real range of reaction possibilities, promote insertion of internal olefins (hydroalumination) and
particularly given the complexity of the 1H NMR spectrum chain walking isomerisation at [Al(Oct)2H]. The reaction
after decomposition. However they do illustrate that alkane between [Cp2ZrCl2] and [Al(Oct)2H] in non-polar solvent leads
elimination from zirconium(IV) alkyls may be facile, and fol- preferentially to clusters such as 4,35 with bridging hydride
lowed by activation of the cyclopentadienyl ligand, can lead to ligands, plus a number of other unidentified clusters in equili-
diamagnetic Zr(IV) and Zr(III) binuclear complexes featuring brium. In toluene the precatalyst [Cp2ZrCl2] promotes hydro-
η5,η1-Cp ligation. The presence of alkylaluminium reagents alumination of 1-octene, but is largely ineffective for internal
may of course increase the complexity through formation of octenes. One reason seems to be that the cluster 4 is relatively
Zr/Al clusters, and Schwartz56 has also suggested that com- inert towards insertion,28 hence its formation, which is
plexes such as 13 may exist as nonuniform polymers, which we favoured at high Al/Zr ratios, limits the use of the Zr reagent in
have not considered at this stage. There are also a number of catalytic amounts.
other reduction mechanism possibilities. Bercaw has
Somewhat more success was achieved in tetrahydrofuran.
suggested reduction from homobinuclear Zr(IV) complexes,37 While it seems Zr/Al hydride clusters are still formed to an
20104 | Dalton Trans., 2015, 44, 20098–20107
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
extent, it was possible to catalyze the insertion and isomerisa- was added to an NMR tube and dissolved in ca. 0.40 mL of
tion of internal olefins to primary alkyls (albeit with very high toluene-d8. To this solution [Al(Oct)2H] (75 μL, 150 μmol
catalyst loadings, ca. 25% Zr). However, this transformation is hydride, 66 μmol [Al(Oct)3]) was added. The solution became
accompanied by catalyst decomposition/deactivation and pink in colour. 1H NMR (toluene-d8): δ 5.69 (s, 10H, CpH); 1.64
alkane formation. Furthermore, the presence of the isomerisa- (m, 12H, CH2); 1.25–1.55 (br, m, 60H, CH2); 0.92 (t, J = 7 Hz,
tion catalyst interferes with the final step of the cycle, leading 18H, CH3); 0.38 (t, J = 8 Hz, 12H, AlCH2); −0.98 (br s → t, J = 7
to back-isomerisation during olefin elimination (whereas this Hz, 1H, Zr–H); −2.11 (br s → d, J = 7 Hz, 2H, Zr–H). 13C NMR
step is highly α-selective in the absence of Zr14). The elimin- (toluene-d8): δ 104.7 (Cp); 36.2, 32.5, 30.1, 30.0, 26.0, 23.2
ation step is also accompanied by further catalyst decompo- (CH2); 14.4 (CH3); 10.5 (br, Al–CH2). The 1H and 13C NMR
sition. Preliminary observations and theoretical studies spectra are shown in Fig. S13 and S14.†
suggest that decomposition occurs via reduction of Zr and
Preparation and characterisation of [Cp2ZrCl(Oct)] (5). This
alkane elimination. From here, binuclear Zr(III) and Zr(IV) com- complex can be prepared cleanly in situ from a 1 : 1 : 1 mixture
plexes may form. We are currently studying the decomposition [Cp2ZrCl2] : [Al(Oct)2H] : 1-octene in toluene or thf. Thus,
pathways of Zr(IV)–alkyls in more detail, and investigating alter- [Cp2ZrCl2] (15.4 mg, 53 μmol) was taken up in 0.5 mL of
nate catalysts which may be more compatible with the cycle toluene-d8 in an NMR tube and [Al(Oct)2H] (25 μL, 50 μmol
shown in Scheme 1.
hydride, 22 μmol [Al(Oct)3]) and 1-octene (8.0 μL, 51 μmol)
added. H NMR (toluene-d8): δ 5.79 (s, 10H, CpH); 1.2–1.7 (m,
1
br, CH2, overlapping with Al–octyl signals); 1.08 (m, 2H,
ZrCH2); 0.90 (m, CH3 overlapping with Al–octyl signals). 13C
NMR (toluene-d8): δ 113.0 (Cp); 60.7 (ZrCH2); 36.4 (γ-CH2); 34.6
(β-CH2), 32.6, 29.8, 23.2 (CH2); the CH2 signal at 30.0 ppm and
Experimental and theoretical methods
General
All manipulations were performed under an argon atmosphere the CH3 at 14.4 ppm are masked by the corresponding Al–octyl
using standard Schlenk techniques or carried out in a glove- signals. Complex 5, displaying the same spectral properties,
box. Solvents were purified by passage through an Innovative can be independently prepared by reacting Cp2ZrHCl with
Technologies solvent purification system and, where appropri- 1-octene, as reported by Schwartz.25 The 13C NMR spectrum in
ate, were stored over a sodium mirror. 1-Octene was dried by toluene-d8 is shown in Fig. S15–S17.† In thf, several of the CH2
1
distillation over sodium and stored over a sodium mirror. signals in the H NMR spectrum are better resolved from the
[Al(Oct)3] and [Al(Oct)2H] (70–80% hydride, 20–30% [Al(Oct)3], signals due to Al–octyl species, such that the β- and γ-methyl-
quantified by 1H NMR) were prepared as described pre- ene signals can be assigned in addition to the α-resonance; 1H
viously.14 Elimination of octene from [Al(Oct)3] in the presence NMR (thf-d8): δ 6.21 (s, 10H, CpH); 1.54 (m, 2H, β-CH2);
of Zr catalyst was carried out according to the procedure 1.25–1.50 (br, CH2 Zr–octyl and Al–octyl); 1.16 (m, 2H, γ-CH2);
detailed previously,14 but at lower temperature (refer to 0.99 (m, 2H, ZrCH2); 0.88 (m, CH3 Zr–octyl and Al–octyl).
Table 1). NMR measurements were recorded on a Bruker Assignment of the α, β and γ 1H and 13C NMR signals was
Avance III HD NMR spectrometer with a 5 mm BBFO probe based on COSY correlations between the α and β hydrogens
operating at 400 MHz (1H) or 100 MHz (13C). GC–MS analysis and between β with both α and γ, and with HSQC correlations
was carried out on a Varian 3800 GC coupled to a Varian 1200 for the 13C assignments.
triple quadrupole mass spectrometer in single quadrupole
mode.
Olefin insertion. A representative example corresponds to
attempted insertion of a mixture of octene isomers into
Octylaluminium NMR assignments. In solution with both [Al(Oct)2H] with 25 mol% [Cp2ZrCl2] as catalyst: inside the
Al–octyl and Zr–Octyl species present, the NMR assign- glove box, [Cp2ZrCl2] (18.1 mg, 62 μmol) was dissolved in
ments for [Al(Oct)3(thf)], [Al(Oct)2Cl(thf)] (in thf-d8) and 0.5 mL THF-d8 in a J-Young valve NMR tube. [Al(Oct)2H] (80%
[Al(Oct)3−nCln]2 (n ≈ 1, in toluene or benzene) were identified hydride, 20% [Al(Oct)3], 100.0 μL, 0.257 mmol of hydride) was
by comparison to independent samples. In the last case, the added followed by the addition of the internal octene mixture
approximate composition expected from complete insertion of (40.0 μL, 0.255 mmol). The solution became light pink over
1-octene into a 1 : 1 mixture of [Cp2ZrCl2] : [Al(Oct)2H] was pre- time. The NMR tube was taken out of the glove box and placed
pared by reacting [Al(Oct)3] with [AlCl3] at a 3 : 1 ratio in in a 40 °C water bath. Within 1 h, the solution became dark
1
toluene. The most characteristic 1H NMR signal distinguishing purple. Heating was continued for 48 h. Analysis by H NMR
the different species is the Al–CH2 multiplet, which resonates indicated that 74% of the octene isomers inserted. Continued
at: [Al(Oct)3]2, 0.56 ppm (toluene-d8); [Al(Oct)3−nCln]2 (n ≈ 1), heating up to 96 h did not improve the extent of reaction
0.44 ppm (toluene-d8); [Al(Oct)3(thf)], −0.16 ppm (thf-d8); (corresponding to Fig. S10†).
[Al(Oct)2Cl(thf)], −0.05 ppm (thf-d8). The 13C NMR assignments
D2O quench. Following internal olefin insertion attempts,
for [Cp2ZrCl(Oct)] (5), [Al(Oct)2Cl(thf)] and [Al(Oct)3(thf)] in a the contents of the NMR tube were placed under vacuum and
mixed system (after 1-octene has inserted) are shown in the un-inserted octenes and solvent removed. The resulting
Fig. S12† (full assignments for 5 are detailed below).
dark purple residue was dissolved in C6D6 and quenched with
Preparation and characterisation of [Cp2ZrH(μ-H)2Al3(Oct)6- D2O. The organics were extracted into CDCl3 and analyzed by
(μ-Cl)2] (4). Inside a glove box, Cp2ZrCl2 (15.3 mg, 52 μmol) 13C NMR spectroscopy (Fig. S9†).
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 20098–20107 | 20105

View Article Online
Paper
Dalton Transactions
2 P.-A. R. Breuil, L. Magna and H. Olivier-Bourbigou, Catal.
Lett., 2015, 145, 173.
Theoretical methods
Theoretical calculations throughout this paper were carried
out with the Gaussian0959 program. Geometry optimisations
were performed without symmetry constraints using the
M0660–62 functional in combination with the Stuttgart-
Dresden63 (SDD) double-ζ valence basis set and effective core
potential (ECP) for Zr and the 6-31G(d) basis set for all other
atoms (referred to as BS1). Analytical frequency calculations
were carried out to verify structural optimisations and to
obtain free energy corrections at the same level of theory
(298 K and 1 atm standard state, free energy estimates will
differ somewhat in solution and at elevated temperature). All
3 A. Forestière, H. Olivier-Bourbigou and L. Saussine, Oil Gas
Sci. Technol., 2009, 64, 649.
4 H. Olivier-Bourbigou, A. Forestière, L. Saussine, L. Magna,
F. Favre and F. Hugues, Oil Gas, 2010, 36, 97.
5 G. P. Belov and P. E. Matkovsky, Pet. Chem., 2010, 50,
283.
6 W. Keim, Angew. Chem., Int. Ed., 2013, 52, 12492.
7 J. T. Dixon, M. J. Green, F. M. Hess and D. H. Morgan,
J. Organomet. Chem., 2004, 689, 3641.
8 D. S. McGuinness, Chem. Rev., 2011, 111, 2321.
9 T. Agapie, Coord. Chem. Rev., 2011, 255, 861.
minima contained no imaginary frequencies, and all tran- 10 P. W. N. M. van Leeuwen, N. D. Clément and
sition structures contained only one imaginary frequency that M. J.-L. Tschan, Coord. Chem. Rev., 2011, 255, 1499.
exhibited vibrational modes consistent with the anticipated 11 D. F. Wass, Dalton Trans., 2007, 816.
reaction pathway (further verified by optimisation of slightly 12 P. R. Pujado, in Handbook of Petroleum Refining Processes,
relaxed transition structures in each direction and IRC calcu-
lations). Single point energies for all structures were calculated 13 N. M. Weliange, D. S. McGuinness and J. Patel, Organo-
with the M06 functional with the quadruple-ζ valence def2- metallics, 2014, 33, 4251.
QZVP64,65 basis set and SDD ECP on Zr, and the 6-311+G(2d,p) 14 N. M. Weliange, D. S. McGuinness, M. G. Gardiner and
basis set on all other atoms (referred to as BS2). The Gibbs free J. Patel, Dalton Trans., 2015, 44, 15286.
energy (ΔG) values reported in this paper were obtained at the 15 H. C. Brown, Organic syntheses via boranes, Wiley,
M06/BS2//M06/BS1 level. New York, 1975.
Monometallic Zr(IV) and Zr(II) complexes were calculated as 16 H. C. Brown and B. C. Subba Rao, J. Org. Chem., 1957, 22,
singlets, as is found experimentally. For bimetallic Zr species, 1137.
the closed shell singlet, triplet and open shell singlet (anti- 17 A. de Klerk, S. W. Hadebe, J. R. Govender, D. Jaganyi,
ed. R. A. Meyers, McGraw Hill, 3rd edn, 2004, p. 5.11.
ferromagnetically coupled Zr) were considered. It was found
that the triplet and antiferromagnetically coupled singlet com-
A. B. Mzinyati, R. S. Robinson and N. Xaba, Ind. Eng. Chem.
Res., 2007, 46, 400.
plexes had very similar geometries and energies (for complexes 18 A. B. Mzinyati and D. Jaganyi, Ind. Eng. Chem. Res., 2009,
10 and 11 these are included in the ESI†). In Fig. 3, all singlet 48, 2770.
stationary points refer to the closed shell singlet state. 19 U. M. Dzhemilev and A. G. Ibragimov, Russ. Chem. Rev.,
Minimum energy crossing points (MECPs) between the two 2000, 69, 121.
surfaces were located using the methodology of Harvey and co- 20 U. M. Dzhemilev and V. A. D’yakonov, Top. Organomet.
workers.57 The barriers for spin state crossing shown in Fig. 3
Chem., 2013, 41, 215.
are estimated from electronic energies (Eelec) at the M06/BS1 21 L. V. Parfenova, L. M. Khalilov and U. M. Dzhemilev, Russ.
level.
Chem. Rev., 2012, 81, 524.
22 F. Asinger, B. Fell, R. Janssen, G. Zoche, E. W. Mueller and
F. W. A. G. K. Korte, US 3322806 (Shell Oil), 1967.
23 E. W. Muller, G. Zoche and F. W. A. G. K. Korte, US
3475477 (Shell Oil), 1969.
Acknowledgements
We thank the Australian Research Council for financial
support through grant DP1211558 and the Australian National
Computational Infrastructure (NCI) and Tasmanian Partner-
ship for Advanced Computing (TPAC) for provision of comput-
ing resources. Dr Bun Chan is thanked for useful advice. Dr
James Horne and A/Prof. Noel Davies of the UTas Central
Science Laboratory are thanked for assistance with NMR and
GC-MS measurements, respectively.
24 E. J. M. de Boer, H. Van der Heijden, I. Oosterveld and
A. Van Zon, WO 2006/005762 (Shell), 2006.
25 D. W. Hart and J. Schwartz, J. Am. Chem. Soc., 1974, 96,
8115.
26 J. Schwartz and J. A. Labinger, Angew. Chem., Int. Ed. Engl.,
1976, 15, 333.
27 L. V. Parfenova, A. V. Balaev, I. M. Gubaidullin,
L. R. Abzalilova, S. V. Pechatkina, L. M. Khalilov,
S. I. Spivak and U. M. Dzhemilev, Int. J. Chem. Kinet., 2007,
39, 333.
28 L. V. Parfenova, S. V. Pechatkina, L. M. Khalilov and
U. M. Dzhemilev, Russ. Chem. Bull. Int. Ed., 2005, 54, 316.
Notes and references
1 D. Vogt, in Applied Homogenous Catalysis with Organometal- 29 L. V. Parfenova, R. F. Vil’danova, S. V. Pechatkina,
lic Compounds, ed. B. Cornils and W. A. Herrmann, VCH,
Weinheim, Germany, 1996, vol. 1, p. 245.
L. M. Khalilov and U. M. Dzhemilev, J. Organomet. Chem.,
2007, 692, 3424.
20106 | Dalton Trans., 2015, 44, 20098–20107
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
30 E. Y. Pankratyev, T. V. Tyumkina, L. V. Parfenova, 51 K. I. Gell and J. Schwartz, J. Chem. Soc., Chem. Commun.,
L. M. Khalilov, S. L. Khursan and U. M. Dzhemilev, Organo-
metallics, 2009, 28, 968.
31 E. Y. Pankratyev, T. V. Tyumkina, L. V. Parfenova,
1979, 244.
52 D. R. McAlister, D. K. Erwin and J. E. Bercaw, J. Am. Chem.
Soc., 1978, 100, 5966.
S. L. Khursan, L. M. Khalilov and U. M. Dzhemilev, Organo- 53 P. C. Wailes, R. S. P. Coutts and H. Weigold, Organometallic
metallics, 2011, 30, 6078.
Chemistry of Titanium, Zirconium and Hafnium, Academic
32 L. V. Parfenova, P. V. Kovyazin, I. E. Nifant’ev,
Press, New York and London, 1974.
L. M. Khalilov and U. M. Dzhemilev, Organometallics, 2015, 54 T. Gibson and L. Tulich, J. Org. Chem., 1981, 46, 1821.
34, 3559.
55 K. I. Gell, B. Posin, J. Schwartz and G. M. Williams, J. Am.
Chem. Soc., 1982, 104, 1846.
56 K. I. Gell, T. V. Harris and J. Schwartz, Inorg. Chem., 1981,
20, 481.
57 J. N. Harvey, M. Aschi, H. Schwarz and W. Koch, Theor.
Chem. Acc., 1998, 99, 95.
58 E.-i. Negishi, D. Y. Kondakov, D. Choueiry, K. Kasai and
T. Takahashi, J. Am. Chem. Soc., 1996, 118, 9577.
33 L. I. Shoer, K. I. Gell and J. Schwartz, J. Organomet. Chem.,
1977, 136, C19.
34 D. B. Carr and J. Schwartz, J. Am. Chem. Soc., 1979, 101,
3521.
35 S. M. Baldwin, J. E. Bercaw and H. H. Brintzinger, J. Am.
Chem. Soc., 2008, 130, 17423.
36 S. M. Baldwin, J. E. Bercaw and H. H. Brintzinger, J. Am.
Chem. Soc., 2010, 132, 13969.
59 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand,
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaus-
sian 09, Revision D01, Gaussian, Inc., Wallingford CT, 2009.
37 T. N. Lenton, J. E. Bercaw, V. N. Panchenko, V. A. Zakharov,
D. E. Babushkin, I. E. Soshnikov, E. P. Talsi and
H. H. Brintzinger, J. Am. Chem. Soc., 2013, 135, 10710.
38 E.-i. Negishi and T. Yoshida, Tetrahedron Lett., 1980, 21,
1501.
39 D. E. Van Horn and E.-i. Negishi, J. Am. Chem. Soc., 1978,
100, 2252.
40 S. Beck and H. H. Brintzinger, Inorg. Chim. Acta, 1998, 270,
376.
41 P. C. Wailes, H. Weigold and A. P. Bell, J. Organomet.
Chem., 1972, 43, C29.
42 R. A. Petros and J. R. Norton, Organometallics, 2004, 23,
5105.
43 J. M. Camara, R. A. Petros and J. R. Norton, J. Am. Chem.
Soc., 2011, 133, 5263.
44 M. Bochmann and S. J. Lancaster, Angew. Chem., Int. Ed.
Engl., 1994, 33, 1634.
45 M. Bochmann and S. J. Lancaster, J. Organomet. Chem., 60 Y. Zhao, N. E. Schultz and D. G. Truhlar, J. Chem. Theory
1995, 497, 55. Comput., 2006, 2, 364.
46 E.-i. Negishi, S. J. Holmes, J. M. Tour, J. A. Miller, 61 Y. Zhao and D. G. Truhlar, J. Chem. Phys., 2006, 125,
F. E. Cederbaum, D. R. Swanson and T. Takahashi, J. Am.
Chem. Soc., 1989, 111, 3336.
47 E.-i. Negishi, F. E. Cederbaum and T. Takahashi, Tetra-
hedron Lett., 1986, 27, 2829.
194101.
62 Y. Zhao and D. G. Truhlar, J. Phys. Chem. A, 2006, 110,
13126.
63 D. Andrae, U. Häussermann, M. Dolg, H. Stoll and
H. Preuss, Theor. Chim. Acta, 1990, 77, 123.
48 T. Cuenca and P. Royo, J. Organomet. Chem., 1985, 295, 159.
49 S. L. Buchwald, B. T. Watson and J. C. Huffman, J. Am. 64 F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005,
Chem. Soc., 1987, 109, 2544. 7, 3297.
50 S. Beck and H. H. Brintzinger, Inorg. Chim. Acta, 1998, 270, 65 F. Weigend, F. Furche and R. Ahlrichs, J. Chem. Phys., 2003,
376.
119, 12753.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 20098–20107 | 20107