Investigations on the catalytic carboxylation of olefins with CO2 towards α,β-unsaturated carboxylic acid salts: characterization of intermediates and ligands as well as substrate effects

Article abstract of DOI:10.1039/c5dt01040c

The carboxylation of olefins beyond ethylene towards α,β-unsaturated carboxylic acid salts and a detailed investigation on the critical steps in the catalysis are reported. The influence of two chelating phosphine ligands and several olefins on the elemental steps of the catalysis is shown. The work focusses on the formation of intermediate olefin complexes, lactone formation and base induced elimination of the lactone. The direct carboxylation of olefins is possible using nickel catalysts, which opens a new route towards the desired α,β-unsaturated carboxylic acid salts. The reaction works particularly well for 1,3-dienes and proceeds via the formation of allyl-carboxylates. The ability to form such allyl-type lactone complexes seems in this case to be the most challenging step towards satisfactory turnover numbers.

Full text of DOI:10.1039/c5dt01040c

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Investigations on the catalytic carboxylation of
olefins with CO2 towards α,β-unsaturated
carboxylic acid salts: characterization of
intermediates and ligands as well as substrate
effects†
Cite this: Dalton Trans., 2015, 44,
11083
Ivana Jevtovikj,a Simone Manzini,a Matthias Hanauer,b Frank Romingerc and
Thomas Schaub*a,d
The carboxylation of olefins beyond ethylene towards α,β-unsaturated carboxylic acid salts and a detailed
investigation on the critical steps in the catalysis are reported. The influence of two chelating phosphine
ligands and several olefins on the elemental steps of the catalysis is shown. The work focusses on the for-
mation of intermediate olefin complexes, lactone formation and base induced elimination of the lactone.
The direct carboxylation of olefins is possible using nickel catalysts, which opens a new route towards the
desired α,β-unsaturated carboxylic acid salts. The reaction works particularly well for 1,3-dienes and pro-
ceeds via the formation of allyl-carboxylates. The ability to form such allyl-type lactone complexes seems
in this case to be the most challenging step towards satisfactory turnover numbers.
Received 16th March 2015,
Accepted 3rd May 2015
DOI: 10.1039/c5dt01040c
www.rsc.org/dalton
be obtained in many cases due to reduction of the double
bond,7 isomerization8 or telomerisation.9
Introduction
The carboxylation of olefins with CO2 in the presence of bases
Despite this early work, by the catalytical method, only telo-
to the corresponding α,β-unsaturated carboxylic acid salts pro- merisation reactions,10 the use of Zn-alkyls11 as reductants or
vides atom efficient access to this class of compounds. The the reductive carboxylation of allyl esters12 were reported,
abundant and cheap CO2 can be used in this case as the whereby α,β-saturated carboxylic acid salts are formed. The
C1-building block.1 This carboxylation to the desired acids or catalytic synthesis of the desired α,β-unsaturated carboxylic
salts respectively, mediated by several metals, has been investi- acids based on olefins and CO2 in the presence of a base has
gated over the last few decades, based on the early results of been achieved just recently by the appropriate choice of bases
Hoberg et al.1b The transformations were thereby mainly and ligands. The first catalytic carboxylation of ethylene to
limited to stoichiometric reactions with respect to the metal sodium acrylate was described by our laboratory using a two-
complexes to obtain α,β-unsaturated carboxylic acids:2 using stage process with Ni-dtbpe (dtbpe = 1,2-bis(di-tert-butylphos-
nickel(0), CO2 and olefins followed by acidic workup or alkyl- phino)ethane) as the catalyst and NaOtBu as the base.13 The
ation of cyclopentene,3 styrene,4 piperylene,5 or ethylene.6 two-stage mode was necessary because the lactone-formation
Under similar conditions, α,β-unsaturated carboxylic could not and the acrylate-formation have to be separated in two steps,
but is, in the overall reaction, catalytic towards the nickel.
When the steps were repeated, roughly 10 turn-overs could be
achieved by this method. Vogt et al. were able to achieve the
catalytic formation of lithium acrylates in a single step cataly-
sis by the use of LiI, an amine base, Zn as a reductant and a
Ni-dcpp (dcpp = 1,3-bis(di-cyclohexylphosphino)propane) cata-
lyst with a TON up to 21, but the reaction was not reported for
other olefins (Scheme 1).14
By further variation of the base (to avoid carbonate for-
mation) and change of the ligands, we were also able to
achieve the formation of sodium acrylate in a single step
aCaRLa (Catalysis Research Laboratory), Im Neuenheimer Feld 584, D-69120
Heidelberg, Germany. E-mail: thomas.schaub@basf.com; Fax: (+)49 621 60-6648957
bBASF SE, Quantum Chemistry, Carl-Bosch-Strasse 38, D-67056 Ludwigshafen,
Germany
cRuprecht-Karls-Universität Heidelberg, Organisch-Chemisches Institut,
Im Neuenheimer Feld 270, D-69120 Heidelberg, Germany
dBASF SE, Synthesis and Homogeneous Catalysis, Carl-Bosch-Strasse 38,
D-67056 Ludwigshafen, Germany
†Electronic supplementary information (ESI) available: Computational details
on the isomers of complex 3c. CCDC 1052343–1052346. For the ESI and crystal-
lographic data in CIF or other electronic format see DOI: 10.1039/c5dt01040c
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 11083–11094 | 11083

View Article Online
Paper
Dalton Transactions
Scheme 1 Catalytic carboxylation of ethylene to acrylates with CO2.
catalysis. In this system, Ni-BenzP* (BenzP* = (R,R)-1,2-bis(tert-
butyl-methyl-phosphino)benzene) as the catalyst and sodium-
2-fluorophenolate as the base together with Zn were used to
achieve a TON up to 107.15 It was also possible to apply the
reaction for the first time towards the synthesis of a series of
α,β-unsaturated carboxylic acid salts via the catalytic carboxyl-
ation with CO2 of olefins like styrenes, butadiene, piperylene,
isoprene or cyclohexadiene. For a systematic improvement of
the reaction, it is essential to understand the critical steps of
the catalysis and we have investigated the carboxylation of
olefins beyond ethylene in more detail. Herein, we report the
influence of two chelating phosphines and several olefins on
the elemental steps of the catalysis, mainly the formation of
intermediate olefin complexes, lactone formation and the base
induced elimination of the lactone.
Scheme 2 Main species in the catalytic cycle on the carboxylation of
olefins with CO2 towards α,β-unsaturated carboxylic acid salts.
Scheme 3 Synthesis of olefin complexes with the phosphine ligands
dtbpe and dcpe.
Results and discussion
According to our previous work on the formation of sodium
acrylate,13,15 the main intermediates in the olefin carboxyl-
ation are the olefin complex, the lactone and the carboxylate
complexes according to Scheme 2.
For a deeper understanding, we focused on the synthesis of
the species shown in Scheme 2, in order to obtain insight into
which the crucial steps are in the catalytic conversion with
nickel as the catalyst while using different olefins like
1-hexene, 3-hexene, styrene, butadiene, methyl-2,4-pentadieno-
ate, piperylene and cis-3-hexene.
Table 1 Synthesis of olefin complexes
1: dtbpe
(conv. in %)
2: dcpe
(conv. in %)
Alkene
a
100a
100a
23c
100a
b
c
91c
94b
95c mixture of isomers
97b
d
In our previous work, we just reported the catalytic conver-
sion of several olefins by using the Ni-BenzP* catalyst, without
any characterisation of intermediates or variation of the
ligand. Within these more systematic investigations, we also
focused on the use of the more common and easily available
chelating phosphines dcpe (1,2-bis(dicyclohexylphosphino)-
ethane) and dtbpe (Scheme 3).
e
f
95c mixture of
isomers
Only from (L)NiCl2
and NaBHEt3
5c
Only from (L)NiCl2
and N-Selectride
a Ni(COD)2 (0.1 mmol), ligand (0.1 mmol), olefin (0.3 mmol) at 60 °C,
overnight in d8-THF. b [Ni(COD)2] (0.1 mmol), ligand (0.1 mmol),
olefin (0.3 mmol) at room temperature, overnight in d8-THF.
c Ni(COD)2 (0.1 mmol), ligand (0.1 mmol), olefin (0.3 mmol) at 100 °C,
overnight in d8-THF (0.6 mL).
As given in Table 1, both ligands immediately formed the
corresponding olefin complexes in clean reactions for the
11084 | Dalton Trans., 2015, 44, 11083–11094
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Scheme 4 Proposed isomers for the [(dcpe)Ni(1,3-butadiene)] (2c)
complex.
31P NMR of the mixture at RT shows an expected AB spin
2
system (δA = 77.5, δB = 76.8, JAB = 54.1 Hz) for the isomer
where a 1,3-butadiene molecule is bonded in a single η2-trans
manner. Another isomer gives two broad signals, difficult to
assign at room temperature. Upon reducing the temperature to
−40 °C the two broad signals sharpen and move to a slightly
2
lower field to give a set of doublets (δA = 52.6, δB = 78.1, JAB
=
Fig. 1 Molecular structure of [(dcpe)Ni(C6H5CHCH2)] (2b). Ellipsoids
represent 50% probability. Hydrogen atoms are omitted for clarity.
Selected bond lengths (Å) and angles (°): Ni1–P1 2.159(1), Ni–P2 2.152(1),
Ni–C1 1.959(5), Ni1–C2 1.981(4), C1–Ni1–C2 42.0(2), C1–Ni1–P1
154.3(2), P1–Ni1–P2 91.9(1).
51.2 Hz). At an elevated temperature, the two sets of signals
slowly broaden, eventually becoming a single line suggesting
fluxional behaviour where most likely a η2-trans butadiene
molecule adopts a single cis-configuration via η4-bonded buta-
diene. Such isomerisations are not uncommon and were
observed previously for several transition metal butadiene
complexes.16 Therefore, while using the dienes in entries d
and e, coordination on the nickel through the primary olefin
bond was observed. Most likely, due to the presence of the
methyl- and carboxylic methyl-ester groups, coordination
through the primary olefin bond is favoured, which excludes
further the formation of allyl-carboxylato type complexes. This
reflects the poor reactivity of piperylene and methyl-2,4-penta-
dienoate in comparison to 1,3-butadiene.
According to the catalytic cycle, the formation of the
lactone from the olefin-complex with CO2 follows in the olefin
carboxylation. In the cases where the synthesis of the olefin
complex starting form [Ni(COD)2] was not successful, we
attempted the one-pot lactone formation from [Ni(COD)2]/
ligand/olefin under CO2-pressure.
The oxidative coupling of ethylene and CO2 using Ni(0)
systems has been reported for both dcpe and dtbpe
ligands.7d,13 However, whereas the ethylene complexes could
be easily converted to the desired nickelalactones in good to
excellent yields depending on the ligand, the subsequent
reaction of the olefin complexes with CO2 seems to be not
that straightforward (Scheme 5). Experiments were performed
under CO2 pressure (5 bar) with [Ni(COD)2] (0.1 mmol), ligand
(0.1 mmol) and olefin (0.3 mmol) at 100 °C overnight. While
entries a–e. The reactions of [Ni(COD)2]/dcpe with 1-hexene
and 1,3-pentadiene were relatively slow even at elevated temp-
eratures resulting in low yields after the chosen reaction time
(overnight), whereas high conversions were obtained for the
other olefin/ligand combinations. Almost all of these com-
plexes were synthesized at higher temperatures, except for the
olefin complexes of methyl-2,4-pentadienoate. Crystals of the
[(dcpe)Ni(C6H5CHCH2)] (2b) complex could be isolated by the
NMR experiment in d8-THF at ambient temperature (Fig. 1).
These were characterised by X-ray analysis. By the use of
[Ni(COD)2] as a convenient and commercially available precur-
sor with either dtbpe or dcpe and cis-3-hexene, the substi-
tution of the COD ligand was not possible. It should be noted
that the synthesis of these complexes has been reported by
Hoberg et al., however, it was only achieved starting from the
labile 16 electron [Ni(CDT)] (CDT = all trans-1,5,9-cyclododeca-
triene) complex.16 In our case, the chelating COD is a far
better ligand in comparison to 3-hexene, which makes the sub-
stitution difficult and the π-olefin complex inaccessible under
these conditions. This was also reflected in the catalysis,
where the carboxylation of cis-3-hexene could not be achieved
under the conditions we have investigated. Under the con-
ditions we used, we could also never observe any carboxylation
products of the COD, which underlines that internal olefins
are not accessible for the carboxylation in this system. Never-
theless, both complexes of 3-hexene (1f and 2f) could be syn-
thesised independently in a reaction of (L)NiCl2, cis-3-hexene
(5 equiv.) and NaBHEt3/N-selectride (3/2 equiv.) for dtbpe and
dcpe respectively. These were tested in the lactone formation
step of the overall reaction.
VT NMR experiments on the reaction of [Ni(COD)2]/dcpe
with 1,3-butadiene enabled us to distinguish between two
different isomers of the [(dcpe)Ni(1,3-butadiene)] (2c) complex
(Scheme 4).
Scheme 5 Lactone formation from [Ni(COD)2]/ligand/olefin under
CO2-pressure.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 11083–11094 | 11085

View Article Online
Paper
Dalton Transactions
the dcpe/Ni(0) butadiene complex is being carboxylated under
CO2 pressure to give the nickelalactone [(dcpe)Ni(OC(vO)
C4H6)] in a straightforward and fast reaction even at RT, only
traces of the lactone could be detected in the 31P NMR spectra
in the case of dtbpe/Ni(0) systems with 1-hexene and methyl-
2,4-pentadienoate. The carboxylation reaction on the other
olefin complexes was not successful under these conditions.
As expected, the coupling of 1,3-butadiene with CO2 in
nickel(0) does not proceed via a cyclic nickelalactone complex,
but rather yields an allyl-carboxylato complex. This is not sur-
prising, as several allyl-nickel complexes obtained via carboxy-
lation of 1,3-butadiene and 1,3-piperylene were described
before.5a It seems that the selectivity for such allyl-type com-
plexes is dependent on the ligand used in the system.17 In the
molecular structure, the remaining carbon atoms of the unsa-
turated butadiene unit form an η3-allylic structure with the
nickel atom. These findings were confirmed with X-ray analysis
of [(dcpe)Ni(OC(vO)C4H6)] obtained in THF at ambient tem-
perature. Due to disorder in the molecule, we abstained from a
quantitative discussion of the bond length in the molecule
(Fig. 2).
Scheme 6 Stoichiometric reaction of [Ni(COD)2]/ligand with cinnamic
acid, 2-heptenoic acid and 2,4-pentadienoic acid.
Unable to access many of the compounds using the pre- In order to avoid the formation of the decarboxylated product,
viously described direct route, we investigated the stoichio- the reaction was also performed under 3 bar of CO2. Although
metric reaction of [Ni(COD)2] and the ligand with trans- the decarboxylation was disfavoured, less formation of the lac-
cinnamic acid, 2-heptenoic acid as well as 2,4-pentadienoic tones was detected. On increasing the temperature to 80 °C
acid (Scheme 6). These reactions were performed in order to under the same CO2 pressure, the decarboxylated product was
obtain insights into the mechanism and an idea about the also formed as the major product. In this case, it seems that
stability of the intermediates towards decarboxylation in the the insufficient lactone stability towards the reverse reaction
overall reaction as well as to check, if the lack of lactone for- can be an explanation for the lack of catalytic activity in the
mation is preventing the catalysis.
styrene carboxylation while using Ni-dcpe. Heating the reac-
The reactions were monitored by 31P NMR spectroscopy tion mixture with 2-heptenoic acid at 40 °C for 1 day leads to
(Fig. 3). It can be seen that the reactions with both trans-cin- decomposition of the π-complex, suggesting that the nickela-
namic and 2-heptenoic acid are slow at room temperature, lactone is the more stable species. Also in this case, perform-
yielding a mixture of nickelalactone and π-carboxylic acid ing the reaction under a CO2 atmosphere (3 bar) inhibits the
complex (Fig. 4). However, at temperatures above 40 °C the formation of the nickelalactone. Cooling down the mixture
complexes resulted from the reaction with trans-cinnamic acid again, only the signals for the nickelalactone complex could be
decarboxylate to give the π-styrene [(dcpe)Ni(0)] (2b) complex. assigned, indicating simple decarboxylation of the π-carboxylic
acid complex and excluding temperature dependent equili-
brium between the two species.
Performing the reaction with cinnamic acid and
[Ni(COD)2]/BenzP*, one could observe that the cinnamic acid
complex and the nickelalactone bearing this ligand are much
more stable and decarboxylate only after heating at 100 °C.
This most likely explains why this ligand is much more active
in the catalysis with 1,3-butadiene (TON 116) and also suitable
for the carboxylation of other olefins.15
Adding a slight excess (2 equiv.) of 2,4-pentadienoic acid to
[Ni(COD)]/ligand gives quantitatively a allyl-carboxylato [(dcpe)-
Ni(OC(vO)C4H6)] complex (4c). The reaction includes hydro-
gen migration from the Ni-atom to the α-C atom of the carboxy-
late. Such oxidative additions have been described before for
Ni(0) and Pd(0).18,19 The compound shows the same spectral
features as the coupling product of [(dcpe)Ni(1,3-butadiene)]
with carbon dioxide. While dcpe gives exclusively one product,
Fig. 2 Molecular structure of [(dcpe)Ni(OC(vO)C4H6)] (4c). Ellipsoids
the reaction with dtbpe results in the formation of two
different compounds, presumably two isomeric nickelalac-
represent 50% probability. Hydrogen atoms and carbon ellipsoids in the
ligand are omitted for clarity.
11086 | Dalton Trans., 2015, 44, 11083–11094
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Fig. 4 Molecular structure of [(dcpe)Ni(C6H5CHCHCO2H)] (5b). Ellip-
soids represent 50% probability. Hydrogen atoms (except for H1) are
omitted for clarity. Selected bond lengths (Å) and angles (°): Ni1–P1
2.174(3), Ni–P2 2.168(4), Ni–C1 1.976(7), Ni1–C2 1.991(7), C1–Ni1–C2
43.1(2), C2–C3–O2 122.7(7), O2–C3–O1 120.6(7).
Fig. 5 31P{H} NMR spectra of lactone complexes synthesised from 2,4-
pentadienoic acid and [Ni(COD)2]/ligand: dcpe (left) and dtbpe (right).
Fig. 3 Time and temperature dependence in the reaction of [Ni(COD)2]/
dcpe with 2-heptenoic acid (top) and trans-cinnamic acid (bottom).
alkene rest is located at α- or β-position with respect to the car-
bonyl group, seems to be thermodynamically less favourable.
The reaction mixtures depicted in Fig. 5 were used without
purification in a subsequent reaction with a base and the
olefin (1,3-butadiene) to show the feasibility of the next
steps in the catalytic cycle: deprotonation (II) and olefin
exchange (III). In order to obtain any catalysis, the choice of
the base is also crucial: the base should not irreversibly react
with CO2 to form carbonates, but should be basic enough to
allow cleavage of the lactone to form sodium acrylate. Sodium-
tones. The small JPP coupling constants of 9 and 10 Hz respect-
ively, excludes the presence of a Ni(0) species and suggests
again an oxidative addition of the doubly unsaturated acid, fol-
lowed by a hydrogen migration step. Unfortunately, separating
and purifying these two species was unsuccessful and dis-
1
tinguishing between species in the H NMR is really difficult
due to overlapping regions with CH3 groups from dtbpe. There-
fore, assuming that these isomers have a similar structures to
4c (due to the corresponding chemical shift range and similar
JPP coupling constants), we have calculated the relative ener-
gies for various isomers such as η1- and η3-allylic structures as
well as 4- and 5-membered cyclic ethers for both dtbpe and
dcpe (Fig. 6; details in the ESI†). Surprisingly, for dtbpe, a
4-membered β-nickelalactone seems to be the most stable
isomer. As expected, η3-allylic carboxylates are also low in
energy, whereas the five-membered cyclic ethers, where the
Fig. 6 Relative energies of different isomeric nickelalactones for 3c cal-
culated in kJ mol−1 (ligand = dtbpe).
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 11083–11094 | 11087

View Article Online
Paper
Dalton Transactions
Scheme 8 Reduction of Ni(II) complexes with elemental Zn in the pres-
ence of styrene.
Scheme 7 Base-mediated cleavage of [(dcpe)Ni(lactone)] into a carb-
oxylate complex [(dcpe)Ni(RCHvCHCO2Na)] and a in situ exchange
reaction with olefin.
pounds were therefore synthesized separately, isolated and
used in the catalysis of butadiene and CO2 with the dcpe/
dtbpe ligand systems, in order to evaluate the role of Zn. Prior
to that, we performed stoichiometric reactions with these two
complexes, Zn (10 equiv.) and olefin (styrene/10 equiv.) under
standard conditions (THF, 20 h, 100 °C) and investigated the
reduction of Ni(0) species (Scheme 8).
2-fluorophenolate was identified to be an appropriate base for
this reaction (Scheme 7).15 However, sodium-2-fluorophenolate
did not give a clean reaction. For this reason, NaOtBu was
used as a stronger base, to determine whether the lactone
could be cleaved at all.
By adding a base, the coordinated sodium carboxylate can
be formed by reductive elimination of the lactone. The stability
and the reactivity of the investigated nickelalactones are rather
surprising. The allyl-carboxylato (dcpe)Ni(II) complex under-
went an exchange reaction with NaOtBu to give, by 31P NMR
[(dcpe)Ni(OC6H4F)2] was easily reduced, affording the
desired olefin complex (65% conversion after 20 h by 31P
NMR). The [(dtbpe)Ni(O2CO)] carbonate complex however, was
not reduced with Zn under these conditions, suggesting that
its formation could be one deactivation pathway of the cataly-
sis. With these assumptions, the next step was testing the com-
plexes in the catalysis using styrene with and without Zn
(Scheme 9). In the reactions without Zn, both complexes
showed minor activity and very low turnovers were achieved:
[(dcpe)Ni-(OC6H4F)2] gave a TON of 1 and [(dtbpe)Ni(O2CO)]
gave a TON of 3. After the addition of powdered Zn, slightly
higher TONs were observed: 5 vs. 1 for [(dcpe)Ni(OC6H4F)2]
and 4 vs. 3 for [(dtbpe)Ni(O2CO)]. One could argue that these
results are in agreement with the stoichiometric reactions and
that possibly Zn has a role of a reductant, at least in the case
of [(dcpe)Ni-(OC6H4F)2]. However, the very low TON values and
the given error make these arguments questionable. Therefore,
despite the certain beneficial effect of the powdered Zn in the
catalytic reactions, the mechanism of its reactivity is still not
very clear. This opens a further question, how necessary are
such large quantities of Zn in the reaction medium and is it
worth pursuing the advantages of having a homogeneous
system for a process development, albeit the slightly lower
turnover numbers.
spectroscopy,
a characteristic broadening that could be
assigned as a π-butadiene Ni(0) complex with only one equi-
valent of olefin present in the mixture. In contrast, the mixture
of isomeric (dtbpe)Ni(II) complexes mainly remained
unreacted even with an excess of base/olefin.
In the earlier studies on the catalytic reaction of ethylene
and CO2 with the BenzP* ligand, the addition of finely pow-
dered Zn proved to be beneficial for the coupling (TON 69 vs.
39), most likely reducing Ni(II) intermediates formed during
the reaction.15 A similar behaviour was observed in the cata-
lytic carboxylation of 1,3-butadiene with BenzP* (TON 137 vs.
117) and dcpe (TON 37 vs. 22). Taking our standard conditions
into consideration, we assume that Ni(II) species such as
the phenoxide complex [(L)Ni(OC6H4F)2] and the carbonate
complex [(L)Ni(O2CO)] (Fig. 7) can be formed during this reac-
tion and elemental Zn is regenerating the active catalytic
species from these complexes in the catalysis. These two com-
Catalysis was performed with [Ni(COD)2] (0.1 mmol), ligand
(0.11 mmol), 2-F-phenoxide (300 mmol), Zn (100 mmol), CO2
(20 bar) and the corresponding olefin in THF (30 mL) at
100 °C overnight. The reaction under CO2 pressure (20 bar),
Fig. 7 Molecular structure of [(dtbpe)Ni(O2CO)]. Ellipsoids represent
50% probability. Hydrogen atoms are omitted for clarity. Selected bond
lengths (Å) and angles (°): Ni1–P1 2.177(7), Ni–P2 2.165(7), Ni–O1 1.882(2), Scheme 9 Catalytic carboxylation of olefins to acrylates with CO2
Ni1–O2 1.889(2), O1–Ni1–O2 70.0(8), O1–Ni1–P2 98.4(6).
towards α,β-saturated carboxylic acid salts.
11088 | Dalton Trans., 2015, 44, 11083–11094
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
1,3-butadiene (300 equiv.) with sodium 2-fluorophenoxide (300 for Lorentz and polarisation effects.22 CCDC 1052343 (2b),
equiv.) and dcpe as the ligand gave a TON of 37. A low TON of 1052344 (4c), 1052345 (5b) and 1052346 (Ni-dtbpe carbonate)
3 could be observed using dtbpe, a result consistent with the contain the supplementary crystallographic data for this
earlier observations, in which it was shown that the corres- paper.
ponding nickelalactone is not accessible under these con-
Formation of olefin complexes 1a–f and 2a–f, NMR experi-
ditions. Both ligands showed a low activity in the catalytic ments. [Ni(COD)2] (27.5 mg, 0.1 mmol), phosphine ligand
reaction with 1,3-pentadiene (100 equiv. of olefin) giving 4 and (0.1 mmol, 1 equiv., dcpe 42.2 mg, dtbpe 38.2 mg) and olefin
11 turnovers respectively. A similar reaction was observed in (0.3 mmol, 3 equiv.) were dissolved in d8-THF (0.6 mL) and
the case of methyl-2,4-pentadienoate, where both ligands gave placed in a J-young NMR tube. The mixture was heated
a comparably low TON of 3 (100 equiv. of olefin). Assuming (60–100 °C for 24 h accordingly) to form the desired olefin
that the allyl-nickel(II) complex is not only an intermediate in complex. After the time had elapsed, the mixture was cooled
the C–C coupling reaction between butadiene and CO2, but down to room temperature and the reaction was studied by
also a catalytically active species, the complex was used in the NMR spectroscopy.
[(dtbpe)Ni(1-hexene)] (1a). 31P NMR spectroscopic yield:
catalysis under standard conditions to give 33 turnovers,
1
100%. H NMR (600 MHz, d8-THF): δ 0.88 (t, 3H, CH3, JHH
=
which is comparable to the TON achieved using the [Ni(COD)2]/
dcpe/1,3-butadiene system. Unlike the coupling of ethylene
with CO2, where the cleavage of nickelalactones [(L)Ni-
(CH2CH2CO2)] to form acrylate π-complexes [(L)Ni(η2-
CH2vCHCO2R)] is considered the most challenging step in
the catalytic cycle,13 it appears that in the case of substituted
alkenes, the lactone formation is a crucial step that influences
the outcome of the overall reaction.
7.2 Hz), 1.20 (m, 36H, C(CH3)3), 1.40 (m, 4H, CH), 1.66 (ddd,
1H, CH, J = 3.4, 6.7, 12.6 Hz), 1.72 (m, 4H, PCH2CH2P), 1.88
(m, 1H, CH), 1.95 (tdd, 1H, CH, J = 3.9, 5.9, 9.7 Hz), 2.18 (m,
1H, CH), 2.59 (m, 1H, CH); 13C NMR (151 MHz, d8-THF) δ 13.7
(s, CH3), 22.7 (dd, PCH2CH2P, JCP = 11.8, 18.8 Hz), 24.0 (dd,
PCH2CH2P, JCP = 13.8, 21.8 Hz), 29.2 (t, CH3, JCP = 7.36 Hz),
29.8 (d, C(CH3)3, JCP = 6.6 Hz), 30.0 (d, C(CH3)3, JCP = 6.8 Hz),
30.2 (d, C(CH3)3, JCP = 7.6 Hz), 30.3 (d, C(CH3)3, JCP = 7.6 Hz),
30.9 (dd, CH3, JCP = 9.2, 13.4 Hz), 34.1 (m, CH3), 34.2 (m, CH3),
34.1 (d, CH2, JCP = 4.9 Hz), 36.6 (d, CH2, JCP = 8.3 Hz), 37.0 (t,
CH2, JCP = 2.7 Hz), 38.9 (dd, CH2 from 1-hexene, JCP = 3.5, 19.3
Hz), 49.9 (dd, CH from 1-hexene, JCP = 2.5, 23.3 Hz); 31P NMR
Conclusions
We have shown that the direct carboxylation of olefins is poss-
ible using a nickel catalyst, which opens a new route towards
the desired α,β-unsaturated carboxylic acid salts. The reaction
works particularly well for 1,3-dienes and proceeds via the for-
mation of allyl-carboxylates. Investigations showed that the
ability to form such allyl-type lactone complexes represents in
this case the most challenging step towards satisfactory turn-
over numbers.
3
3
(81 MHz, d8-THF) δ 91.0 (d, JPP = 75.1 Hz), 88.6 (d, JPP = 75.1
Hz) ppm; IR after evaporation of all volatiles (ν, cm−1): 660 (w,
P–C), 728 (m), 912 (s), 1073 (s), 1179 (w), 1261 (w), 1380 (m),
1461 (s, C–H aliph.), 1974 (w), 2364 (w), 2872 (vs, C–H aliph.),
2959 (s, C–H aliph.).
[(dcpe)Ni(1-hexene)] (2a). 31P NMR spectroscopic yield: 23%.
3
31P NMR (81 MHz, d8-THF) δ 62.3 (d, JPP = 72.2 Hz), 57.7 (d,
3JPP = 71.8 Hz) ppm.
[(dtbpe)Ni(styrene)] (1b). 31P NMR spectroscopic yield: 100%.
1H NMR (600 MHz, d8-THF): δ 1.15 (m, 36H, C(CH3)3), 1.64 (m,
4H, PCH2CH2P), 2.19 (td, 2H, CH2, J = 4.8, 9.7 Hz), 3.53 (qt,
1H, CH, J = 14.7, 29.3 Hz), 6.95 (m, 5H, CH); 13C NMR
(151 MHz, d8-THF) δ 20.9 (dd, PCH2CH2P, JCP = 13.6, 18.1 Hz),
21.7 (dd, PCH2CH2P, JCP = 14.1, 20.1 Hz), 27.4 (t, CH3,
JCP = 7.0 Hz), 27.7 (d, C(CH3)3, JCP = 7.2 Hz), 28.0 (d, C(CH3)3,
JCP = 6.3 Hz), 28.3 (d, C(CH3)3, JCP = 7.0 Hz), 28.6 (d, C(CH3)3,
JCP = 7.0 Hz), 29.1 (dd, CH3, JCP = 9.4, 14.1 Hz), 32.2 (dd, CH2
from styrene, JCP = 3.4, 21.9 Hz), 32.4 (dd, CH3, JCP = 5.4,
8.7 Hz), 35.5 (dd, CH3, JCP = 4.1, 8.3 Hz), 48.2 (dd, CH from
styrene, JCP = 2.08, 19.2 Hz), 118.3 (d, CHPh, JCP = 2.3 Hz),
122.1 (d, CHPh, JCP = 2.1 Hz), 125.5 (d, CHPh, JCP = 1.7 Hz),
(148.8 (dd, ipso-CPh, JCP = 1.6, 5.6 Hz); 31P NMR (81 MHz, THF)
Experimental
General considerations
All manipulations were performed under an inert atmosphere
of dry argon by using Schlenk techniques or by working in the
glovebox. All autoclave reactions were performed under an
inert atmosphere of argon by working in a glovebox. Solvents
were purchased from Sigma-Aldrich. 1H, 31P, and 13C NMR
spectra were recorded on Bruker Avance 200, 400, 500, or
600 MHz spectrometers and were referenced to the residual
proton (1H) or carbon (13C) resonance peaks of the solvent.
Chemical shifts (δ) are expressed in ppm. 31P NMR was refer-
enced to triphenylphosphine oxide as an internal standard.
Elemental analyses were recorded by the analytical service
of the chemistry department of the University of Heidelberg.
Bis(di-tert-butylphosphino)ethane17b and [(dcpe)NiCl2]20 were
synthesised following reported procedures. X-ray structures
were solved by direct methods and refined against F2 with a
full-matrix least squares algorithm by using the SHELXTL
(Version 2008/4) software package.21 Intensities were corrected
3
3
δ 95.1 (d, JPP = 68.2 Hz), 84.4 (d, JPP = 68.2 Hz) ppm; IR after
evaporation of all volatiles (ν, cm−1): 658 (m, P–C), 911 (s),
1070 (s), 1177 (m), 1364 (m), 1459 (s, C–H aliph.), 1967 (w),
2082 (m), 2235 (m), 2682 (m), 2860 (vs, C–H aliph.), 2974 (vs,
C–H aliph.).
[(dcpe)Ni(styrene)] (2b). 31P NMR spectroscopic yield: 100%.
1H NMR (600 MHz, d8-THF): δ 1.34 (m, 44H, CH and CH2 (Cy)),
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 11083–11094 | 11089

View Article Online
Paper
Dalton Transactions
3
1.96 (m, 4H, PCH2CH2P), 2.21 (m, 2H, CH2), 3.81 (m, 1H, CH), (162.0 MHz, THF, −40 °C) δ 76.1 (d, JPP = 54.8 Hz), 75.4 (d,
6.91 (m, 5H, CH); 13C NMR (151 MHz, d8-THF) δ 19.9 (ddd, 3JPP = 54.8 Hz), δ 78.12 (d, JPP = 51.2 Hz), 52.6 (d, JPP = 51.2
PCH2CH2P, JCP = 25.8, 38.1, 26.7 Hz), 24.2–28.4 (m, CH2 from Hz) ppm. IR after evaporation of all volatiles (ν, cm−1): 657 (m,
Cy), 31.9 (dd, CH from Cy, JCP = 4.5, 13.3 Hz), 33.1 (m, CH P–C), 912 (s), 1074 (s), 1180 (s), 1364 (m), 1458 (s, C–H aliph.),
from Cy), 33.1 (d, CH2 from styrene overlapping with CH from 1810 (w), 1967 (m), 2084 (w), 2235 (w), 2361 (w), 2681 (m),
Cy, JCP = 14.4 Hz), 33.6 (dd, CH from Cy, JCP = 3.7, 15.6 Hz), 2856 (vs, C–H aliph.), 2973 (vs, C–H aliph.).
3
3
[(dtbpe)Ni(Me-2,4-pentadienoate)] (1d). 31P NMR spectro-
scopic yield: 94%. 1H NMR (600 MHz, d8-THF): δ 1.10 (m, 36H,
C(CH3)3), 1.55 (m, 4H, PCH2CH2P), 1.69 (m, 2H, CH2), 1.93 (m,
1H, CH), 3.32 (s, 3H, OCH3), 3.52 (m, 1H, CH), 4.84 (m, 1H,
CH); 13C NMR (151 MHz, d8-THF) δ 21.5 (m, PCH2CH2P), 26.8
47.7 (d, CH from styrene, JCP = 17.9 Hz), 117.8 (d, CHPh, JCP
=
2.1 Hz), 121.7 (d, CHPh, JCP = 2.5 Hz), 125.4 (d, CHPh, JCP = 1.8
Hz), 147.6 (dd, ipso-CPh, JCP = 1.4, 5.6 Hz); 31P NMR (81 MHz,
3
3
d8-THF) δ 68.7 (d, JPP = 64.4 Hz), 56.2 (d, JPP = 64.4 Hz) ppm;
IR after evaporation of all volatiles (ν, cm−1): 657 (m, P–C), 911
(s), 1069 (s), 1178 (m), 1364 (m), 1459 (s, C–H aliph.), 1812 (w),
1967 (w), 2083 (m), 2235 (m), 2682 (m), 2862 (vs, C–H aliph.),
2977 (s, C–H aliph.). Elemental analysis of the crystals isolated
from the NMR experiment calcd C 69.75%, H 9.64; found C
70.46%, H 9.75%. Single crystals suitable for X-ray analysis
were isolated from the NMR experiment. Orange crystals (poly-
hedron), dimensions 0.20 × 0.15 × 0.15 mm3, crystal system
monoclinic, space group P21/c, Z = 4, a = 10.6643(18) Å, b =
9.5838(16) Å, c = 31.345(6) Å, β = 97.564(5)°, V = 3175.7(9) Å3,
ρ = 1.224 g cm−3, T = 200(2) K, θmax = 25.090°, radiation Mo Kα,
λ = 0.71073 Å, 0.5° ω scans with a CCD area detector, covering
the asymmetric unit in reciprocal space; reflections:
17 547 measured, 5635 unique (Rint = 0.0606), 4308 observed
(I > 2σ(I)); μ = 0.73 mm−1, min/max transmission: 0.77/0.92,
346 parameters refined, hydrogen atoms were treated using
appropriate riding models, except for the olefinic ones, which
were refined isotropically, goodness of fit: 1.10 for observed
reflections, final residual values R1(F) = 0.058, wR(F2) = 0.088
for observed reflections, residual electron density −0.37 to
(t, CH3, JCP = 7.2 Hz), 27.6 (m, C(CH3)3), 28.5 (dd, CH3, JCP
=
8.9, 13.6 Hz), 32.5 (m, CH3), 33.2 (m, CH3), 35.2 (m, CH2 from
Me-2,4-pentadienoate), 46.3 (s, OCH3), 48.1 (m, CH from Me-
2,4-pentadienoate), CH from Me-2,4-pentadienoate could not
be located. 169.5 (m, CvO); 31P NMR (81 MHz, d8-THF) δ 93.4
(d, 3JPP = 53.7 Hz), 88.5 (d, 3JPP = 57.4 Hz) ppm; IR after evapor-
ation of all volatiles (ν, cm−1): 660 (m, P–C), 749 (m), 842 (m),
911 (s), 1069 (s), 1160 (m), 1268 (s, C–O), 1365 (m, –CH3), 1460
(s), 1583 (w), 1681 (m, –CvC–), 1723 (m, CvO), 1885 (w), 1970
(w), 2083 (s), 2235 (s), 2682 (w), 2863 (vs, C–H aliph.), 2976 (vs,
C–H aliph.).
[(dcpe)Ni(Me-2,4-pentadienoate)] (2d). 31P NMR spectroscopic
yield: 97%. 1H NMR (600 MHz, d8-THF): δ 1.66 (m, 44H,
CH2 (Cy)), 1.61 (m, 4H, PCH2CH2P), 2.65 (m, 1H, CH), 3.00 (m,
2H, CH2), 3.33 (s, 3H, OCH3), 3.52 (m, 1H, CH), 4.58 (m, 1H,
CH); 13C NMR (151 MHz, d8-THF) δ 13.3 (m, PCH2CH2P),
24.2–27.9 (m, CH2 from Cy), 32.2 (m, CH from Cy), 32.5 (m,
CH from Cy), 33.3 (m, CH from Cy), 33.9 (m, CH from Cy), CH2
from Me-2,4-pentadienoate overlapping with CH from Cy, 46.7
(s, OCH3), 48.6 (m, CH from Me-2,4-pentadienoate), 81.0 (br. s,
CH from Me-2,4-pentadienoate), 94.5 (br. s, CH from Me-2,4-
pentadienoate), 168.2 (s, CvO); 31P NMR (81 MHz, d8-THF) δ
0.45 e Å−3
.
[(dtbpe)Ni(1,3-butadiene)]17c (1c). 31P NMR spectroscopic
yield: 91%. 1H NMR (600 MHz, d8-THF): δ 1.04 (m, 36H,
C(CH3)3), 1.56 (dd, 4H, PCH2CH2P, J = 7.0, 15.5 Hz), 3.00 (d, 1H,
CH2, JHH = 8.6 Hz), 3.16 (d, 1H, CH2, JHH = 13.4 Hz), 4.46 (m,
3
3
80.0 (d, JPP = 29.2 Hz), 60.0 (d, JPP = 31.0 Hz) ppm; IR after
evaporation of all volatiles (ν, cm−1): 655 (m, P–C), 747 (m),
842 (m), 911 (s), 1069 (s), 1163 (m), 1268 (m, C–O), 1364 (m),
1459 (s), 1667 (m, –CvC–), 1723 (m, CvO), 1968 (w), 2082 (s),
2235 (s), 2682 (m), 2859 (vs, C–H aliph.), 2974 (vs, C–H aliph.).
[(dtbpe)Ni(piperylene)] (1e). 31P NMR spectroscopic yield:
95%. 1H NMR (600 MHz, d8-THF): δ 1.11 (m, 36H, C(CH3)3),
1.31 (d, 3H, CH3, JHH = 8.3 Hz), 2.19 (m, 2H, CH2), 3.42 (m, 1H,
CH), 4.66 (m, 1H, CH), 5.25 (m, 1H, CH); PCH2CH2P
bridge could not be assigned due to overlapping with signals
from the unreacted starting material; 13C NMR (151 MHz,
1H, CH), 4.93 (d, 1H, CH, JHH = 9.4 Hz), 5.05 (d, 1H, CH2, JHH
=
15.5 Hz), 6.16 (m, 1H, CH); 13C NMR (151 MHz, d8-THF) δ 20.8
(t, PCH2CH2P, JCP = 16.6 Hz), 26.8 (t, CH3, JCP = 7.3 Hz), 27.7
(d, C(CH3)3, JCP = 6.9 Hz), 28.6 (dd, CH3, JCP = 9.0, 13.7 Hz),
30.4 (t, CH2 from 1,3-butadiene, JCP = 3.4 Hz), 31.8 (t, CH from
1,3-butadiene, JCP = 5.8 Hz); 31P NMR (162.0 MHz, d8-THF,
3
3
−40 °C) δ 90.7 (d, JPP = 68.6 Hz), 83.8 (d, JPP = 68.6 Hz) ppm;
IR after evaporation of all volatiles (ν, cm−1): 657 (m, P–C), 733
(m), 913 (s), 1067 (s), 1168 (s), 1364 (m), 1459 (s, C–H aliph.),
1601 (w), 1808 (w), 1967 (m), 2082 (s), 2235 (s), 2682 (m), 2862
(vs, C–H aliph.), 2971 (vs, C–H aliph.).
d8-THF)
δ 15.3 (s, CH3 from 1,3-pentadiene), 18.6 (t,
PCH2CH2P, JCP = 21.2 Hz), 19.8 (t, PCH2CH2P, JCP = 19.8 Hz),
24.5–27.8 (m, C(CH3)3), 30.2 (t, CH from 1,3-pentadiene,
JCP = 3.4 Hz), 32.1 (m, CH3), 32.7 (m, CH3), 33.4 (t, CH2 from
[(dcpe)Ni(1,3-butadiene)]17c (2c). 31P NMR spectroscopic
yield: 95%. 1H NMR (600 MHz, d8-THF): δ 1.27 (m, 44H, CH
and CH2 (Cy)), 1.51 (d, 4H, PCH2CH2P, J = 8.7 Hz), 1.66 (d, 4H,
PCH2CH2P, J = 10.9 Hz), 2.78 (d, 1H, CH2, JHH = 12.6 Hz), 2.86
(d, 1H, CH2, JHH = 7.3 Hz), 4.71 (m, 1H, CH), 5.05 (d, 1H, CH2,
JHH = 8.9 Hz), 5.17 (d, 1H, CH, JHH = 15.4 Hz), 6.32 (m, 1H,
1,3-pentadiene, JCP = 7.4 Hz), 76.6 (t, CH from 1,3-pentadiene,
3
JCP = 2.3 Hz); 31P NMR (81 MHz, d8-THF) δ 98.8 (d, JPP
=
3
3
53.7 Hz), 97.5 (d, JPP = 53.1 Hz), δ 89.6 (d, JPP = 71.7 Hz),
86.5 (d, JPP = 71.6 Hz) ppm; IR after evaporation of all
3
volatiles (ν, cm−1): 655 (m, P–C), 747 (m), 909 (s), 1069 (s),
1177 (s), 1364 (m), 1459 (s, C–H aliph.), 1817 (w), 1967 (m),
2083 (s), 2235 (s), 2682 (m), 2860 (vs, C–H aliph.), 2975 (s, C–H
aliph.).
CH); 13C NMR (151 MHz, d8-THF) δ 19.9 (t, PCH2CH2P, JCP
19.94 Hz), 24.6–27.5 (m, CH2 from Cy), 30.2 (t, CH2 from 1,3-
butadiene, JCP = 3.5 Hz), 33.2 (dd, CH from Cy, JCP = 4.2, 14.5
Hz), 76.7 (t, CH from 1,3-butadiene, JCP = 2.6 Hz); 31P NMR
=
11090 | Dalton Trans., 2015, 44, 11083–11094
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
3
[(dcpe)Ni(piperylene)] (2e). 31P NMR spectroscopic yield: 5%. 71.5 (d, JPP = 10.9 Hz) ppm. Due to very low conversion, the
3
31P NMR (81 MHz, d8-THF) δ 63.6 (d, JPP = 61.5 Hz), 57.2 (d, formed lactones [(dtbpe)Ni(OC(vO)C4H6)] and [(dcpe)Ni-
3JPP = 61.5 Hz) ppm.
[(dtbpe)Ni(cis-3-hexene)] (1f). A solution of NaBHEt3 in THF
(300 μL, 1 M, 0.3 mmol) was added at room temperature to a
suspension of (dtbpe)NiCl2 (44.8 mg, 0.1 mmol) and cis-3-
(OC(vO)C5H5OMe)] could not be isolated and further charac-
terised. For [(dcpe)Ni(OC(vO)C4H6)] a dedicated synthesis was
achieved.
[(dcpe)Ni(OC(vO)C4H6)] (4c). [Ni(COD)2] (0.36 mmol,
hexene (42.0 μL, 0.5 mmol) in 0.6 mL of d8-THF. The color 100 mg) and dcpe (0.36 mmol, 153 mg) were suspended in
changed from red to brown and release of gas was observed. THF (30 mL). A solution of 1,3-butadiene (1.08 mmol, 360 μL)
The mixture was stirred (the reaction vessel was open to an Ar was added in toluene (20 wt%). The mixture was added in an
line) at room temperature for 1 hour. 31P NMR spectroscopic autoclave and stirred (300 rpm) at 100 °C for 24 h. When the
yield: 65%. 1H NMR (600 MHz, d8-THF): δ 1.08 (t, 3H, CH3, time had elapsed, the mixture was cooled down to room tem-
JHH = 7.3 Hz), 1.21 (m, 36H, C(CH3)3), 1.49 (m, 2H, PCH2CH2P), perature and pressurised with CO2 (20 bar, 15 min equilibration
1.65 (m, 4H, CH2), 2.17 (m, 2H, PCH2CH2P), 2.68 (m, 2H, CH); time) at 25 °C. The mixture was stirred (300 rpm) at RT for
13C NMR (151 MHz, d8-THF) δ 15.3 (t, CH3 from cis-3-hexene, additional 20 h. When the time had elapsed, the pressure was
JCP = 4.9 Hz), 20.8 (t, CH2 from cis-3-hexene, overlapping with released and the autoclave was opened under an inert argon
CH2 of dtbpe), 20.9 (t, PCH2CH2P, JCP = 16.7 Hz), 27.3 (t, CH3, atmosphere in the glove box. The solution which was concen-
JCP = 3.6 Hz), 28.1 (m, C(CH3)3), 31.9 (m, CH3), 54.5 (t, CH trated and precipitated by addition of pentane (10 mL) yielded
from cis-3-hexene, JCP = 9.4 Hz); 31P NMR (81 MHz, d8-THF) δ the compound as a fine red solid. Yield: 146 mg, 70%. 1H
83.5 ppm, (s); IR after evaporation of all volatiles (ν, cm−1): 662 NMR (400 MHz, d8-THF) δ 1.58 (m, 44H, CH and CH2, Cy), 2.70
(w, P–C), 748 (m), 841 (s), 910 (s), 1048 (s), 1100 (s), 1160 (s), (m, 2H, CH2 next to carbonyl), 3.51 (m, 1H, CH), 4.55 (m, 2H,
1385 (m), 1460 (m, C–H aliph.), 1596 (w), 1885 (w), 2083 (s), CH2), 4.84 (m, 1H, CH); 13C NMR (151 MHz, d8-THF) δ 20.7 (m,
2235 (s), 2794 (m, C–H aliph.), 2850 (s, C–H aliph.), 2892 (s, PCH2CH2P), 21.8 (m, PCH2CH2P), 25.4–29.4 (m, CH2 from Cy),
C–H aliph.), 2929 (s, C–H aliph.).
33.6 (t, CH from Cy, JCP = 11.5 Hz), 35.0 (dd, CH from Cy, JCP =
[(dcpe)Ni(cis-3-hexene)]16 (2f). A solution of Na selectride in
THF (40 μL, 1 M, 0.04 mmol) was added at room temperature
to a suspension of (dcpe)NiCl2 (11.3 mg, 0.02 mmol) and cis-3-
hexene (12.6 μL, 0.1 mmol) in 0.6 mL of d8-THF. NMR analysis
after 15 min at RT showed full conversion of the starting
materials and selective formation of the desired complex. 31P
NMR spectroscopic yield: 20%. 31P NMR (81 MHz, d8-THF) δ
58.0 ppm (s).
All olefin complexes were used in the next step of reaction
without being isolated and without further purification. Due to
low conversion or instability, [(dcpe)Ni(1-hexene)], [(dcpe)Ni-
(piperylene)] and [(dcpe)Ni(cis-3-hexene)] were identified only
by 31P NMR spectroscopy.
4.4, 14.2 Hz), 37.8 (d, CH, JCP = 3.4 Hz), 58.7 (d, CH, JCP = 21.9
Hz), 60.0 (d, CH, JCP = 15.3 Hz), 70.3 (d, CH2, JCP = 17.7 Hz),
177.4 (d, CvO, JCP = 11.4 Hz); 31P NMR (162.0 MHz, d8-THF) δ
77.1 (d, 3JPP = 6.4 Hz), 71.3 (d, 3JPP = 6.4 Hz) ppm; IR (ν, cm−1):
658 (m, P–C), 910 (s), 1069 (s), 1181 (m), 1364 (m), 1459 (s,
C–H aliph.), 1621 (m, CvO), 1804 (w), 1967 (m), 2364 (w),
2682 (m), 2861 (vs, C–H aliph.), 2977 (vs, C–H aliph.). Elemen-
tal analysis calcd C 64.56%, H 9.39; found C 65.10%, H 9.32%.
Crystals suitable for X-ray analysis could be obtained from con-
centrated THF solution at RT. Orange crystals (polyhedron),
dimensions 0.13 × 0.12 × 0.10 mm3, crystal system ortho-
rhombic, space group Pnma, Z = 8, a = 31.6368(16) Å, b =
21.6435(12) Å, c = 9.1291(4) Å, V = 6251.0(5) Å3, ρ = 1.231 g
cm−3, T = 200(2) K, θmax = 24.984°, radiation Mo Kα, λ =
0.71073 Å, 0.5° ω-scans with a CCD area detector, covering
the asymmetric unit in reciprocal space; reflections:
40 500 measured, 4469 unique (Rint = 0.0843), 2880 observed
(I > 2σ(I)); μ = 0.73 mm−1, min/max transmission: 0.88/0.94,
409 parameters refined, hydrogen atoms were treated using
appropriate riding models, goodness of fit: 1.08 for observed
reflections, final residual values R1(F) = 0.076, wR(F2) = 0.173
for observed reflections, residual electron density −0.56 to
Formation of nickelalactones 3a–f and 4a–f in an high-
pressure (HP) NMR experiments. [Ni(COD)2] (27.5 mg,
0.1 mmol), phosphorus ligand (0.1 mmol, 1 equiv., dcpe
42.2 mg, dtbpe 38.2 mg) and olefin (0.3 mmol, 3 equiv.) were
dissolved in d8-THF (0.6 mL) and placed in a HP NMR tube.
The mixture was heated for 24 h accordingly, to form the
desired olefin complex. After the time had elapsed, the
mixture was cooled down to room temperature and the NMR
tube was consecutively pressurized with CO2 (5 bar, 15 min
equilibration time). The mixture was heated at 100 °C for an
additional 24 h, except in the case of 1,3-butadiene where the
reaction proceeds at RT. The reaction was studied by 31P NMR
spectroscopy.
0.53 e Å−3
.
Stoichiometric reactions between [Ni(COD)]/ligand with
α,β-unsaturated carboxylic acids. [Ni(COD)2] (27.5 mg,
0.1 mmol), ligand (0.1 mmol, 1 equiv., dcpe 42.2 mg, dtbpe
38.2 mg), 2-heptenoic acid (12.8 mg, 0.1 mmol)/cinnamic
acid (14.8 mg, 0.1 mmol)/2,4-pentadienoic acid (24.9 mg,
0.3 mmol) were dissolved in d8-THF (0.6 mL) in a NMR tube.
The mixtures was analysed by 31P NMR spectroscopy after
given times and at given temperatures. The reactions with
trans-cinnamic acid and 2-heptenoic acid are very slow and the
conversions to lactones and acid-π-complexes are very low in
[(dtbpe)Ni(OC(vO)C6H12)] (3a). 31P NMR spectroscopic
3
yield: 6%. 31P NMR (81 MHz, d8-THF) δ 83.5 (d, JPP = 9.1 Hz),
3
79.5 (d, JPP = 8.7 Hz) ppm. [(dcpe)Ni(OC(vO)C4H6)] (4c):
31P NMR spectroscopic yield: 100%. 31P NMR (81 MHz,
3
3
d8-THF) δ 77.1 (d, JPP = 6.4 Hz), 71.4 (d, JPP = 6.4 Hz) ppm.
[(dcpe)Ni(OC(vO)C5H5OMe)] (4d): 31P NMR spectroscopic
yield: 4%. 31P NMR (81 MHz, d8-THF) δ 78.2 (d, 3JPP = 10.6 Hz),
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 11083–11094 | 11091

View Article Online
Paper
Dalton Transactions
comparison to the unreacted [(dcpe)Ni(COD)] complex For at 80 °C for the trans-cinnamic acid complex whereas the
these reasons, it was extremely difficult to separate the mix- 2-heptenoic acid complex did not decompose.
tures, and the compounds were identified only by 31P NMR
Stoichiometric reactions between [Ni(COD)]/ligand with
spectroscopy. Due to low solubility, it was possible to isolate 2,4-pentadienoic acid and a subsequent reaction with a base.
the [(dcpe)Ni(trans-cinnamic acid)] (5b) complex from the reac- [Ni(COD)2] (27.5 mg, 0.1 mmol), ligand (0.1 mmol, 1 equiv.,
tion mixture and have it fully characterised. [(dcpe)Ni((C4H9)- dcpe 42.2 mg, dtbpe 38.2 mg), 2,4-pentadienoic acid (24.9 mg,
3
C–CC(vO)O)] (4a): 31P NMR (81 MHz, d8-THF) δ 69.6 (d, JPP
=
0.3 mmol) were dissolved in d8-THF (2 mL) in a pressure NMR
tube and reacted at RT for 2 hours to give [(dtbpe)Ni(OC(vO)-
C4H6)] (3c), [(dcpe)Ni(OC(vO)C4H6)] (4c) respectively as
described in the previous section. 1 equivalent of base (Na-2-
fluorophenoxide (0.1 mmol, 13.4 mg)/NaOtBu (0.1 mmol,
9.8 mg)) and 1 equivalent of 1,3-butadiene (16 wt% solution
in toluene, 30 µL) were added and the reaction mixtures
were analysed by 31P NMR. [(dtbpe)Ni(OC(vO)C4H6)] (3c) gives
no reaction with sodium 2-fluorophenoxide/NaOtBu and
1,3-butadiene and only signals for 3c could be observed in the
31P NMR spectra. [(dcpe)Ni(OC(vO)C4H6)] (4c): The reaction
of 4c with sodium 2-fluorophenoxide/1,3-butadiene results
in decomposition. The reaction of 4c with NaOtBu/1,3-buta-
diene gives a broad signal that corresponds to [(dcpe)Ni(1,3-
butadiene)]17c (2c): 31P NMR (81 MHz, d8-THF, 25 °C) δ
70.0 ppm, br s.
8.0 Hz), 63.3 (d, 3JPP = 8.4 Hz) ppm.
[(dcpe)Ni(2-heptenoic acid)] (5a). 31P NMR (81 MHz, d8-THF)
δ 64.3 (d, 3JPP = 55.5 Hz), 59.0 (d, 3JPP = 56.1 Hz) ppm.
[(dcpe)Ni(PhC–CC(vO)O)] (4b). 31P NMR (81 MHz, THF) δ
3
3
69.7 (d, JPP = 2.6 Hz), 61.9 (d, JPP = 2.7 Hz) ppm; [(dcpe)Ni-
(trans-cinnamic acid)] (5b): H NMR (600 MHz, d8-THF) δ 1.37
1
(m, 44H, CH and CH2 (Cy)), 3.39 (m, 1H, CH), 4.05 (m, 1H,
CH), 6.80 (m, 1H, CH), 6.95 (m, 2H, CH), 7.02 (m, 2H, CH),
9.30 (s, 1H, COOH); 13C NMR (151 MHz, d8-THF) δ 19.1 (dd,
PCH2CH2P, JCP = 20.1 Hz), 19.7 (dd, PCH2CH2P, JCP = 20.7 Hz),
23.5–28.6 (m, CH2 from Cy), 31.3 (dd, CH from Cy, JCP = 4.8,
16.0 Hz), 31.6 (dd, CH from Cy, JCP = 4.7, 15.1 Hz), 33.0 (dd,
CH from Cy, JCP = 1.8, 17.9 Hz), 33.7 (dd, CH from Cy, JCP
1.9, 17.9 Hz), 39.5 (d, CH, JCP = 15.1 Hz), 47.8 (d, CH, JCP
=
=
20.7 Hz), 119.3 (s, CHPh), 122.6 (d, CHPh, JCP = 1.9 Hz),
125.7 (s, CHPh), 145.8 (d, ipso-CPh, JCP = 5.7 Hz), 172.5 (d,
C(vO)OH, JCP = 3.6); 31P NMR (81 MHz, d8-THF) δ 65.5
[(dcpe)Ni(OC6H4F)2]. [(dcpe)NiCl2] (55.2 mg, 0.1 mmol) and
sodium-2-fluorophenoxide (27 mg, 0.2 mmol) were suspended
in THF (1 mL). The orange suspension turned red within
15 min and was stirred for an additional hour. The formed
NaCl was filtered off and the bright red solution was dried
under vacuum yielding the phenoxide complex as a fine red
3
3
(d, JPP = 50.5 Hz), 62.6 (d, JPP = 50.3 Hz) ppm; IR after
evaporation of all volatiles (ν, cm−1): 657 (m, P–C), 910 (s),
1068 (s), 1180 (s), 1364 (m), 1459 (s, C–H aliph.), 1802 (w),
1967 (m), 2363 (w), 2682 (m), 2862 (vs, C–H aliph.), 2977 (vs,
C–H aliph.). Elemental analysis of the crystals isolated from
the NMR experiment calcd C 66.78%, H 8.97%; found C
66.95%, H 8.87%. Single crystals suitable for X-ray analysis
were isolated from the NMR experiment. Yellow crystals
(needle), dimensions 0.14 × 0.08 × 0.06 mm3, crystal system
monoclinic, space group P21/c, Z = 4, a = 16.78(3) Å, b = 8.692
(16) Å, c = 27.13(5) Å, β = 103.38(3)°, V = 3850(13) Å3, ρ =
1.210 g cm−3, T = 200(2) K, θmax = 22.463°, radiation Mo Kα,
λ = 0.71073 Å, 0.5° ω-scans with a CCD area detector,
cover the asymmetric unit in reciprocal space; reflections:
18 748 measured, 5010 unique (Rint = 0.1386), 2827 observed
(I > 2σ(I)); μ = 0.62 mm−1, min/max transmission 0.86/0.98;
410 parameters refined, hydrogen atoms were treated using
appropriate riding models, except for H1 of the carboxy group,
which was refined isotropically, goodness of fit of 1.02
for observed reflections, final residual values of R1(F) = 0.066,
wR(F2) = 0.127 for observed reflections, and residual electron
density −0.31 to 0.62 e Å−3. [(dcpe)Ni(OC(vO)C4H6)] (4c):
1
powder. H NMR (300 MHz, d8-THF) δ 1.64 (m, 44H, CH and
CH2 (Cy)), 5.92 (m, 1H, CH), 6.39 (m, 2H, CH), 6.96 (m, 1H,
CH); 13C NMR (151 MHz, d8-THF) δ 18.6 (t, PCH2CH2P, JCP
=
19.5 Hz), 24.1 (s, CH2 from Cy), 25.1 (m, CH2 from Cy), 26.4 (s,
CH2 from Cy), 27.1 (s, CH2 from Cy), 31.9 (t, CH from Cy, JCP
11.2 Hz), 109.6 (d, CHPh JCP = 6.5 Hz), 111.6 (d, CHPh, JCP
19.7 Hz), 120.8 (dd, CHPh, JCP = 2.8, 39.4), 152.6 (d, CHPh, JCP
=
=
=
11.2 Hz), 153.0 (s, CPh–O), 155.4 (CPh–F); 31P NMR (162.0 MHz,
d8-THF) δ 69.6 ppm (s); 19F (282 MHz, d8-THF) δ −138.6 ppm
(s); IR (ν, cm−1): 657 (m, P–C), 910 (s), 1069 (s), 1181 (m), 1364
(m), 1459 (s, C–H aliph.), 1597 (w, C–C arom.), 1800 (w), 1967
(m), 2364 (w), 2682 (m), 2861 (vs, C–H aliph.), 2976 (vs, C–H
aliph.). Elemental analysis calcd C 64.88%, H 8.02%; found C
64.84%, H 8.10%.
Reduction reactions with Zn. [(dcpe)Ni(OC6H4F)2] (10 mg,
0.014 mmol), Zn powder (9 mg, 0.14 mmol) and styrene
(16 μL, 0.14 mmol) were suspended in d8-THF (1 mL) and
heated at 100 °C. After 20 h, the mixture was filtered to remove
the excess of Zn and characterised by 31P NMR spectroscopy.
The product was identified as a [(dcpe)Ni(styrene)] complex.
31P NMR (81 MHz, d8-THF) δ 77.1 (d, JPP = 6.4 Hz), 71.4
(d, JPP = 6.4 Hz) ppm. Full characterisation of this compound
3
3
3
31P NMR (81 MHz, d8-THF) δ 68.7 (d, JPP = 64.4 Hz), 56.2 (d,
has been described in the previous section.
[(dtbpe)Ni(OC(vO)C4H6)] (3c). 31P NMR (81 MHz, d8-THF) δ
3JPP = 64.4 Hz) ppm. The conversion was 65%.
98.1 (d, 3JPP = 10.1 Hz), 91.2 (d, 3JPP = 10.1 Hz), δ 93.0 (d, 3JPP
8.6 Hz), 88.4 (d, 3JPP = 8.6 Hz) ppm.
=
[(dtbpe)Ni(O2CO)].23 [Ni(COD)2] (120 mg, 0.43 mmol) and
dtbpe (137 mg, 0.43 mmol) were suspended in DMF (10 mL).
The mixture was placed in an autoclave under an inert atmos-
phere in the glovebox. Outside the glovebox, the autoclave was
pressurised with ethylene (5 bar, 15 min equilibration time)
and CO2 (35 bar, 15 min equilibration time) at 25 °C. The
The same reactions were also performed under a CO2
atmosphere (3 bar) in a pressure NMR-tube. Under these con-
ditions, the formation of the nickelalactone is inhibited. The
acid-π-complexes were observed and decarboxylation occurred
11092 | Dalton Trans., 2015, 44, 11083–11094
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
mixture was stirred (300 rpm) and heated at 65 °C for 2 h. reaction mixture as an internal standard. The vial was washed
After the time had elapsed, the mixture was cooled down to with additional D2O (5 mL). Then, the reaction mixture was
room temperature and the pressure was released. The auto- layered with Et2O (40 mL) and the aqueous phase (2 mL) was
clave was carefully opened under an argon flow and degassed centrifuged to improve the phase separation. The corres-
water (5 mL) was added to the mixture. The autoclave was ponding sodium salt was then quantified by 1H-NMR
closed under an argon atmosphere and pressurised again with (200 MHz, 70 scans) and the turnover number (TON) was cal-
ethylene (5 bar, 15 min equilibration time) and CO2 (35 bar, culated accordingly.
15 min equilibration time) at 25 °C. The mixture was stirred
(300 rpm) and heated at 100 °C for 18 h. After the time had
elapsed, the mixture was cooled down to room temperature
Acknowledgements
and the pressure was released. The autoclave was carefully
CaRLa (Catalysis Research Laboratory) is co-financed by the
opened in the glove box and the mixture was transferred in a
Ruprechts-Karls-University Heidelberg (University of Heidel-
Schlenk flask. All solvents were evaporated and the orange-
berg) and BASF SE.
brown oily residue was washed with pentane (3 × 10 mL) yield-
ing the carbonate complex as an orange fine powder. Yield:
130 mg, 74%. 1H NMR (600 MHz, d8-THF) δ 1.36 (m, 36H,
Notes and references
C(CH3)3), 1.62 (m, 4H, PCH2CH2P); 13C NMR (151 MHz, d8-THF)
δ 19.7 (m, PCH2CH2P), 20.6 (m, PCH2CH2P), 27.3 (t, CH3, JCP
=
1 Selected reviews containing transformations of olefins with
CO2: (a) B. Yu, Z. F. Diao, C. X. Guo and L. N. He, J. CO2
Util., 2013, 1, 60; (b) M. Cokoja, C. Bruckmeier, B. Rieger,
W. A. Herrmann and F. E. Kühn, Angew. Chem., Int. Ed.,
2011, 50, 8510; (c) P. Braunstein, D. Matt and D. Nobel,
Chem. Rev., 1988, 88, 747; (d) Y. Tsuji and T. Fujihara,
Chem. Commun., 2012, 48, 9956.
2 Acrylate formation of metal complexes: (a) R. Alvarez,
E. Carmona, D. J. Cole-Hamilton, A. Galindo, E. Gutierrez-
Puebla, A. Monge, M. L. Poveda and C. Ruiz, J. Am. Chem.
Soc., 1985, 107, 5529; (b) A. Galindo, A. Pastor, P. Pérez and
E. Carmona, Organometallics, 1993, 12, 4443; (c) R. Fischer,
J. Langer, A. Malassa, D. Walther, H. Görls and
G. Vaughan, Chem. Commun., 2006, 2510; (d) D. Jin,
P. G. Williard, N. Hazari and W. H. Bernskoetter, Chem. –
Eur. J., 2014, 20, 3205.
3 A. Hoberg, A. Ballesteros, A. Signan, C. Jegat and
A. Milchereit, Synthesis, 1991, 395.
4 H. Hoberg, Y. Peres and A. Milchereit, J. Organomet. Chem.,
1986, 307, C38.
5 (a) H. Hoberg, D. Schaefer and B. W. Oster, J. Organomet.
Chem., 1984, 266, 313; (b) H. Hoberg and S. Schaefer,
J. Organomet. Chem., 1983, 255, C15.
6 (a) S. Y. T. Lee, M. Cokoja, M. Drees, Y. Li, J. Mink,
W. A. Herrmann and F. E. Kühn, ChemSusChem, 2011, 4,
1275; (b) C. Bruckmeier, M. W. Lehenmeier, R. Reichardt,
S. Vagin and B. Rieger, Organometallics, 2010, 29, 2199.
7 (a) H. Hoberg and D. Schaefer, J. Organomet. Chem., 1982,
236, C28; (b) H. Hoberg, K. Jenni, C. Krüger and E. Raabe,
Angew. Chem., Int. Ed. Engl., 1986, 98, 819; (c) H. Hoberg
and K. Jenni, J. Organomet. Chem., 1987, 322, 193;
(d) H. Hoberg and D. Schaefer, J. Organomet. Chem., 1983,
251, C51; (e) A. Behr and U. Kanne, J. Organomet. Chem.,
1986, 317, C41; (f) M. Takimoto and M. Mori, J. Am. Chem.
Soc., 2001, 123, 2895.
7.3 Hz), 27.6 (m, C(CH3)3), 29.1 (dd, CH3, JCP = 9.5, 14.1 Hz),
33.4 (dd, CH3, JCP = 6.8, 9.3 Hz), 34.0 (dd, CH3, JCP = 8.5, 15.5
Hz), 162.5 (d, CvO, JCP = 4.1 Hz); 31P NMR (243.0 MHz, d8-
THF) δ 90.3 ppm (s). IR (ν, cm−1): 658 (m, P–C), 912 (s), 1068
(s), 1181 (m), 1364 (m), 1459 (s, C–H aliph.), 1563 (w), 1627
(w), 1636 (m, CvO), 1967 (m), 2682 (m), 2870 (vs, C–H aliph.),
2970 (vs, C–H aliph.). Crystals suitable for X-ray analysis could
be obtained from concentrated THF solution at RT. Yellow
crystals (plate), dimensions 0.17 × 0.07 × 0.04 mm3, crystal
system monoclinic, space group P21/c, Z = 4, a = 9.9880(9) Å,
b = 14.3032(12) Å, c = 16.3855(13) Å, β = 106.4993(14)°, V =
2244.4(3) Å3, ρ = 1.294 g cm−3, T = 200(2) K, θmax = 29.612°,
radiation Mo Kα, λ = 0.71073 Å, 0.5° ω-scans with a CCD area
detector, covering the asymmetric unit in reciprocal space;
reflections: 19 179 measured, 6295 unique (Rint = 0.0581), 4243
observed (I > 2σ(I)); μ = 1.02 mm−1, min/max transmission:
0.86/0.96, 238 parameters refined, hydrogen atoms were
treated using appropriate riding models, goodness of fit: 1.04
for observed reflections, final residual values R1(F) = 0.050,
wR(F2) = 0.098 for observed reflections, residual electron
density −0.38 to 0.53 e Å−3
.
General procedure for catalytic synthesis of acrylate deriva-
tives. Ni precursor (0.1 mmol, [Ni(COD)2] 27.5 mg, [(dcpe)Ni-
(OC6H4F)2] 70.3 mg, [(dtbpe)Ni(O2CO)] 43.7 mg) ligand
(0.1 mmol, 1 equiv., dcpe 42.2 mg, dtbpe 38.2 mg), sodium
2-fluorophenolate (30.0 mmol, 300 equiv.), Zn (10.0 mmol,
100 equiv.) and alkene (30.0 mmol, 300 equiv.) were intro-
duced into the autoclave together with 30 mL of THF under an
argon atmosphere in the glove box. Outside the glovebox, the
autoclave was pressured with CO2 (20 bar, 15 min equilibration
time) at 25 °C. The reaction mixture was therefore stirred
(300 rpm) at 100 °C for 20 h. When the time had elapsed, the
autoclave was cooled down to room temperature and pressure
was released. It was opened and the reaction crude was
transferred to a 100 mL glass bottle. D2O (15 mL) was used
to wash the autoclave and added to the reaction mixture.
8 H. Hoberg, A. Ballesteros and A. Signan, J. Organomet.
Chem., 1991, 403, C19.
3-(Trimethylsilyl)propionic-2,2,3,3-d4
(0.125 mmol) was dissolved in D2O (5 mL) and added to the
acid
sodium
salt
9 (a) H. Hoberg, Y. Peres, A. Milchereit and S. Gross, J. Orga-
nomet. Chem., 1988, 345, C17; (b) H. Hoberg, Y. Peres,
This journal is © The Royal Society of Chemistry 2015
Dalton Trans., 2015, 44, 11083–11094 | 11093

View Article Online
Paper
Dalton Transactions
C. Krüger and Y. H. Tsay, Angew. Chem., Int. Ed. Engl., 1987, 17 (a) A. Döhring, R. Goddard, G. Hopp, P. W. Jolly, N. Kokel
99, 799; (c) H. Hoberg and D. Bärhausen, J. Organomet.
Chem., 1989, 379, C7; (d) H. Hoberg, Y. Peres and
A. Milchereit, J. Organomet. Chem., 1986, 307, C41.
10 H. Hoberg, S. Groß and A. Milchreit, Angew. Chem., 1987,
99, 567.
11 C. M. Willimas, J. B. Johnson and T. Rovis, J. Am. Chem.
Soc., 2008, 130, 14936.
12 T. Moragas, J. Cornella and R. Martin, J. Am. Chem. Soc.,
2014, 136, 17702.
and C. Krüger, Inorg. Chim. Acta, 1994, 222, 179;
(b) R. Benn, P. W. Jolly, T. Joswig, R. Mynott and
K.-P. Schick, Z. Naturforsch., B: Chem. Sci., 1986, 41, 680;
(c) R. Benn, P. Betz, R. Goddard, P. W. Jolly, N. Kokel,
C. Krüger and I. Topalović, Z. Naturforsch., B: Chem. Sci.,
1991, 46, 1395.
18 P. S. Schulz, O. Walter and E. Dinjus, Appl. Organomet.
Chem., 2005, 19, 1176.
19 T. Yamamoto, K. Sano, K. Osakada, S. Komiya,
A. Yamamoto, Y. Kushi and T. Tada, Organometallics, 1990,
9, 2396.
13 M. L. Lejkowski, R. Lindner, T. Kageyama, G. É. Bódizs,
P. N. Plessow, I. B. Müller, A. Schäfer, F. Rominger,
P. Hofmann, C. Futter, S. A. Schunk and M. Limbach, 20 I. Bach, R. Goddard, C. Kopiske, K. Seevogel and
Chem. – Eur. J., 2012, 18, 14017. K.-R. Pörschke, Organometallics, 1999, 18, 10.
14 C. Hendriksen, E. A. Pidko, G. Yang, B. Schäffner and 21 Software package SHELXTL 2008/4 for structure solution
D. Vogt, Chem. – Eur. J., 2014, 20, 12037.
15 N. Huguet, I. Jevtovikj, A. Gordillo, M. L. Lejkowski,
and refinement, see: G. M. Sheldrick, Acta Crystallogr., Sect.
A: Fundam. Crystallogr., 2008, 64, 112.
R. Linder, M. Bru, A. Y. Khalimon, F. Rominger, 22 Program SADABS 2008/1 for absorption correction, see:
S. A. Schunk, P. Hofmann and M. Limbach, Chem. – Eur. J.,
2014, 20, 1.
G. M. Sheldrick, Bruker Analytical X-ray Division, Madison,
Wisconsin, 2008.
16 H. Hoberg, D. Schaefer, G. Burkhart, C. Krüger and 23 L. Gonzáles-Sebastián, M. Flores-Alamo and J. J. García,
M. J. Romao, J. Organomet. Chem., 1984, 266, 203. Organometallics, 2012, 31, 8200.
11094 | Dalton Trans., 2015, 44, 11083–11094
This journal is © The Royal Society of Chemistry 2015