Reductive functionalization of carbon dioxide to methyl acrylate at zerovalent tungsten

Article abstract of DOI:10.1039/c2dt31032e

Alkali metal reduction of tungsten tetrachloride in the presence of excess trimethylphosphite and ethylene affords moderate yields of trans- tetrakis(trimethylphosphite)tungsten bis(ethylene). This easily prepared species bearing inexpensive ancillary lig

Full text of DOI:10.1039/c2dt31032e

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 10763
www.rsc.org/dalton
PAPER
Reductive functionalization of carbon dioxide to methyl acrylate at zerovalent
tungsten†
Justin M. Wolfe and Wesley H. Bernskoetter*
Received 11th May 2012, Accepted 24th June 2012
DOI: 10.1039/c2dt31032e
Alkali metal reduction of tungsten tetrachloride in the presence of excess trimethylphosphite and ethylene
affords moderate yields of trans-tetrakis(trimethylphosphite)tungsten bis(ethylene). This easily prepared
species bearing inexpensive ancillary ligands promotes the oxidative coupling of carbon dioxide and
ethylene at ambient temperature to produce two isomeric tetrakis(trimethylphosphite)tungsten acrylate
2
3
hydride complexes. These isomers vary by the κ -O,O and κ -C,C,O coordination mode of the acrylate
ligand, and swiftly interconvert in solution as detected by 2D NMR spectroscopy. The CO -derived
2
acrylate fragment may be released from the tungsten coordination sphere by treatment with methyl iodide
to afford modest quantities of free methyl acrylate.
Introduction
The growing scarcity and expense of petroleum resources has
motivated numerous endeavors to exploit renewable carbon
1
sources in the production of industrially desirable chemicals.
Among these efforts, carbon dioxide has emerged as a promising
small molecule carbon source owing to its incredible abundance,
2
relative ease of transport and low toxicity. Unfortunately, the
substantial stability of CO2 has thus far limited its thermo-
chemical utilization to a small set of commodities including
3
urea, salicylic acid and carbonates, despite the feasibility of
4
many other reductive functionalization reactions. The prospect
of reducing CO with light olefins to generate acrylates is one
2
2
Fig. 1 Reductions of CO with ethylene reported by Hoberg and
Carmona.
such example which has garnered attention since the early 1980s
when Hoberg and Carmona reported some of the first carbon
dioxide–ethylene couplings at zerovalent nickel and molyb-
5
8
denum species, respectively (Fig. 1).
(Fig. 2). The extrusion of functionalized CO
2
from the nickel
Acrylates are widely employed polar monomers integral to a
range of materials applications, such as super absorbent poly-
complex also represents one of the few cases where a late tran-
sition metallalactone has induced β-hydride elimination to create
9
mers, and thus present alluring targets for CO functionalization.
the salient CvC bond of acrylate. Our laboratory has recently
2
However, among the now numerous transition metal complexes
investigated acrylate formation mechanisms at group VI metal
observed to mediate C–C bond formation between CO2 and
centers which appear to engender more facile β-hydride elimin-
ation pathways. Herein we report the coupling of carbon
dioxide with ethylene at a simple tungsten(0) complex and the
subsequent extrusion of methyl acrylate derived from reductive
CO functionalization.
2
6
10
ethylene, only a select few are known to produce acrylate struc-
5
b,7
tures.
Furthermore, liberation of CO -derived acrylate
2
products from these metal species has seen limited success.
Carmona and co-workers have demonstrated that [(PMe ) -
3
2
(
C H )MoH(CO CHvCH )] may be deprotonated by butyl-
2 4 2 2 2
5b
lithium to afford lithium acrylate, and more recently Rieger
and others found adding large excesses of methyl iodide to nick-
elalactone complexes produces 29–56% yields of methyl acrylate
Experimental
General considerations
All manipulations were carried out using standard vacuum,
Schlenk, cannula or glovebox techniques. Ethylene and carbon
dioxide were purchased from Corp Brothers and stored over 4 Å
molecular sieves in heavy walled glass vessels prior to use.
Department of Chemistry, Brown University, Providence, RI 02912,
USA. E-mail: wb36@brown.edu
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c2dt31032e
1
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 10763–10768 | 10763

View Article Online
Fig. 2 Methyl iodide induced β-H elimination and methyl acrylate liberation from nickelalactones.
Argon and nitrogen were purchased from Corp Brothers and
mixture was stirred at ambient temperature for two days to afford
a pale orange solution. The volatiles were removed in vacuo and
the residue rinsed with hexamethyldisiloxane, extracted with
pentane and diethyl ether, and filtered through a Pasteur pipette.
Chilling the orange solution at −35 °C for several days afforded
0.061 g (41%) of 2-C H O as a beige powder. Anal. Calcd for
used as received. WCl was prepared according to literature pro-
4
11
cedures. All other chemicals were purchased from Aldrich,
VWR, Strem, Fisher Scientific or Cambridge Isotope
Laboratories. Volatile, liquid chemicals were dried over 4 Å
molecular sieves and distilled prior to use. Solvents were dried
3
3
2
1
2
and deoxygenated using literature procedures.
C H O P W: C, 23.95; H, 5.36. Found: C, 23.98; H, 4.93.
15 40 14 4
1
13
31
2
1
H, C, and P NMR spectra were recorded on Bruker DRX
00 Avance and 300 Avance MHz spectrometers. H and
C chemical shifts are referenced to residual solvent signals;
κ -2-C H O isomer: H NMR (23 °C, C D ): δ −8.13 (tt, 13.1,
3
3
2
6 6
1
4
68.0 Hz, W–H), 3.66 (virtual t, 4.9 Hz, 18H, P(OCH ) ), 3.69
3 3
1
3
3
(d, 10.9 Hz, 18H, P(OCH ) ), 5.15 (dd, 2.1, 10.4 Hz,
3
3
1
P chemical shifts are referenced to an external standard of
CO CHvCH ), 5.89 (dd, 10.4, 17.3 Hz, CO CHvCH ), 6.12
2 2 2 2
13
1
H PO . Probe temperatures were calibrated using ethylene glycol
(dd, 2.1, 17.4 Hz, CO CHvCH ). C{ H} NMR (23 °C,
2 2
3
4
1
3
and methanol as previously described. Unless otherwise noted,
all NMR spectra were recorded at 23 °C. IR spectra were
recorded on a Jasco 4100 FTIR spectrometer. GC-MS data were
recorded using a Hewlett-Packard (Agilent) GCD 1800C
GC-MS spectrometer. Elemental analyses were performed at
Robertson Microlit Laboratories, Inc., in Madison, NJ or Atlantic
Microlab, Inc., in Norcross, GA.
toluene-d ): δ 47.7, 48.2 (P(OCH ) ), 121.2 (CO CHvCH ),
8 3 3 2 2
3
1
1
128.6 (CO CHvCH ), 174.3 (CO CHvCH ). P{ H} NMR
2
2
2
2
2
1
(23 °C, C D ): δ 173.70 (t, J
= 76 Hz, 2P, J
= 486 Hz),
W–P
6
6
P–P
1
2
3
W–P 3 3 2
149.41 (t, J
= 76 Hz, 2P, J
= 486 Hz). κ -2-C H O
P–P
1
isomer: H NMR (23 °C, C D ): δ −4.47 (quintet, 1H, W–H),
2.21 (br, 1H, CO CHvCH ), 2.75 (br, 1H, CO CHvCH ),
6
6
2
2
2
2
4.38 (br, 1H, CO CHvCH ), 3.67 (br, 36H, P(OCH ) ).
2
2
3 3
1
H NMR (−60 °C, toluene-d ): δ −4.69 (m, 1H, W–H), 2.19 (br,
8
1
H, CO CHvCH ), 2.73 (br, 1H, CO CHvCH ), 4.33 (br dd,
2 2 2 2
Preparation of trans-[P(OMe) ] W(C H ) (2-C H )
3
4
2
4 2
2
4
8
.5, 8.7 Hz, 1H, CO CHvCH ), 3.09 (d, 9H, P(OCH ) ), 3.42
2
2
3 3
(
d, 9H, P(OCH ) ) 3.50 (d, 9H, P(OCH ) ) 3.59 (d, 9H,
A 100 mL heavy walled glass reaction vessel was charged with
.245 g (0.76 mmol) of WCl , 0.281 g (2.26 mmol) of P(OMe)3
and approximately 10 mL of tetrahydrofuran. The resulting
mixture was stirred at ambient temperature for two days. The
volatiles were removed to afford a dark solid presumed to be
3 3 3 3
1
13
P(OCH3)3).
C{ H} NMR (23 °C, toluene-d ): δ 35.0
8
0
4
(
(
4
1
CO CHvCH ), 48.6 (P(OCH ) ), 49.8 (CO CHvCH ), 179.0
2 2 3 3 2 2
1
31
CO CHvCH ). P{ H} NMR (23 °C, C D ): δ 147.47 (br s,
2 2 6 6
3
1
1
P). P{ H} NMR (−60 °C, toluene-d ): δ 141.02 (m, 1P),
8
42.05 (m, 1P), 143.95 (m, 1P), 146.13 (m, 1P). IR for both
(
P(OMe) ) WCl . The material was rinsed with pentane. To this
3 2 4
−1
isomers (KBr): v
1
= 1638, 1618, 1367 cm ; v
= 1621,
13C–O
solid 15 mL tetrahydrofuran, 24.0 g (5.23 mmol) of 0.5%
C–O
−1
587, 1355 cm .
sodium amalgam and 0.216 g (1.74 mmol) of P(OMe) were
3
added. On a vacuum line approximately 20 equiv of ethylene
were admitted at −196 °C. The reaction mixture was stirred at
ambient temperature for two days. The volatiles were removed
in vacuo from the dark solution. The residue was extracted
with approximately 15 mL of diethyl ether and filtered through
Celite. Removal of the ether and washing with 5 mL of
hexamethyldisiloxane afforded 0.307 g (55%) of 2-C H as a
3 3 2
Treatment of 2-C H O with iodomethane
A J. Young NMR tube was charged with 0.020 g (0.027 mmol)
of 2-C H O and approximately 400 μL of benzene-d . On a
3
3
2
6
vacuum line, 5 equiv of methyl iodide were admitted at −196 °C
via a calibrated gas bulb (85 torr in 28.9 mL). The reaction
mixture was thawed, shaken and allowed to sit at ambient temp-
erature for 2 days. The conversion of 2-C H O and methyl
2
4
light brown solid of good purity. Further purification may be
conducted by recrystallization from pentane to yield nearly
colorless flakes. Anal. Calcd for C H O P W: C, 26.10; H,
3
3
1
2
iodide to methyl acrylate was monitored by H NMR spec-
troscopy versus an internal standard of hexamethyldisiloxane
which indicated a yield of 31%. The identity of methyl acrylate
1
6 44 12 4
1
6
.02. Found: C, 26.55; H, 5.76. H NMR (23 °C, C D ): δ 3.50
6 6
(
br s, 36H, P(OCH ) ), 1.97 (quintet, 5.8 Hz, 8H, C H ).
3
3
2 4
1
3
1
1
1
C{ H} NMR (23 °C, C D ): δ 18.2 (C H ), 51.4 (P(OCH ) ).
was confirmed by comparison to H NMR spectra of authentic
6
6
2
4
3 3
3
1
1
P{ H} NMR (23 °C, C D ): δ 133.2 (s, J
= 411 Hz).
samples and by GC-MS analysis of the reaction volatiles which
afforded a m/z peak of 85 at a retention time identical to that of
an authentic sample.
6
6
W–P
Preparation of [P(OMe) ] W(CO CHvCH )H (2-C H O )
3
4
2
2
3
3
2
A 100 mL heavy-walled glass reaction vessel was charged with
.142 g (0.193 mmol) of 2-C H and approximately 10 mL of
Results and discussion
0
2
4
diethyl ether. On a vacuum line approximately 20 equiv of CO2
Within the relatively few examples of CO –ethylene coupling to
2
were admitted to the vessel at −196 °C. The resulting reaction
acrylates at transition metals, zerovalent molybdenum and
10764 | Dalton Trans., 2012, 41, 10763–10768
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 3 Oxidative coupling of carbon dioxide and ethylene at molyb-
denum(0) supported by trimethylphosphite ligands.
Fig. 4 Preparation of tetrakis(trimethylphosphite)tungsten bis-
(ethylene).
tungsten complexes have afforded some of the most promising
5
b,7,10
examples reported thus far.
However, most of these com-
plexes either require expensive phosphine ligands to stabilize the
low-valent metal or produce functionalized carbon dioxide com-
plexes in rather low total yields from the commercially available
precursors. This has directed our efforts toward identifying
easily prepared group VI metal complexes supported by simple,
inexpensive donor ligands which could serve as more practical
carbon dioxide products were isolated together as a beige
powder from pentane or diethyl ether in modest yield, and were
consistently obtained in an approximate 2 : 1 ratio over multiple
2
synthetic trials, with [P(OMe)
]
3
W(H)(CO
CHvCH
-κ -O,O)
4
2
2
as the major species. The persistence of two isomeric forms of
2-C hindered our ability to obtain single crystals of suit-
H O
3 2
3
platforms to study the reductive functionalization of CO to acry-
able quality for X-ray diffraction, and suggested that the two
acrylate species may interconvert in solution. This prompted a
series of isotopic labeling studies, infrared analyses and multi-
dimensional/variable temperature NMR experiments to confirm
2
lates and methodologies for extrusion of the acrylate fragment.
One complex which shares some of these desired features is
trans-[P(OMe) ] Mo(C H ) (1-C H ) (Fig. 3) which was pre-
3
4
2
4 2
2
4
7
b
2
3
viously described by Galindo and co-workers in 1998.
Encouragingly, complex 1-C H mediates acrylate ligand for-
the equilibration of the κ and κ coordination modes and verify
the structural assignments.
2
4
1
31
mation upon treatment with carbon dioxide, but was itself syn-
thesized in relatively low yield (∼15%). We sought to leverage
this attractive, inexpensive ancillary ligand by employing a
tungsten metal center which should provide enhanced reduction
potential for functionalizing carbon dioxide, possibly afford
zerovalent metal ethylene compounds in greater yields and con-
veniently accommodate further study of acrylate liberation from
the metal.
The ambient temperature H and P NMR spectra of
2-C display multiple signals as expected for a mixture
H O
3 2
3
of two tungsten acrylate hydride species. In the upfield region of
1
the H NMR spectrum an apparent triplet of triplets resonance
2
was observed at −8.13 ppm ( JP–H = 13.1, 68.0 Hz) consistent
2
with the metal hydride of κ -2-C
H
3
O
coupling to two pairs of
3
2
phosphorus nuclei (Fig. 6). Additionally, three downfield vinylic
protons were observed, resonating at 5.15, 5.89, and 6.12 ppm,
similar to the signals of free acrylic acid. A corresponding pair
of triplet peaks observed at 173.70 and 149.41 ppm in the
The tungsten congener of 1-C H was prepared by alkali
2
4
metal reduction of WCl in the presence of trimethylphosphite
4
3
1
and excess ethylene (Fig. 4). Best results were obtained when
P NMR spectrum is also consistent with the proposed structure
2
WCl was treated with trimethylphoshite in tetrahydrofuran to
of κ -2-C
H O , and is analogous to the report of the isostruc-
3 3 2
4
7
b
generate [P(OMe) ] WCl prior to addition of sodium amalgam,
tural molybdenum congener. In addition to these NMR signals
for κ -2-C , a second set of resonances were evident for
the κ -2-C H O isomer. These included vinylic resonances
3 3 2
3
2
4
2
ethylene and more phosphite. Extraction of the reduction
3
H
3
O
2
3
mixture with diethyl ether afforded trans-[P(OMe) ] W(C H )
2 4 2
3
4
(
2-C H ) in moderate yield. Complex 2-C H was characterized
observed at 2.13, 2.23, and 2.77 ppm and a metal hydride at
−4.47 ppm. H COSY NMR spectra indicated coupling between
2
4
2
4
1
by a combination of combustion analysis and multinuclear NMR
spectroscopy. H NMR spectra of the bis(ethylene) species
exhibited a quintet signal at 1.97 ppm ( J
1
these vinylic peaks and the tungsten hydride, confirming that
these resonances correspond to one isomer. The upfield shifting
of the vinylic signals is consistent with olefin coordination to the
3
= 5.8 Hz) assigned
P–H
to the protons of the bound olefin, with a corresponding
1
3
1
13
15
3
C NMR resonance located at 18.2 ppm (confirmed by H– C
metal. Notably, the κ -2-C
H O resonances were all distinctly
3 3 2
1
13
2
HSQC NMR). Both these H and C chemical shift values are
typical of ethylene bound to zerovalent tungsten, though
slightly upfield of those reported for the molybdenum analogue,
1
broadened with respect to those of the κ -2-C
H
3
3
O
2
isomer,
may undergo an intramolecular
process which interchanges the four P(OMe) ligands
7
a
3
suggesting that κ -2-C H O
3 3 2
3
7
b
-C H .
(vide infra). This hypothesis is supported by the quintet splitting
2
4
2
Treatment of ethereal or hydrocarbon solutions of 2-C H4
pattern ( JP–H = 49.0 Hz) exhibited by the W–H signal and a
2
31
with 1 atm of carbon dioxide over two days resulted in
complete conversion to a mixture of two tungsten acrylate
hydride isomers, [P(OMe) ] W(H)(CO CHvCH -κ -O,O) and
broad singlet at 147.47 ppm observed in the P NMR spectrum
(Fig. 6).
2
Variable temperature and two-dimensional exchange NMR
spectroscopy were employed to assess the intra- and intermole-
3
4
2
2
3
[P(OMe) ] W(H)(CO CHvCH -κ -C,C,O) (collectively denoted
3 4 2 2
14
2
3
2
-C H O ) (Fig. 5). The coupling reaction with 2-C H pro-
cular dynamics of κ -2-C
3
H
3
O
2
and κ -2-C
3
H
O
3
2
in solution.
3
3
2
2
4
1
ceeded with no observable intermediates and with a qualitatively
H NOESY NMR experiments (27 °C; mixing time 350 ms)
slower rate than 1-C H , suggesting a sluggish ligand dis-
sociation may be key to starting the reaction. The functionalized
exhibited exchange correlations between the two isomers’ metal
hydride and olefinic resonances. Additional exchange between
2
4
16
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 10763–10768 | 10765

View Article Online
Fig. 5 CO
2 3 3 2
–ethylene coupling at tungsten and isomerization of 2-C H O .
1
Fig. 6 Partial H NMR spectrum of 2-C
3
H
3
O
2
.
the P(OMe) signals was anticipated, but could not be resolved
Confirmation of the incorporation of carbon dioxide into
2-C H O was obtained by C isotopic labeling studies. Treat-
3 3 2
3
13
due poor signal separation. The observations clearly establish an
2
3
isomerization between κ -2-C H O and κ -2-C H O at a rate
ment of a benzene solution of 2-C H with one atmosphere of
2 4
3
3
2
3
3
2
1
3
13
slightly slower than the 1D NMR acquisition timescale. Acquir-
ing 1D NMR spectra at lower temperatures in toluene-d8
CO afforded two enhanced resonances in the C NMR spec-
2
3
trum at 179.0 and 174.3 ppm assigned to κ -2-C H O and
3
3
2
3
2
resolved and sharpened some of the resonances of κ -2-C H O
κ -2-C H O , respectively. These assignments were established
3 3 2
3
3
15
2
2
1
13
while leaving those of κ -2-C H O essentially unchanged. At
temperatures around −20 °C, the peaks of κ -2-C H O appear
to undergo decoalescence, distorting the signals into broad, near-
featureless peaks in the H and P NMR spectra. At −60 °C,
the resonances of κ -2-C H O sharpen considerably, and
by a H– C HMBC NMR spectrum which showed correlation
3
3
2
3
between the labeled carbons, the respective metal hydride and
3
3
2
13
vinylic resonances. Incorporation of the C labels also altered to
the bands in the infrared spectrum, shifting signals at 1638, 1618
1
31
3
−1
and 1367 cm observed for unlabeled 2-C H O to 1621, 1587
3
3
2
3
3
2
−1
exhibit four separate doublet peaks in the P(OMe) region of the
and 1355 cm , respectively. The intense bands in the
1600–1650 cm region suggest that the carboxylates are not
bound in bridging modes within dimeric structures. Addition-
3
1
−1
H NMR spectrum, confirming that all four phosphite ligands
17
are inequivalent in the static structure. The concurrent obser-
3
1
vation of four multiplet signals in the P NMR spectrum sup-
ports this characterization. Additionally, the tungsten hydride
ally, the presence of four P(OMe) ligands about each metal (as
3
1
31
indicated by H and P NMR spectroscopy) strongly implies
1
signal observed in the H NMR spectrum devolves into an un-
that 2-C H O is monomeric as it is coordinately saturated.
3
3
2
resolved multiplet, as may be anticipated for coupling with four
inequivalent phosphorus nuclei with similar chemical shifts.
Although a more detailed mechanism of the intramolecular
The monomeric aggregation state of the molybdenum and
tungsten acrylate species 1-C H O and 2-C H O differ from
3
3
2
3
3
2
most of the other of group VI acrylate complexes reported,
which typically adopt a dimeric form with a bridging acrylate
3
dynamic process in κ -2-C H O cannot be made solely from
3
3
2
5
b,7a,10
the variable temperature NMR data, a rapid site change of the
ligand.
The preference for the monomer may originate
P(OMe) ligands does appear to be the origin of its broadened
from the modest π-acid character of P(OMe) . The PMe -
3
3
3
spectral features.
supported molybdenum and tungsten acrylate species have both
10766 | Dalton Trans., 2012, 41, 10763–10768
This journal is © The Royal Society of Chemistry 2012

View Article Online
3 3 2
Fig. 7 Liberation of methyl acrylate from 2-C H O using methyl iodide.
5
b
been crystallographically characterized as dimeric structures.
methyl acrylate by carbon monoxide addition did not afford
additional free acrylate. Similar attempts to liberate the carboxy-
late fragment by addition of alternative electrophiles including
methyl triflate and trimethylsilyl iodide to 2-C H O did not
afford detectable quantities ( H NMR) of free acrylate. Although
limited in yield, this example of obtaining free acrylate products
The modest back-bonding with P(OMe) should strengthen the
3
metal–ligand bond, preventing its loss (as observed for PMe3;
Fig. 1) and thus blocking access to a coordinatively unsaturated
species necessary for a bridging acrylate. The modest steric per-
turbation of switching from the smaller PMe3 to the larger
3
3
2
1
18
P(OMe) ligand may also influence the observed differences.
from CO –ethylene coupling is one of rather few examples.
3
2
Given the parallel in aggregation state of 1-C H O and
3
3
2
2
-C H O , the subtle variation in acrylate coordination mode
3 3 2
between these two congeners was somewhat surprising. The
spectroscopic data indicate that the tungsten acrylate complex
Concluding remarks
2
can adopt a κ -coordination mode identical to that described for
The direct coupling of carbon dioxide and ethylene at a tran-
sition metal to afford acrylate remains a challenging transform-
ation, but one which offers promise for the chemical reduction
3
molybdenum; however, an alternative κ -coordination mode is
sufficiently near in energy to permit its observation as the minor
3
isomer of 2-C H O . In general, the κ -acrylate coordination
of CO to commodity chemicals. To date, most platforms which
3
3
2
2
geometry is less common for transition metals compared to the
are known to mediate acrylate formation on group VI metals are
somewhat limited from broader exploration by the poor yields of
the necessary organometallic species or by the cost of the ancil-
lary ligands. Since considerable advancements are still needed to
master both acrylate formation at transition metals and their
reductive removal to complete catalytic cycles, we have sought
inexpensive, easily prepared acrylate complexes which are
2
7a,19
κ -mode.
Its presence here may originate from a competing
preference to bind the π-acidic olefin ligand at the more reducing
tungsten center.
Following the successful preparation of a CO –ethylene
2
derived tungsten acrylate complex, our laboratory was eager to
employ this easily prepared complex in studies to liberate a
desirable acrylate product from the metal center. Given the
strong oxophilicity of tungsten, this presented a considerable
challenge. Fortunately, addition of a small excess of methyl
iodide to the mixture of the 2-C H O isomers afforded quan-
obtained from CO –ethylene coupling. Reported here is one
2
such tungsten complex supported by trimethyphosphite ligands
which functionalizes CO to acrylate. In solution, the acrylate
2
3
3
2
hydride complex 2-C H O quickly interconverts between two
3
3
2
tities of methyl acrylate (Fig. 7). The yield of free methyl acry-
late ( judged by H NMR) was comparable to those achieved by
isomeric forms as judged by 2D NMR spectroscopy. This equili-
1
2
3
bration affords access to both the κ -O,O and κ -C,C,O mono-
meric acrylate coordination modes, in contrast to observations
made with the exact molybdenum congener as well as the
closely related trimethylphosphine–tungsten counterpart. These
finding suggests that the electrophilicity of the complex imparted
by metal and ligand selection may be used to attenuate both the
aggregation state and coordination mode of the functionalized
nickelalactone complexes, but under more mild additions of
8
electrophile. Addition of greater excesses of methyl iodide did
not significantly alter or improve the yield of acrylate product.
As with the liberation for methyl acrylate from the nickel com-
plexes, the tungsten species convert into an intractable mixture
of organometallic products, including at least three metal
1
hydride-containing species as observed by H NMR spec-
CO fragment. Such features should be of interest in the removal
2
20
troscopy. The formation of methyl acrylate was confirmed by
of acrylate products. Using 2-C H O , free methyl acrylate
3
3
2
1
3
13
GC-MS and C isotopic labeling, using both CO2 and
derived from CO –ethylene coupling was obtained by treatment
2
1
3
13
CH I. Interestingly, while the doubly C-labeled experiment
produced enhanced peaks in the C NMR spectrum for methyl
acrylate (doublets at 166.4 and 51.4 ppm; J
with methyl iodide. While the yields of functionalized CO2
remain modest, the tungsten acrylate hydride species affords
methyl acrylate in comparable quantities and with considerably
milder additions of electrophiles than those obtained using late
3
1
3
2
= 2.6 Hz), it
C–C
also produced a second set of doublet resonances of comparable
2
15
intensity at 174.2 and 56.6 ppm ( J
= 2.8 Hz). The simi-
metal CO activation agents. The electrophilic addition in this
C–C
2
larity of this second set of resonances to free methyl acrylate and
the observation of C–C coupling suggests that one of the tung-
sten byproducts retains a methyl acrylate fragment, and the
actual yield of methyl acrylate formed during electrophilic
addition may be substantially higher than the 31% observed as
free organic. However, attempts to liberate this second portion of
case as much slower than acrylate isomer interconversion,
obscuring the role of acrylate coordination mode. In on-going
investigations, our laboratory seeks to employ this readily acces-
sible tungsten acrylate hydride complex in efforts to synthesize
other CO -derived acrylate products, identify the role that acry-
2
late coordination mode may play in product removal, as well as
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 10763–10768 | 10767

View Article Online
study reductive acrylate elimination techniques which may ulti-
E. Gutierrez-Puebla, A. Monge, M. L. Poveda and C. Ruiz, J. Am. Chem.
Soc., 1985, 107, 5529.
(a) S. A. Cohen and J. E. Bercaw, Organometallics, 1985, 4, 1006;
mately yield new catalytic CO fixation methods.
2
6
(b) H. Hoberg, Y. Peres, C. Kruger and Y. H. Tsay, Angew. Chem., Int.
Ed. Engl., 1987, 26, 771; (c) H. G. Alt and C. E. Denner, J. Organomet.
Chem., 1990, 390, 53; (d) B. Hessen, A. Meetsma, F. Bolhuis, J. Teuben,
G. Helgesson and S. Janger, Organometallics, 1990, 9, 1925;
Acknowledgements
We gratefully acknowledge funding by the Air Force Office of
Scientific Research (Award No. FA9550-11-1-0041), the U.S.
Department of Energy-National Energy Technology Laboratory
(e) H. Hoberg and A. Ballesteros, J. Organomet. Chem., 1991, 411, C11;
(f) H. Hoberg, A. Ballesteros, A. Sigan, C. Jegat, D. Barhausen and
A. Milchereit, J. Organomet. Chem., 1991, 407, C23; (g) M. Aresta and
E. Quaranta, J. Organomet. Chem., 1993, 463, 215.
(a) A. Galindo, A. Pastor, P. Perez and E. Carmona, Organometallics,
1993, 12, 4443; (b) C. Collazo, M. del Mar Conejo, A. Pastor and
A. Galindo, Inorg. Chim. Acta, 1998, 272, 125.
(a) C. Bruckmeier, M. W. Lehenmeier, R. Reichardt, S. Vagin and
B. Rieger, Organometallics, 2010, 29, 2199; (b) S. Y. T. Lee, M. Cokoja,
M. Drees, Y. Li, J. Mink, W. A. Herrmann and F. E. Kuhn,
ChemSusChem, 2011, 4, 1275.
(Award No. DE-FE0004498) and Brown University. J. W. also
7
8
thanks Brown University for a Summer Undergraduate Teaching
and Research Award.
References
9
R. Fischer, J. Langer, G. Malassa, D. Walther, H. Gorls and G. Vaughan,
Chem. Commun., 2006, 2510.
1
(a) Chemicals and Materials from Renewable Resources, ed. J. Bozell,
ACS Symposium Series 784, American Chemical Society, Washington,
DC, 2001; (b) Feedstocks for the Future, ed. J. Bozell and M. K. Patel,
ACS Symposium Series 921, American Chemical Society, Washington,
DC, 2006; (c) P. N. R. Vennestrom, C. M. Osmundsen, C. H. Christensen
and E. Taarning, Angew. Chem., Int. Ed., 2011, 50, 10502; (d) Renewable
Raw Materials: New Feedstocks for the Chemical Industry, ed. R. Ulber,
D. Sell and T. Hirth, Wiley-VCH, Weinheim, 2011.
(a) A. Behr, Angew. Chem., Int. Ed. Engl., 1988, 27, 661; (b) M. Aresta,
Carbon dioxide reduction and uses as a chemical feedstock, in Activation
of Small Molecules: Organometallic and Bioinorganic Perspectives, ed.
W. B. Tolman, Wiley-VCH, Weinheim, 2006, pp. 1–35;
1
0 W. H. Bernskoetter and B. T. Tyler, Organometallics, 2011, 30, 520.
1 R. R. Schrock, L. G. Sturgeoff and P. R. Sharp, Inorg. Chem., 1983, 22,
801.
1
1
1
1
2
2 A. B. Pangborn, M. A. Giardello, R. H. Grubbs, R. K. Rosen and
F. J. Timmers, Organometallics, 1996, 15, 1518.
3 J. Sandström, Dynamic NMR Spectroscopy, Academic Press, New York,
982.
4 The [P(OMe)
tively be named [P(OMe)
1
2
3
3
3
]
4
W(H)(CO
2
CHvCH
2
-κ -C,C,O) species may alterna-
1 2
3
]
4
W(H)(CO CHvCH -κ -O-η -C,C). The
2 2
former listing is used here for ease in comparing the two isomeric
products in subsequent discussions.
(c) E. A. Quadrelli, G. Centi, J.-L. Duplan and S. Perathoner,
15 F. P. Pruchnik, Organometallic Chemistry of the Transition Elements,
ChemSusChem, 2011, 4, 1194.
(a) P. Braunstein, D. Matt and D. Nobel, Chem. Rev., 1988, 88, 747;
Plenum Press, New York, 1990.
1
1
6 See ESI.†
(
(
b) T. Sakakura, J.-C. Choi and H. Yasuda, Chem. Rev., 2007, 107, 2365;
c) M. Aresta and A. Dibenedetto, Industrial utilization of carbon
7 K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordi-
nation Compounds, Wiley, New York, 4th edn, 1986, pp. 231–233.
8 For updated Tolman cone angle values see: (a) D. White and N. Coville,
Adv. Organomet. Chem., 1994, 36, 95; (b) L. Stahl, W. Trakarnpruk,
J. W. Freeman, A. M. Arif and R. D. Ernst, Inorg. Chem., 1995, 34,
dioxide, in Developments and Innovation in Carbon Dioxide Capture
and Storage Technology, ed. M. M. Maroto-Valer, Woodhead, Cambridge,
1
2
010, pp. 377–410.
4
5
M. Aresta and A. Dibenedetto, Dalton Trans., 2007, 2975; M. Aresta and
A. Dibenedetto, Catal. Today, 2004, 98, 455; Y. Patil, P. J. Tambade,
S. R. Jagtap and B. M. Bhanage, Front. Chem. Eng. China, 2009, 4, 213;
M. Cokoja, C. Bruckmeier, B. Rieger, W. A. Herrmann and F. E. Kuhn,
Angew. Chem., Int. Ed., 2011, 50, 8510.
1810.
1
2
9 U. Koelle, B.-S. Kang and U. Thewalt, Organometallics, 1992, 11, 2893.
0 While the identity of all tungsten products in the mixture cannot be
confirmed do to the inability to separate them, it is presumed based on
the diamagnetism and the observation of metal hydrides that these are
primarily W(II) and/or W(IV) species.
(a) H. Hoberg and D. Schaefer, J. Organomet. Chem., 1983, 251, c51;
(b) R. Alvarez, E. Carmona, D. J. Cole-Hamilton, A. Galindo,
10768 | Dalton Trans., 2012, 41, 10763–10768
This journal is © The Royal Society of Chemistry 2012