Mechanistic Investigations of the Ruthenium-Catalyzed Synthesis of Acrylate Salt from Ethylene and CO2

Article abstract of DOI:10.1021/acs.organomet.9b00659

Here we describe detailed mechanistic analyses of the ruthenium-catalyzed synthesis of acrylate salt from ethylene and CO₂. All of the primary steps of the catalytic cycle, that is (i) oxidative cyclization of ethylene and CO₂ to give ruthenalactones, (ii) thermal β-H elimination of ruthenalactones to hydrido acrylato complexes, and (iii) base-mediated generation of ethylene-coordinated, zerovalent ruthenium complexes from hydrido acrylato complexes, are found to proceed more rapidly on the more electron rich complexes. These results suggest that the last two processes proceed through the cationic ruthenium complexes generated by dissociation of carboxylate anions. As another possible pathway for the catalytic acrylate synthesis, base-promoted cleavage of a ruthenalactone to an ethylene-coordinated, zerovalent ruthenium complex is also studied, and the reaction is found to proceed smoothly, where the rate of the reaction is dependent on the basicity of the base. In contrast to the stoichiometric reactivities, electron-deficient ruthenium complexes exhibited higher activity in the catalytic acrylate synthesis.

Full text of DOI:10.1021/acs.organomet.9b00659

pubs.acs.org/Organometallics
Article
Mechanistic Investigations of the Ruthenium-Catalyzed Synthesis of
Acrylate Salt from Ethylene and CO2
Kohei Takahashi, Yo Hirataka, Tatsuyoshi Ito, and Nobuharu Iwasawa*
Cite This: https://dx.doi.org/10.1021/acs.organomet.9b00659
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Here we describe detailed mechanistic analyses of
the ruthenium-catalyzed synthesis of acrylate salt from ethylene
and CO2. All of the primary steps of the catalytic cycle, that is (i)
oxidative cyclization of ethylene and CO2 to give ruthenalactones,
(ii) thermal β-H elimination of ruthenalactones to hydrido acrylato
complexes, and (iii) base-mediated generation of ethylene-
coordinated, zerovalent ruthenium complexes from hydrido
acrylato complexes, are found to proceed more rapidly on the
more electron rich complexes. These results suggest that the last
two processes proceed through the cationic ruthenium complexes
generated by dissociation of carboxylate anions. As another
possible pathway for the catalytic acrylate synthesis, base-promoted
cleavage of a ruthenalactone to an ethylene-coordinated, zerovalent ruthenium complex is also studied, and the reaction is found to
proceed smoothly, where the rate of the reaction is dependent on the basicity of the base. In contrast to the stoichiometric
reactivities, electron-deficient ruthenium complexes exhibited higher activity in the catalytic acrylate synthesis.
Scheme 1. Proposed Mechanisms for the Acrylate
Formation from Ethylene and CO2
INTRODUCTION
■
3,4,6
Oxidative cyclization of ethylene and CO2 has attracted
considerable attention as a promising method to prepare
acrylic acid from abundant carbon resources.1 Hoberg et al.
reported a pioneering work that the oxidative cyclization of
alkenes and CO2 took place on zerovalent nickel complexes
bearing a bipyridine or a bisphosphine ligand to give five-
membered nickelalactones.2 On the basis of that work, many
studies on stoichiometric reactions of nickelalactones3 have
been carried out and acrylate salt formation has been noted in
several cases.3d−f More recently, the catalytic synthesis of
acrylate salts has been realized utilizing Ni4 and Pd4d,f,5
complexes, and theoretical analyses of the reaction pathways
have also been carried out to clarify the mechanism of the
reaction.6 The simplified two major proposed catalytic cycles
are shown in Scheme 1. As a common step, oxidative
cyclization of ethylene and CO2 takes place on a zerovalent
complex to give a metallalactone (step A). From here, two
pathways are proposed. The first one is the β-H elimination of
the metallalactone to a hydrido acrylato complex (step B, cycle
A), which further undergoes base-promoted reductive
elimination of acrylic acid salt with regeneration of the
zerovalent, ethylene-coordinated complex (step C, cycle A).
The other pathway is the direct deprotonation of the
metallalactone by a base to an acrylate-coordinated complex
(step B′, cycle B), which further undergoes ligand exchange
with ethylene to give an acrylic acid salt with regeneration of
the zerovalent complex (step C′, cycle B).
With regard to the nickel-based systems, the most
problematic step is the cleavage of the five-membered
metallalactone (step B or B′), and an additional reagent such
as a Lewis acid,3d,e methylating agent,3a−c,g phosphine ligand,3f
or base4a has commonly been employed to promote cleavage
of nickelalactones. On the other hand, metallalactones of group
6 metals are known to undergo β-H elimination easily.7 For
example, a zerovalent Mo complex, Mo(PMe3)4(ethylene)2,
Special Issue: Organometallic Chemistry for Enabling
Carbon Dioxide Utilization
Received: October 6, 2019
https://dx.doi.org/10.1021/acs.organomet.9b00659
© XXXX American Chemical Society
Organometallics XXXX, XXX, XXX−XXX
A

Organometallics
pubs.acs.org/Organometallics
Article
undergoes oxidative cyclization of ethylene and CO2 to afford a
molybdalactone, which spontaneously undergoes β-H elimi-
nation to give a hydrido acrylato complex.7a This is an
interesting result which might lead to a highly active catalyst
for acrylic acid synthesis; however, the catalytic reaction has
not been realized with group 6 metal based systems due to the
difficulty of regenerating the zerovalent complex. There have
also been several reports of the oxidative cyclization of
ethylene and CO2 utilizing other metals, but none of them
produced acrylate salt catalytically.8−12
Recently, with the expectation that a phosphine-coordinated
Ru(0) complex would have sufficient reducing ability due to
the high electron-donating ability of the ligand and
regeneration of Ru(0) would be more facile than that of
zerovalent group 6 metals, we examined the use of a Ru
complex for this reaction and found that a zerovalent Ru
complex bearing a tetradentate phosphine ligand smoothly
underwent oxidative cyclization of ethylene and CO2 to afford
a ruthenalactone as the first example of utilizing ruthenium for
this type of reaction (step A, Scheme 2a).13 Importantly, the
based system, where the key intermediates (zerovalent
complex, ruthenalactone, and hydrido acrylato complex) are
isolable and each step of the catalytic cycle could be followed
by stoichiometric reactions as shown in Scheme 2a.
RESULTS AND DISCUSSION
■
Design, Synthesis, and Properties of New Ruthenium
Complexes. In order to investigate the electronic and steric
effects of the ligand on each step of the catalytic cycle, four
kinds of zerovalent, ethylene-coordinated ruthenium com-
plexes RuL1_E−RuL4_E, including the previously investigated
RuL3_E, bearing tetradentate phosphine ligands L1,15 L2, L3,16
and L417 were selected (Figure 1). We expected that L1−L4
Scheme 2. Ruthenium-Catalyzed Synthesis of Acrylate Salt
Figure 1. Ru complexes and ligands investigated in this study.
would exhibit different properties because the electron-
donating ability of an alkyl-substituted phosphorus atom is
higher than that of an aryl-substituted atom and L1 and L3,
which have ethylene tethers, are more flexible than L2 and L4
with phenylene tethers.
The zerovalent, ethylene-coordinated complexes RuL1_E
−
RuL4_E were prepared from RuH(OAc)(PPh3)3 in two steps
according to our previously reported method (Scheme 3).13
First, RuH(OAc)(PPh3)3 was reacted with the corresponding
tetradentate phosphine ligand to give the hydrido acetato
complexes RuL_OAc (Scheme 3a, 38−87% yield). Then they
were treated with KOtBu under an ethylene atmosphere to
afford the corresponding zerovalent, ethylene-coordinated
ruthenium complexes RuL_E in good yields (Scheme 3b, 62−
cleavage of the ruthenalactone was found to proceed simply by
heating without any additives to give a hydrido acrylato
complex and this process turned out to be reversible (step B,
Scheme 2a).14 Regeneration of the zerovalent complex was
realized by the reaction of the hydrido acrylato complex with
KOEt under an ethylene atmosphere (step C, Scheme 2a). It
was also found that the complex catalyzes the acrylate salt
synthesis although the TON was not very high (Scheme 2b).
This result suggests that the cleavage of a ruthenalactone
proceeds more smoothly in comparison to nickelalactones.
Also, this system is the first example of directly observing
conversion of a metallalactone to a hydrido acrylato complex
without using an additive, which would be advantageous for
the mechanistic study of the conversion of metallalactones.
In the present study, in order to improve the catalytic
activity of the ruthenium system, we performed detailed
mechanistic investigations for the conversion of ethylene and
CO2 into acrylate salts using various zerovalent ruthenium
complexes. We have focused on the effect of the ligand
structure on each step of the catalytic cycle, especially on the
mechanism of the cleavage of the ruthenalactone. These
studies are realized by utilizing the feature of the ruthenium-
Scheme 3. Synthesis of Ru(ethylene) Complexes Bearing a
Tetradentate Phosphine Ligand
B
https://dx.doi.org/10.1021/acs.organomet.9b00659
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
82% yield). The structures of these complexes were confirmed
by the similarity of NMR spectra with those of RuL3_E, whose
structure was established by X-ray analysis.13 This protocol is
practically useful for the synthesis of these types of electron-
rich, zerovalent ruthenium complexes because the conventional
methods required strong reductants such as metallic Na and
LiAlH4 for the reduction of Ru(II) complexes.18
To evaluate the electron density of the ruthenium centers of
RuL1_E−RuL4_E,
13C NMR chemical shifts of the coordinated
ethylene were compared (Figure 2). The observed chemical
Figure 2. Evaluation of the electron density on the ruthenium center
of RuL_E on the basis of their 13C NMR chemical shifts of coordinated
ethylene.
shifts of the carbon atom of the coordinated ethylene were as
follows: RuL1_E, 18.6 ppm; RuL2_E, 22.1 ppm; RuL3_E, 34.8
ppm; RuL4_E, 43.1 ppm. Since a higher field shift indicates
stronger metallacyclopropane character due to the greater
back-donation, the electron density on the ruthenium center is
19
thought to be RuL1_E > RuL2_E > RuL3_E > RuL4_E
.
Accordingly, the electron-donating ability of tetradentate
phosphine ligands should be L1 > L2 > L3 > L4, as expected.
Although an evaluation of the steric hindrance of these ligands
is not necessarily easy, it is roughly estimated to be L1 ≈ L2 >
L3 ≈ L4 on the basis of the following two assumptions: (i) Cy
and iPr groups are larger than a Ph group,20 and (ii) the type of
ligand backbone (−CH2CH2− or −C6H4−) does not make a
significant difference.21
Evaluation of the Ligand Effect on the Reactivity of
Zerovalent Ruthenium Complexes with Ethylene and
CO2 (Steps A and B in Cycle A). With these complexes in
hand, the effect of ligand structure on the oxidative cyclization
of ethylene and CO2 was investigated first (step A in Scheme
1). A DMA or THF/DMA (1/1)22 solution of each of the
ethylene-coordinated complexes RuL_E was heated at 60 °C
under an atmospheric pressure of ethylene and CO2 (1/1), and
the progress of the reaction was monitored by 31P NMR
spectroscopy (Figure 3). As already described in our previous
report,13 RuL3_E was slowly converted to ruthenalactone
RuL3_lac (completed after 15 h, as indicated with △ in Figure
3a).23 On the other hand, the reaction of the electron-rich
complex RuL1_E was much more rapid (completed in 1 h) than
that of RuL3_E, and more importantly, the hydrido acrylato
complex RuL1_acr was obtained as a single major product and
ruthenalactone RuL1_lac was not observed at all (the amount of
RuL1_acr is indicated with ○ in Figure 3b). Similarly, RuL2_E
was cleanly converted to the hydrido acrylato complex RuL2_acr
Figure 3. Progress of the reaction of RuL_E with ethylene/CO2: (a)
△
conversion of RuL3_E to RuL3_lac
conversion of RuL1_E and RuL2_E to RuL1_acr and RuL2_acr ( ,
(
, RuL3_lac) in DMA; (b)
○
RuL1_acr; ×, RuL2_acr) in THF/DMA (1/1).
acrylate salts. In contrast, the reaction of the electron-poor
complex RuL4_E did not take place at 60 °C. From these
results, it is concluded that the initial reaction rate of oxidative
24
cyclization is RuL1_E > RuL2_E > RuL3_E > RuL4_E
,
which
with a lower reaction rate (completed after 9 h) than RuL1_E
,
coincides with the electron-donating ability of the ligands (L1
> L2 > L3 > L4). It is reasonable that oxidative cyclization
proceeds more quickly with the more electron rich ruthenium
center.
With regard to the ligand effect for the cleavage of
ruthenalactones, hydrido acrylato complexes RuL_acr were
and in this case also, ruthenalactone RuL2_lac was not detected
during the reaction (the amount of RuL2_acr is indicated with ×
in Figure 3b). The fact that the cleavage of these
ruthenalactones occurred simply by heating without any
additives is very promising for the catalytic synthesis of
C
https://dx.doi.org/10.1021/acs.organomet.9b00659
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
directly obtained in high yields by heating RuL1_E and RuL2_E
at 60 °C under ethylene and CO2 (1/1, 1 atm) as already
described above. On the other hand, as we have reported
previously, the hydrido acrylato complex RuL3_acr was
produced via cleavage of ruthenalactone RuL3_lac when it was
heated to 100 °C (in DMA, 4 h, 24% (46% conversion),
Scheme 2a, step B). In addition, side reactions such as retro-
oxidative cyclization of RuL3_lac to RuL3_E (14%) and formation
of the carbonate complex Ru(L3)(CO3) (7%) were noted with
this substrate under these conditions.13,14 As RuL1_acr and
RuL2_acr were formed in 88% and 29% yields after heating for 1
h at 60 °C, the cleavage of RuL1_lac and RuL2_lac is thought to
be more facile than that of RuL3_lac
.
As previously proposed for Ni-based systems in the
literature,3d−f,4b,6b,g conversion of a ruthenalactone to a
hydrido acrylato complex is considered to proceed by
dissociation of the oxygen atom from ruthenium first to give
a cationic ruthenium complex and successive β-H elimination
affords a hydrido acrylato complex (Scheme 4). Formation of
Scheme 4. Possible Reaction Mechanism for the Cleavage of
Ruthenalactone via β-H Elimination
○
Figure 4. Progress of the conversion of RuL3_lac to RuL3_acr ( ,
RuL3_lac; ×, RuL3_acr).
monitored by 31P NMR spectroscopy (Figure 5). It was found
that the rate of generation of RuL1_E was the fastest (completed
such a cationic intermediate is thought to be facilitated with
the stronger electron donation by L1 and L2 and/or steric
bulkiness of L1 and L2, which accelerates the dissociation of a
carboxylate group.25
As both RuL3_lac and RuL3_acr were observed, evaluation of
the conversion rate and equilibrium ratio of cleavage of
ruthenalactone RuL3_lac to RuL3_acr was carried out. A THF/
DMA solution of RuL3_lac was heated at 80 °C, and the
reaction progress was followed by 31P NMR spectroscopy
(Figure 4; the amounts of RuL3_lac and RuL3_acr are indicated
with ○ and ×, respectively). After 7 h, the reaction reached
equilibrium and RuL3_acr was formed in about 16% yield with
the recovery of RuL3_lac in about 75% yield (Keq = [RuL3_acr]/
[RuL3_lac] ≈ 0.2). Thus, the equilibrium ratio at 80 °C was
estimated to be RuL3_lac:RuL2_acr ≈ 80:20. From the initial rate
of the conversion of RuL3_lac to RuL3_acr, the pseudo-first-order
rate constant k (k: d[RuL3_acr]/dt = k[RuL3_lac]) at 80 °C was
estimated to be 3.9 × 10−4 min−1.26
As a summary of this section, both the oxidative cyclization
of ethylene and CO2 and cleavage of ruthenalactones (steps A
and B in cycle A, Scheme 1) are found to proceed more
smoothly on the complex bearing a more electron-donating
ligand, L1 or L2. With these two ligands, the ethylene-
coordinated complex directly gave the hydrido acrylato
complex via ruthenalactone intermediates.
Evaluation of the Ligand Effect on the Regeneration
of Zerovalent Ruthenium Complexes from Hydrido
Acrylato Complexes and Its Mechanistic Study (Step C
in Cycle A). Next, the base-promoted regeneration of
ethylene-coordinated, zerovalent ruthenium complex RuL_E
from hydrido acrylato complex RuL_acr was investigated (step
C in cycle A, Scheme 1). A THF/DMA (1/1) solution of each
of the isolated hydrido acrylato complexes RuL_acr (for L4,
hydridoacetato complex RuL4_OAc was used as a model
compound) was heated at 50 °C in the presence of 10 equiv
of KOtBu under an ethylene atmosphere and the progress of
the formation of ethylene-coordinated complex RuL_E was
○
Figure 5. Progress of the conversion of RuL_acr to RuL_E ( , RuL1_E
;
◇
, RuL4_E).
△
×, RuL2_E
;
, RuL3_E
;
in 5 min, indicated with ○ in Figure 5), followed by RuL2_E
(completed after 1 h, indicated with × in Figure 5) and RuL3_E
(completed after 9 h, indicated with △ in Figure 5). The
reaction of RuL4_OAc was too slow to be observed at 50 °C
(indicated with ◇ in Figure 5), although the reaction
smoothly took place at 120 °C in 2 h (Scheme 3b). Therefore,
D
https://dx.doi.org/10.1021/acs.organomet.9b00659
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
the order of the rate of base-promoted reductive elimination of
acrylic acid (or acetic acid) turned out to be RuL1_acr > RuL2_acr
> RuL3_acr > RuL4_OAc. Interestingly, this order is also
consistent with that of the electron-donating ability of the
ligand, as discussed above (L1 > L2 > L3 > L4), and the
electron-rich ruthenium center facilitated this step. The same
tendency was also observed with RuL_OAc complexes bearing
ligands L1−L4 (see the Supporting Information). Through the
conversion of hydrido acrylato complexes RuL_acr to ethylene-
coordinated complex RuL_E, the ruthenium center is reduced
from Ru(II) to Ru(0), which is expected to be preferred on the
more electron poor complex. Therefore, the reaction
mechanism of the step is intriguing.
Scheme 5. Possible Mechanisms for the Generation of
Ethylene-Coordinated Complexes RuL_E from Hydrido
Acrylato Complexes RuL_acr
To further obtain information on the reaction mechanism
for the conversion of hydrido acrylato complex RuL_acr to
ethylene-coordinated complex RuL_E, we examined the effect
of the concentration of KOtBu on this step by using RuL3_acr
.
The rate constant for the formation of RuL3_E under an
ethylene atmosphere was measured with various amounts of
KOtBu (2, 5, 10, and 20 equiv) to ruthenium and plotted as
shown in Figure 6. It indicates that the reaction rate was close
5b). The last pathway shown in Scheme 5c is that via a cationic
ruthenium intermediate. The acrylate anion is dissociated from
the hydrido acrylato complex RuL_acr first to generate a cationic
hydrido complex (step A). Then, deprotonation of the
ruthenium by KOtBu (step B) or coordination of OtBu to
the ruthenium center (step C) followed by reductive
elimination of HOtBu (step D) gives the zerovalent ruthenium
complex. Finally, coordination of ethylene to the ruthenium
regenerates RuL_E (step E). There have been several reported
examples of the generation of cationic hydrido rutheniums by
the dissociation of chloride anion as a reaction similar to that
of step A,28 and conversion of a cationic hydridoruthenium
complex to a zerovalent ruthenium complex by PhLi is also
known as a reaction similar to step B or steps C and D.29
We then evaluated the validity of these pathways on the
basis of the experimental results. First of all, the concerted
reductive elimination pathway (Scheme 5a) is not consistent
with the observation that the generation of ethylene-
coordinated complex RuL_E is more rapid with more electron
rich hydrido acrylato complexes (RuL1_acr > RuL2_acr > RuL3_acr
> RuL4_OAc). The initial reductive elimination step is thought
to be preferred with more electron poor complexes, because
ruthenium is reduced from Ru(II) to Ru(0). Next, the initial
deprotonation pathway (Scheme 5b) can be discarded from
the results that the reaction rate was close to zero-order with
the concentration of KOtBu and RuL3_acr and RuL3_E were
Figure 6. Rate constants (k: d[RuL3_acr]/dt = −k[RuL3_acr]) for the
generation of RuL3_E from RuL3_acr with various amounts of KOtBu.
to zero order with the concentration of KOtBu.27 It is also
noted that, during the conversion of RuL3_acr to RuL3_E, these
two complexes were observed as the two major species by 31P
NMR.
Here we considered the following three pathways for the
conversion of RuL_acr to RuL_E as shown in Scheme 5. The first
is the concerted reductive elimination of acrylic acid by
forming an O−H bond to give a zerovalent ruthenium
complex, followed by coordination of ethylene. The liberated
acrylic acid is converted to its salt with KOtBu (Scheme 5a).
Such reductive elimination of carboxylic acid has been
proposed on the basis of DFT calculations for the nickel-
catalyzed synthesis of acrylic acid.6b,c,e,f The second pathway is
the initial deprotonation of the hydrido acrylato complex to
give an anionic intermediate, which undergoes dissociation of
acrylate anion, followed by coordination of ethylene (Scheme
E
https://dx.doi.org/10.1021/acs.organomet.9b00659
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
observed as the two major species during the reaction. If the
deprotonation is the rate-determining step or pre-equilibrium,
the reaction rate should depend on KOtBu concentration. So
far, the pathway in Scheme 5c is consistent with the
experimental results, with the assumption that the dissociation
of acrylate anion (step A) is the rate-determining step. In this
case, the reaction rate is independent of the concentration of
KOtBu, and the generation of the cationic intermediate [LRu-
H]+ (step A) would proceed more rapidly with electron-rich
complexes, which coincides with the experimental results.30 As
another support for the pathway via the cationic intermediate,
the generation of RuL3_E by KOtBu was accelerated in the
presence of 5 equiv of NaBArF, which is thought to interact
with the acrylate group and facilitate its dissociation as a Lewis
acid (see Figure S17, SI-4).3d
complex RuL3_E was directly generated and hydrido acrylato
complex RuL3_acr was not observed as the intermediate.31
To reveal the order of KOtBu on the reaction, the reaction
rate for the consumption of RuL3_lac was determined by 31P
NMR using various amounts of KOtBu (2−30 equiv) and the
pseudo-first-order rate constants (k: d[RuL3_lac]/dt =
−k[RuL3_lac]) were plotted against the amount of KOtBu, as
shown in Figure 7a. Although saturation of the reaction rate
was observed at higher concentration, first-order dependence
on the amount of KOtBu was observed at lower concen-
tration.32 If RuL3_acr was indeed formed during the reaction
(path A in Scheme 6), it would show no first-order
dependence on KOtBu; however, this is observed upon the
As a summary of the examination of each elementary step of
the Ru-mediated acrylate synthesis, oxidative cyclization of
ethylene and CO2 (step A, Scheme 1), cleavage of the
ruthenalactone to the hydrido acrylato complex (step B in
cycle A, Scheme 1), and regeneration of the zerovalent
ruthenium complex from the hydrido acrylato complex (step C
in cycle A, Scheme 1) all turned out to proceed more rapidly
with the complexes bearing the more electron rich ligands L1
and L2. These results can be explained by presuming that the
reactions proceed through dissociation of a carboxylate anion
from ruthenalactones or hydrido acrylato complexes to give
cationic Ru species as the rate-determining steps and the
reactions proceed more rapidly with Ru complexes having
electron-rich ligands due to the stabilizing effect of the
generated cationic ruthenium intermediates.
Elucidation of the Mechanism of the Cleavage of
Ruthenalactones in the Presence of Base (Steps B′ and
C′ in Cycle B). As shown in Scheme 1, two major mechanisms
are proposed concerning the acrylate synthesis from ethylene
and CO2, and cycle A has been lent support as a plausible
pathway in the Ru-mediated reaction as described above. Now
we examine the possibility of cycle B: that is, the direct
abstraction of the α-proton of the carbonyl group of the
ruthenalactone by a base to form an acrylate-coordinated
complex (step B′ in cycle B, Scheme 1), followed by ligand
exchange with ethylene (step C′ in cycle B, Scheme 1). So far,
in nickel systems, both mechanisms (cycles A and B) are
supported by DFT calculations. For example, a direct
deprotonation mechanism was suggested for NaOMe-
promoted cleavage,6c and a β-H elimination pathway was
proposed for DBU.6b However, direct experimental evidence is
still lacking in nickel-based systems. In contrast, in the
ruthenium system, it was already confirmed that some of the
ruthenalactones undergo thermal cleavage via β-H elimination
without any additives to give hydrido acrylato complexes.
However, the behavior of these ruthenalactones with bases
remained unexplored. As the high stability of various
ruthenium intermediates is one of the noteworthy features of
this Ru system, we decided to examine the behavior of the
ruthenalactones toward bases to clarify whether the direct
deprotonation of metallalactone pathway operates or not.
We chose RuL3_lac as a substrate, as this ruthenalactone can
be easily prepared and both hydrido acrylato complex RuL3_acr
and ethylene-coordinated complex RuL3_E were isolated as
stable complexes. When a THF/DMA (1/1) solution of
ruthenalactone RuL3_lac was treated with 10 equiv of KOtBu
under an ethylene atmosphere at 80 °C, ethylene-coordinated
Figure 7. Plot of rate constants (k: d[RuL3_lac]/dt = −k[RuL3_lac]) for
the generation of RuL3_E from RuL3_lac with various amounts of (a)
KOtBu and (b) NaOMes.
F
https://dx.doi.org/10.1021/acs.organomet.9b00659
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
Scheme 6. Proposed Two Pathways for the Cleavage of
Ruthenalactone
Table 1. Catalytic Synthesis of Acrylate Salt Using Various
Ruthenium Complexes
a
TON
entry
cat.
base
140 °C
180 °C
1
2
3
4
5
6
7
RuL1_OAc
RuL2_OAc
RuL3_OAc
RuL4_OAc
RuL4_OAc
RuL4_OAc
RuL4_OAc
KOtBu
KOtBu
KOtBu
KOtBu
K2CO3
CsOMes
Cs2CO3
Cs2CO3
0.3
0.2
1.5
0.7
0.3
0.4
1.1
7.9
0.6
8.8
8.6
15
reaction of RuL3_lac, indicating that a different mechanism is
operational in the presence of KOtBu. Therefore, when KOtBu,
a strong base, was added, the cleavage of ruthenalactone
RuL_lac to RuL_E is likely to proceed via a direct deprotonation
pathway (path B in Scheme 6).33
b
8
RuL4_E
1
a
b
Determined by H NMR. 1000 equiv (2.5 mmol) of Cs2CO3 was
used.
Next, to examine the effect of the basicity of the base in this
step, sodium 2,4,6-trimethylphenoxide (NaOMes) was applied
as a weaker base in comparison to KOtBu. The same
experiment was performed, and initial pseudo-first-order rate
constants were plotted against the amount of NaOMes (Figure
7b). It was revealed that the reaction rate was much slower
than that of KOtBu and linearly increased as the amount of
NaOMes was increased. The dependence of the reaction rate
on the amount of NaOMes is consistent with the direct
deprotonation of the ruthenalactone (path B, Scheme 6) as in
the case of KOtBu. In contrast to the reaction of KOtBu, the
reaction rate was not zero when the fitting line was
extrapolated to zero equivalent of NaOMes, suggesting the
existence of another pathway which is zero order on the
concentration of NaOMes. Although the exact reason for this
phenomenon is not clear, the presence of a competitive β-H
elimination pathway (path A, Scheme 6) and/or the presence
of a retro-oxidative cyclization pathway,14,23 whose reaction
rate may become competitive under the conditions, are
conceivable.
The results described in this section verified the existence of
two pathways (cycles A and B in Scheme 1) experimentally for
the first time in the acrylate salt synthesis and are realized by
utilizing the characteristic properties of the ruthenium system,
where the key intermediates (zerovalent ethylene-coordinated
complex, ruthenalactone, and hydrido acrylato complex) are
isolable. These results would also afford useful insights into
acrylate synthesis using other transition-metal catalysts.
Evaluation of Catalytic Activities of Ruthenium
Complexes. As the elementary steps of Ru-mediated acrylate
formation from ethylene and CO2 have been clarified, we
finally investigated the catalytic activities of ruthenium
complexes bearing L1−L4 (Table 1). A DMA solution of
each of the hydrido acetato complexes RuL_OAc (L1−L4)34 was
heated at 140 or 180 °C under ethylene/CO2 pressure (3.0
MPa, 1/1) in the presence of an excess amount of KOtBu (250
equiv to Ru). While RuL1_OAc and RuL2_OAc, which showed
high reactivity in the elementary steps of acrylate formation,
gave only substoichiometric amounts of acrylate salts at either
temperature (entries 1 and 2), RuL3_OAc gave more than 1
equiv of the acrylate salt at 140 °C, but the yield was not
improved at 180 °C (entry 3).35 In contrast, the TON with
RuL4_OAc, which showed the lowest reactivity in the
stoichiometric reactions, was less than 1 at 140 °C, but it
increased up to 7.9 at 180 °C (entry 4). Next, the scope of
applicable base was investigated by using RuL4_OAc (entries 5−
7). It was found that cesium 2,4,6-trimethylphenoxide
(CsOMes) or cesium carbonate gave acrylate salts with
TONs similar to those of KOtBu (entries 6 and 7). The
possible use of carbonate as a base is important, since it is
much easier to handle than alkoxides or phenoxides but has
not been used commonly in the catalytic acrylate synthesis
using group 10 metal complexes.36 After further optimization
of the conditions, the TON was increased up to 15 by using
1000 equiv of Cs2CO3 with the ethylene-coordinated complex
RuL4_E as a catalyst.37
The applicability of weak bases could be associated with the
reaction mechanism of the cleavage of metallalactone. Under
the catalytic conditions, the system should be weakly basic due
to the existence of a large excess of CO2, which neutralizes the
base by making a CO2 adduct. Therefore, direct abstraction
pathway (cycle B, Scheme 1) would become less likely. On the
other hand, the β-H elimination pathway (cycle A, Scheme 1)
would become favorable because β-H elimination could occur
without base and successive deprotonation from the cationic
species to give a zerovalent Ru species would be able to
proceed with a weaker base.
The higher TON obtained with RuL4_OAc was inconsistent
with the stoichiometric reactivities investigated above, where
electron-rich complexes showed higher activities for each step
of the catalytic cycle. We have carried out several experiments
to compare the stability of the complexes as a possible
explanation for this phenomenon, but no obvious trends were
observed.38
CONCLUSIONS
■
In conclusion, we performed detailed mechanistic investiga-
tions on the ruthenium-mediated direct acrylic acid synthesis
from ethylene and CO2. First, we studied the ligand effect for
each step of the proposed catalytic cycle A: that is, oxidative
cyclization of ethylene and CO2, cleavage of ruthenalactones,
and the regeneration of zerovalent ruthenium complexes from
hydrido acrylato complexes. All of the steps were found to
proceed more smoothly on the more electron rich ruthenium
complexes, which suggested that the generation of cationic
G
https://dx.doi.org/10.1021/acs.organomet.9b00659
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
pressure, and the residue was washed with degassed MeOH. The
obtained white solid was filtered through a short pad of silica gel using
THF as a solvent. The solvent was removed under reduced pressure
to give the product as a white solid (517 mg, 0.73 mmol) in 81%
yield. Spectral data were in good agreement with literature data.15
Synthesis of L2. To a solution of (2-bromophenyl)-
diisopropylphosphine (3.86 g, 14.1 mmol)41 in Et2O (30 mL) cooled
to −78 °C was added dropwise nBuLi (1.6 M in hexane, 8.9 mL, 14.2
mmol). After the mixture was stirred for 4 h at −78 °C, a Et2O (10
mL) solution of P(OPh)3 (1.41 g, 4.54 mmol) was added dropwise.
The reaction mixture was stirred overnight at room temperature and
then heated under reflux for 1.5 h. After the reaction was quenched
with degassed MeOH (2.0 mL), the solvent was removed under
reduced pressure. The resulting crude mixture was filtered through a
short pad of silica gel using THF as a solvent, and the solvent was
removed under reduced pressure. To the mixture was added hexane,
and the resulting white precipitate was collected by filtration and
washed with hexane to give the product as a white solid (0.54 g, 0.88
ruthenium species by dissociation of a carboxylate anion is
important in these steps. Next, the mechanism for the cleavage
of a ruthenalactone in the presence of a base was studied in
detail, and direct deprotonation of the ruthenalactone by a
base was also found to be possible, but its pathway might be
competitive with the β-H elimination pathway when a weaker
base is used. Finally, the catalytic activities were examined with
ruthenium complexes bearing various tetradentate phosphine
ligands. In contrast to the stoichiometric reactivities, electron-
deficient complexes exhibited higher TONs. At the present
stage, many explanations for the difference of stoichiometric
and catalytic activities are possible and clarification of the
deactivation pathways to improve the catalytic efficiency of the
present reaction would be an important topic of future
research. The results obtained here would afford useful insights
not only for the development of better catalysts based on
ruthenium but also for the clarification of the reaction
mechanism of acrylate formation from ethylene and CO2 in
general.
1
mmol) in 19% yield. H NMR (500 MHz, C6D6): δ 7.37−7.30 (m,
3H), 7.09 (dt, J = 1.0, 7.5 Hz, 3H), 7.06−7.00 (m, 3H), 6.97 (t, J =
7.5 Hz, 3H), 2.28 (sep, J = 7.5 Hz, 3H), 2.03 (dsep, J = 7.0, 7.0 Hz,
3H), 1.41 (dd, J = 12.5, 6.5 Hz, 9H), 1.28 (dd, J = 11.0, 7.0 Hz, 9H),
1.01 (dd, J = 14.5, 7.0 Hz, 9H), 0.70 (dd, J = 13.5, 7.0 Hz, 9H). 13C
NMR (125 MHz, C6D6): δ 148.8 (dt, J = 34.8, 15.5 Hz), 142.6 (dd, J
= 29.9, 18.0 Hz), 136.0 (dd, J = 4.8, 3.6 Hz), 131.9 (d, J = 6.0 Hz),
128.6, 127.7, 26.4 (d, J = 15.5 Hz), 22.9 (dd, J = 15.5, 9.5 Hz), 20.9−
20.2 (m). 31P NMR (200 MHz, C6D6): δ 0.29 (d, J = 149.5 Hz, 3P),
− 16.4 (q, J = 154.9 Hz, 1P). IR (ATR): 3040, 2954, 2859, 1460,
1438, 1383, 1356, 1238, 1152, 1102, 1030 cm−1. HRMS (FD+): m/z
calcd for C36H54P4 + H+ 611.3249, found 611.3238.
EXPERIMENTAL SECTION
■
General Methods. All operations were performed under a
nitrogen or an argon atmosphere. 1H, 13C, and 31P NMR spectra
1
were recorded on a JEOL ECS-400 (400 MHz for H and 160 MHz
for 31P), an ECX-500 (500 MHz for H, 125 MHz for 13C, and 200
1
MHz for 31P) an ECZ-500 (500 MHz for H, 125 MHz for 13C, and
1
200 MHz for 31P), or a Bruker DRX-500 (500 MHz for 1H, 125 MHz
for 13C, and 200 MHz for 31P) spectrometer in C6D6, CD3CN/C6D6
(1/1), THF-d8, or D2O as solvent. Chemical shifts are expressed in
parts per million (ppm) referenced to solvent signals (δH 7.16 and δC
128.0 for C6D6, δH 7.16 and δC 128.0 for CD3CN/C6D6 (1/1), δH
3.58 and δC 25.3 for THF-d8, and δH 4.79 for D2O) or H3PO4 (δP
0.00) as an external standard. IR spectra were recorded on an FT/IR-
460 plus instrument (JASCO Co., Ltd.) with ATR PRO450-S
accessory (JASCO Co., Ltd.). Mass spectra were recorded on a Bruker
micrOTOF-QII (ESI-mass) mass spectrometer. Silica gel 60 (Kanto
Chemical Co., Inc.) was used for column chromatography. THF,
Et2O, hexane, and toluene were purified by a Glass-Contour solvent
purification system. Dehydrated DMSO, DMF, and DMA (N,N-
dimethylacetamide) were purchased from Kanto Chemicals and
degassed by freeze−pump−thaw cycles. Diglyme, DMPU, and o-
xylene were purchased from TCI or Aldrich Chemicals, degassed by
freeze−pump−thaw cycles, and dried over molecular sieves. C6D6 and
THF-d8 were purchased from Kanto Chemicals and dehydrated by
the benzophenone ketyl method. CD3CN was purchased from Kanto
Chemicals or Aldrich Chemicals, degassed by freeze−pump−thaw
cycles, and dried over molecular sieves. Other reagents were
purchased from TCI, Kanto Chemicals, Aldrich Chemicals, or
Fujifilm Wako Chemicals and used as received unless otherwise
noted. Cylinders of ethylene, CO2, and a 1/1 mixture of ethylene and
Synthesis of RuL1_OAc. A solution of RuH(OAc)(PPh3)3 (500 mg,
0.527 mmol) and L1 (374 mg, 0.529 mmol) in DMA (15 mL) was
heated at 60 °C for 2 h. The resulting white precipitate was collected
by filtration, washed with DMA and hexane, and dried in vacuo to
give RuL1_OAc (173 mg, 0.200 mmol) as a white solid in 38% yield. 1H
NMR (500 MHz, CD3CN/C6D6 1/1): δ 2.10−2.01 (m, 1H), 2.01−
1.91 (m, 7H), 1.88−0.93 (m, 69H), 0.91−0.80 (m, 2H), 0.80−0.68
(m, 2H), −11.52 (ddt, J = 81.5, 21.5, 28.5 Hz, 1H). 13C NMR (125
MHz, CD3CN/C6D6 1/1): δ 173.7, 44.4 (t, J = 6.0 Hz), 44.2, 37.9
(dt, J = 4.8, 14.4 Hz), 31.3, 30.5, 29.8, 29.5−25.5 (multiple signals for
CH2CH2PCy2 and Ac groups). 31P NMR (200 MHz, CD3CN/C6D6
1/1): δ 147.0 (q, J = 11.1 Hz, 1P), 63.9 (t, J = 12.9 Hz, 2P), 50.5 (q, J
= 11.9 Hz, 1P). IR: 2913, 2844, 1945, 1612, 1447, 1415, 1370, 1320,
1297, 1269, 1255, 1219, 1192, 1173, 1115, 1070, 1044, 1026, 1002
cm−1. HRMS (ESI+): m/z calcd for C44H82O2P4Ru − H−, 867.4225,
found 867.4249.
Synthesis of RuL2_OAc. A solution of RuH(OAc)(PPh3)3 (155 mg,
0.164 mmol) and L2 (100 mg, 0.164 mmol) in o-xylene (5 mL) was
heated at 140 °C for 2 h. After the solvent was removed under
reduced pressure, hexane was added. The resulting yellow precipitate
was collected by filtration and washed with hexane to give RuL2_OAc
(95.4 mg, 0.124 mmol) as a yellow solid in 76% yield. 1H NMR (400
MHz, C6D6): δ 7.88 (t, J = 6.4 Hz, 2H), 7.65 (t, J = 7.2 Hz, 1H), 7.42
(brs 3H), 7.02−6.84 (m, 6 H), 2.92 (brs, 2H), 2.69 (sep, J = 7.6 Hz,
2H), 2.57 (dsep, J = 6.0 Hz, 6.0 Hz, 2H), 2.38 (s, 3H) 1.53 (q, J = 7.6
Hz, 6H), 1.43−1.26 (m, 12H) 1.07 (q, J = 7.2 Hz, 6H), 0.92 (dd, J =
13.6, 6.4 Hz, 6H), 0.64 (q, J = 7.2 Hz, 6H), −8.22 (ddt, J = 92.8, 17.2,
33.2 Hz, 1H). 13C NMR (125 MHz, C6D6): δ 175.5, 149.3−148.5
(m), 146.4 (dd, J = 32.3, 22.6 Hz), 145.1 (dt, J = 47.9 15.5 Hz),
130.2, 130.1, 129.9, 129.3 (t, J = 6.0 Hz), 128.8, 30.2−29.6 (m), 28.8
(d, J = 8.4 Hz), 25.9 (d, J = 3.6 Hz), 20.8, 19.9, 19.7, 19.6 (d, J = 2.4
Hz), 19.5 (d, J = 3.6 Hz), 18.6 (brs) (some of the signals of aromatic
carbons were overlapped with C6D6). 31P NMR (160 MHz, C6D6): δ
126.0−125.5 (m, 1P), 67.8 (brt, J = 17.3 Hz, 2P), 56.2 (brs, 1P). IR
(ATR): 3049, 2959, 2918, 2872, 1976, 1606, 1431, 1369, 1325, 1238,
1107, 1029 cm−1. HRMS (ESI+): m/z calcd for C38H58O2P4Ru −
H− 771.2347, found 771.2374.
CO2 were purchased from Kayama Sanso. RuL3_OAc, RuL3_E, RuL3_lac
,
and RuL3_acr were prepared according to our previously reported
40
methods.13 L439 and RuH(OAc)(PPh3)3 were prepared according
to the reported procedure. Sodium, potassium, and cesium salts of
2,4,6-trimethylphenoxide (NaOMes, KOMes, and CsOMes) were
prepared by the reaction of 2,4,6-trimethylphenol with NaH, KH, and
CsOH respectively, in THF.
Modified Synthesis of L1.15 To a solution of vinylmagnesium
bromide (0.5 M in THF, 6.0 mL, 3.0 mmol) cooled to −78 °C was
added dropwise neat PCl3 (80 μL, 0.9 mmol). The reaction mixture
was stirred overnight at room temperature. The mixture was cooled to
−78 °C again, and HPCy2 (0.87 mL, 4.0 mmol) and a THF solution
n
of LDA (prepared from BuLi (1.6 M in hexane, 3.2 mL, 5.1 mmol)
and diisopropylamine (0.7 mL, 5.0 mmol) in THF 5.0 mL at −78 °C)
were added successively. After the reaction mixture was stirred
overnight at room temperature, the reaction was quenched with
degassed MeOH (10 mL). The solvent was removed under reduced
Synthesis of RuL4_OAc. A solution of RuH(OAc)(PPh3)3 (122 mg,
0.129 mmol) and L42 (120 mg, 0.147 mmol) in toluene (2.5 mL) was
stirred at 120 °C for 2 h. After the solvent was removed under
H
https://dx.doi.org/10.1021/acs.organomet.9b00659
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
reduced pressure, Et2O was added. The obtained yellow precipitate
was collected by filtration and washed with Et2O to give RuL4_OAc
(54.8 mg, 0.056 mmol) as a yellow solid in 44% yield. 1H NMR (500
MHz, C6D6): δ 8.24 (q, J = 7.0 Hz, 4H), 8.20 (t, J = 7.0 Hz, 2H), 7.97
(t, J = 7.0 Hz, 1H), 7.57 (d, J = 4.5 Hz, 2H), 7.34 (t, J = 7.0 Hz, 1H),
7.24 (t, J = 9.0 Hz, 4H), 7.21−7.15 (m, 4H), 7.13 (t, J = 7.0 Hz, 4H),
7.09−6.98 (m, 5H), 6.86 (t, J = 6.5 Hz, 2H), 6.81 (t, J = 7.0 Hz, 1H),
6.78−6.69 (m, 4H), 6.67−6.56 (m, 8H), 2.32 (s, 3H), −6.47 (ddt, J =
94.0, 20.5, 27.5 Hz, 1H). 13C NMR (125 MHz, THF-d8): δ 174.8,
150.8−149.8 (m), 146.0−144.8 (m), 140.9−140.3 (m), 138.8−138.3
(m), 135.4 (t, J = 6.0 Hz), 134.3−133.7 (m), 133.3 (t, J = 5.9 Hz),
131.0 (d, J = 13.3 Hz), 130.4, 129.9, 129.6, 129.3, 128.1−127.3 (m)
(the signal of CH3COO was overlapped with that of THF-d8). 31P
NMR (200 MHz, C6D6): δ 125.9 (t, J = 15.8 Hz, 1P), 60.4 (t, J = 10.6
Hz, 2P), 48.3 (brs, 1P). IR (ATR): 3049, 1928, 1878, 1607, 1586,
1572, 1481, 1452, 1431, 1371, 1323, 1261, 1185, 1166, 1156, 1103,
1087, 1027 cm−1. HRMS (ESI+): m/z calcd for C56H46O2P4Ru − H−
975.1408, found 975.1427.
cm−1. HRMS (ESI+): m/z calcd for C38H58P4Ru − ethylene + H+
713.2292, found 713.2280.
Synthesis of RuL4_E. To a solution of RuL4_OAc (97.6 mg, 0.100
mmol) in DMA (6 mL) was added KOtBu (22.4 mg, 0.200 mmol)
under Ar. The Ar atmosphere was replaced by ethylene using a
balloon, and the mixture was stirred at 120 °C for 2 h. After the
solvent was removed under reduced pressure, THF and hexane were
added successively. The resulting red precipitate was collected by
filtration and washed with hexane to give RuL4_E (77.7 mg, 0.082
mmol) as a red solid in 82% yield. 1H NMR (500 MHz, C6D6): δ 8.21
(t, J = 6.0 Hz, 3H), 7.39 (d, J = 7.5 Hz, 3H), 7.11−7.02 (m, 15H),
6.91 (t, J = 7.5 Hz, 6H), 6.82 (t, J = 7.5 Hz, 15H), 2.90 (brd, J = 3.5
Hz, 4H). 13C NMR (125 MHz, THF-d8): δ 151.2−150.3 (m), 147.4−
146.3 (m), 142.6 (t, J = 16.9 Hz), 133.2, 133.1, 132.7, 131.0−129.9
(m), 129.7, 129.1, 128.2, 127.6, 43.1 (d, J = 10.0 Hz). The peak at
43.1 ppm was assigned as that of coordinated ethylene, since RuL4_E is
supposed to have only one signal derived from coordinated ethylene
at such a low chemical shift, and all the other carbons are aromatic
carbons. 31P NMR (200 MHz, C6D6): δ 138.2 (q, J = 31.8 Hz, 1P),
70.2 (d, J = 31.8 Hz, 3P). IR (ATR): 3068, 3048, 1566, 1476, 1451,
1429, 1406, 1364, 1305, 1258, 1197, 1163, 1106, 1078. 1065, 1043,
1024 cm−1.
Synthesis of RuL1_E. To a solution of RuL1_OAc (246 mg, 0.283
mmol) in DMA (10 mL) was added KOtBu (70 mg, 0.62 mmol)
under Ar. The Ar atmosphere was replaced by ethylene using a
balloon, and the mixture was stirred at 60 °C for 1 h. The resulting
yellow precipitate was collected by filtration and washed with Et2O to
give RuL1_E (190 mg, 0.227 mmol) as a yellow solid in 80% yield. 1H
NMR (400 MHz, C6D6): δ 2.71 (brd, J = 4.0 Hz, 4H), 2.13 (d, J =
12.8 Hz, 6H), 2.03 (d, J = 10.4 Hz, 6H), 1.84 (brs, 18H), 1.74 (d, J =
5.6 Hz, 6H), 1.60−1.14 (m, 42H). 13C NMR (125 MHz, C6D6): δ
46.7, 32.0−30.8 (m), 31.1, 30.4, 30.0−29.2 (m), 28.6 (brs), 27.2, 18.6
(d, J = 8.4 Hz, CH2CH2). The 13C NMR peak for coordinated
Synthesis of RuL1_acr. RuL1_acr was prepared by the following
procedure and used for the NMR experiments for the generation of
RuL1_E from RuL1_acr
.
To a solution of RuL1_E (50.0 mg, 0.060 mmol) in toluene (5 mL)
was introduced an atmospheric pressure of a 1/1 mixture of ethylene
and CO2. The reaction mixture was stirred at 50 °C for 1 h. After the
solvent was removed under reduced pressure, the resulting solid was
washed with a minimum amount of THF (ca. 0.5 mL × 2) to give
1
ethylene was assigned as follows. On the H NMR spectrum, the
1
RuL1_acr (32.3 mg, 0.037 mmol) as a white solid in 61% yield. H
signal at 2.71 ppm was assigned as that of coordinated ethylene on the
basis of the number of protons (4H) and comparison of the chemical
shift with those of related compounds (for example, RuL4_E exhibited
a signal for coordinated ethylene at 2.90 ppm). On the HMQC
spectrum of RuL1_E (see Figure S1), correlation between peaks at 2.71
NMR (500 MHz, CD3CN/C6D6 1/1): δ 6.17 (dd, J = 18.0, 11.0 Hz,
1H), 5.82 (dd, J = 17.5, 3.5 Hz, 1H), 5.10 (dd, J = 9.5, 3.0 Hz, 1H),
2.10−1.88 (m, 8H), 1.88−0.93 (m, 66H), 0.91−0.80 (m, 2H), 0.80−
0.68 (m, 2H), −11.52 (ddt, J = 82.0, 22.5, 27.5 Hz, 1H). 13C NMR
(125 MHz, CD3CN/C6D6 1/1): δ 170.0, 141.7, 118.7, 44.4 (t, J = 6.0
Hz), 44.2, 37.9 (dt, J = 4.8, 14.4 Hz), 31.3, 30.5, 29.8, 29.5−25.5
(multiple signals for CH2CH2PCy2 and AcO groups). 31P NMR (200
MHz, CD3CN/C6D6 1/1): δ 145.0 (q, J = 11.8 Hz, 1P), 61.9 (t, J =
13.4 Hz, 2P), 49.0−47.8 (m, 1P). IR (ATR): 2920, 2846, 1925, 1633,
1588, 1444, 1400, 1335, 1299, 1255, 1190, 1173, 1114, 1072, 1002
cm−1. HRMS (ESI+): m/z calcd for C45H82O2P4Ru − H− 879.4225,
found 879.4255.
1
ppm on the H NMR and 18.6 ppm on the 13C NMR was found,
which indicated that the peak at 18.6 ppm was that of the coordinated
ethylene. 31P NMR (160 MHz, C6D6): δ 157.1 (q, J = 30.6 Hz, 1P),
67.5 (d, J = 30.6 Hz, 3P). IR (ATR): 2917, 2844, 1569, 1407, 1193,
1169, 1001 cm−1. HRMS (ESI+): m/z calcd for C44H82P4Ru −
ethylene + H+ 809.4170, found 809.4137.
Synthesis of RuL2_E. To a solution of RuL2_OAc (354 mg, 0.459
mmol) in DMA (15 mL) was added KOtBu (99.6 mg, 0.888 mmol)
under Ar. The Ar atmosphere was replaced by ethylene using a
balloon, and the mixture was stirred at 60 °C for 1 h. After the solvent
was removed under reduced pressure, THF and hexane were added
successively. The obtained red precipitate was collected by filtration
and washed with hexane to give RuL2_E (210 mg, 0.284 mmol) as a
red solid in 62% yield. When RuL2_E was dissolved in C6D6 in the
absence of an excess amount of ethylene, it was gradually converted to
unknown species, which could be reconverted to RuL2_E under an
ethylene atmosphere. Therefore, NMR spectra were recorded under
Synthesis of RuL2_acr. RuL2_acr was prepared by the following
procedure and used for the NMR experiments for the generation of
RuL2_E from RuL2_acr
.
To a solution of RuL2_E (143.3 mg, 0.194 mmol) in benzene (10
mL) was added acrylic acid (13 μL, 0.19 mmol). The reaction mixture
was stirred at room temperature for 2 h. After the solvent was
removed under reduced pressure, hexane was added. The obtained
yellow precipitate was collected by filtration and washed with hexane
to give RuL2_acr (58.2 mg, 0.074 mmol) as a yellow solid in 38% yield.
1H NMR (500 MHz, C6D6): δ 7.88 (t, J = 6.0 Hz, 2H), 7.64 (t, J =
7.0 Hz, 1H), 7.46−7.38 (m, 3H), 7.02−6.88 (m, 6H), 6.78 (dd, J =
17.5, 10.5 Hz, 1H), 6.49 (dd, J = 17.0, 3.5 Hz, 1H), 5.46 (dd, J = 10.5,
3.0 Hz, 1H), 2.94 (brs, 2H), 2.73 (sept, J = 7.0 Hz, 2H), 2.60 (dsept, J
= 6.0, 6.0 Hz, 2H), 1.50 (q, J = 7.5 Hz, 6H), 1.38−1.26 (m, 12H),
1.04 (q, J = 7.5 Hz, 6H), 0.89 (dd, J = 17.5, 7.0 Hz, 6H), 0.67 (q, J =
6.5 Hz, 6H), −8.25 (ddt, J = 92.5, 18.0, 33.5 Hz, 1H). 13C NMR (125
MHz, C6D6): δ 172.4, 149.4−148.2 (m), 146.3 (dd, J = 32.4, 22.8
Hz), 145.1 (dt, J = 46.8, 15.5 Hz), 138.8 (d, J = 3.6 Hz), 130.4−129.8
(m), 129.3 (t, J = 5.9 Hz), 128.9, 120.7, 30.2−29.7 (m), 28.7 (d, J =
7.3 Hz), 20.9, 19.9, 19.7, 19.5, 18.8. 31P NMR (200 MHz, C6D6): δ
125.7 (q, J = 15.8 Hz, 1P), 67.9 (t, J = 15.8 Hz, 2P), 55.9 (q, J = 10.6
Hz, 1P). IR (ATR): 2946, 2913, 2862, 1962, 1631, 1591, 1402, 1331,
1255, 1104, 1025 cm−1. HRMS (ESI+): m/z calcd for C39H58O2P4Ru
− H− 783.2347, found 783.2354.
1
an ethylene atmosphere. On the H NMR spectrum, the signals for
free ethylene and coordinated ethylene were separately observed and
1
the H NMR chemical shifts for RuL2_E were identical with those in
the absence of ethylene, which indicated the exchange of coordinated
ethylene is slower in comparison to the time scale of 1H NMR
measurement. 1H NMR (500 MHz, C6D6): δ 8.05 (t, J = 6.0 Hz, 3H),
7.35 (d, J = 7.5 Hz, 3H), 7.06 (t, J = 7.0 Hz, 3H), 6.99 (t, J = 7.5 Hz,
3H), 2.92 (q, J = 3.5 Hz, 4H), 2.52 (sept, J = 6.5 Hz, 6H), 1.08−0.98
(m, 18H), 0.73 (brs, 18H). 13C NMR (125 MHz, C6D6): δ 149.3−
147.9 (m), 130.0−129.7 (m), 128.6 (d, J = 19.1 Hz), 127.6 (d, J = 4.8
Hz), 127.5, 31.5 (brs), 22.1 (d, J = 9.6 Hz, CH2CH2), 19.6, 19.4.
The 13C NMR peak for coordinated ethylene was assigned on the
basis of the DEPT135 NMR spectrum (see Figure S2), which
indicated the peak at 22.1 ppm is CH2, which is the only CH2 carbon
of RuL2_E derived from the coordinated ethylene. 31P NMR (200
MHz, C6D6): δ 132.9 (q, J = 32.5 Hz, 1P), 70.9 (d, J = 32.7 Hz, 3P).
IR (ATR): 2977, 2954, 2904, 2866, 1568, 1402, 1198, 1102, 1021
General Procedure for Catalytic Acrylate Synthesis. A
solution of a ruthenium complex (2.5 μmol) and the desired amount
of a base in a solvent (2 mL) was placed in a 20 mL stainless
I
https://dx.doi.org/10.1021/acs.organomet.9b00659
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
autoclave (Taiatsu Techno) under Ar. The system was charged with
ethylene (1.5 MPa) with stirring at room temperature and then with
CO2 (1.5 MPa, total 3.0 MPa). The autoclave was heated at the
desired temperature and time with stirring. The autoclave was
depressurized, and the reaction mixture was transferred into a 20 mL
round-bottom flask by washing with THF, and then the solvent was
removed in vacuo. The amount of potassium acrylate was determined
by 1H NMR in D2O using sodium 3-(trimethylsilyl)-1-propane-
sulfonate (2.2 mg, 10.0 μmol) as an internal standard.
Carbon Dioxide: Preparation, Structure, and Reactivity. Angew. Chem.,
Int. Ed. Engl. 1987, 26, 771−773.
(3) (a) Bruckmeier, C.; Lehenmeier, M. W.; Reichardt, R.; Vagin, S.;
Rieger, B. Formation of Methyl Acrylate from CO2 and Ethylene via
Methylation of Nickelalactones. Organometallics 2010, 29, 2199−
2202. (b) Lee, S. Y. T.; Cokoja, M.; Drees, M.; Li, Y.; Mink, J.;
Herrmann, W. A.; Kuhn, F. E. Transformation of Nickelalactones to
̈
Methyl Acrylate: On the Way to a Catalytic Conversion of Carbon
Dioxide. ChemSusChem 2011, 4, 1275−1279. (c) Plessow, P. N.;
̈
Weigel, L.; Lindner, R.; Schafer, A.; Rominger, F.; Limbach, M.;
ASSOCIATED CONTENT
* Supporting Information
Hofmann, P. Mechanistic Details of the Nickel-Mediated Formation
of Acrylates from CO2, Ethylene and Methyl Iodide. Organometallics
2013, 32, 3327−3338. (d) Jin, D.; Williard, P. G.; Hazari, N.;
Bernskoetter, W. H. Effect of Sodium Cation on Metallacycle β-
Hydride Elimination in CO2−Ethylene Coupling to Acrylates. Chem. -
Eur. J. 2014, 20, 3205−3211. (e) Jin, D.; Schmeier, T. J.; Williard, P.
G.; Hazari, N.; Bernskoetter, W. H. Lewis Acid Induced β-Elimination
from a Nickelalactone: Efforts toward Acrylate Production from CO2
and Ethylene. Organometallics 2013, 32, 2152−2159. (f) Fischer, R.;
■
sı
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.organomet.9b00659.
Details of NMR experiments for kinetic analysis and
NMR spectra for novel compounds (PDF)
AUTHOR INFORMATION
Corresponding Author
̈
Langer, J.; Malassa, A.; Walther, D.; Gorls, H.; Vaughan, G. A key step
■
in the formation of acrylic acid from CO2 and ethylene: the
transformation of a nickelalactone into a nickel-acrylate complex.
Chem. Commun. 2006, 2510−2512. (g) Lee, S. Y. T.; Ghani, A. A.;
Nobuharu Iwasawa − Tokyo Institute of Technology,
Tokyo, Japan; orcid.org/0000-0001-6323-6588;
Email: niwasawa@chem.titech.ac.jp
D’Elia, V.; Cokoja, M.; Herrmann, W. A.; Basset, J.-M.; Kuhn, F. E.
̈
Liberation of methyl acrylate from metallalactone complexes via M−
O ring opening (M = Ni, Pd) with methylation agents. New J. Chem.
2013, 37, 3512−3517.
Other Authors
Kohei Takahashi − Tokyo Institute of Technology, Tokyo,
Japan
Yo Hirataka − Tokyo Institute of Technology, Tokyo,
Japan
Tatsuyoshi Ito − Tokyo Institute of Technology, Tokyo,
Japan
́
́
(4) (a) Lejkowski, M. L.; Lindner, R.; Kageyama, T.; Bodizs, G. E.;
Plessow, P. N.; Muller, I. B.; Schafer, A.; Rominger, F.; Hofmann, P.;
̈
̈
Futter, C.; Schunk, S. A.; Limbach, M. The First Catalytic Synthesis of
an Acrylate from CO2 and an Alkene−A Rational Approach. Chem. -
Eur. J. 2012, 18, 14017−14025. (b) Hendriksen, C.; Pidko, E. A.;
̈
Yang, G.; Schaffner, B.; Vogt, D. Catalytic Formation of Acrylate from
Carbon Dioxide and Ethene. Chem. - Eur. J. 2014, 20, 12037−12040.
(c) Huguet, N.; Jevtovikj, I.; Gordillo, A.; Lejkowski, M. L.; Lindner,
R.; Bru, M.; Khalimon, A. Y.; Rominger, F.; Schunk, S. A.; Hofmann,
P.; Limbach, M. Nickel-Catalyzed Direct Carboxylation of Olefins
with CO2: One-Pot Synthesis of α,β-Unsaturated Carboxylic Acid
Salts. Chem. - Eur. J. 2014, 20, 16858−16862. (d) Manzini, S.;
Huguet, N.; Trapp, O.; Schaub, T. Palladium- and Nickel-Catalyzed
Synthesis of Sodium Acrylate from Ethylene, CO2, and Phenolate
Bases: Optimization of the Catalytic System for a Potential Process.
Eur. J. Org. Chem. 2015, 2015, 7122−7130. (e) Knopf, I.; Tofan, D.;
Beetstra, D.; Al-Nezari, A.; Al-Bahily, K.; Cummins, C. C. A family of
cis-macrocyclic diphosphines: modular, stereoselective synthesis and
application in catalytic CO2/ethylene coupling. Chem. Sci. 2017, 8,
1463−1468. (f) Vavasori, A.; Calgaro, L.; Pietrobon, L.; Ronchin, L.
The coupling of carbon dioxide with ethene to produce acrylic acid
sodium salt in one pot by using Ni(II) and Pd(II)-phosphine
complexes as precatalysts. Pure Appl. Chem. 2018, 90, 315−326.
(g) Hopkins, M. N.; Shimmei, K.; Uttley, K. B.; Bernskoetter, W. H.
Synthesis and Reactivity of 1,2-Bis(di-iso-propylphosphino)benzene
Nickel Complexes: A Study of Catalytic CO2−Ethylene Coupling.
Organometallics 2018, 37, 3573−3580. (h) Takahashi, K.; Cho, K.;
Iwai, A.; Ito, T.; Iwasawa, N. Development of N-Phosphinomethyl-
Substituted NHC-Nickel(0) Complexes as Robust Catalysts for
Acrylate Salt Synthesis from Ethylene and CO2. Chem. - Eur. J. 2019,
25, 13504−13508.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.organomet.9b00659
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This research was supported by JSPS KAKENHI Grant
Numbers 15H05800 and 17H06143.
■
REFERENCES
■
(1) (a) Limbach, M. Acrylates from Alkenes and CO2, the Stuff That
Dreams Are Made of. Adv. Organomet. Chem. 2015, 63, 175−202.
(b) Hollering, M.; Dutta, B.; Kuhn, F. E. Transition metal mediated
̈
coupling of carbon dioxide and ethene to acrylic acid/acrylates. Coord.
Chem. Rev. 2016, 309, 51−67. (c) Wang, X.; Wang, H.; Sun, Y.
Synthesis of Acrylic Acid Derivatives from CO2 and Ethylene. Chem.
2017, 3, 211−228. (d) Schaub, T.; Hashmi, A. S. K.; Paciello, R. A.
Tackling Challenges in Industrially Relevant Homogeneous Catalysis:
The Catalysis Research Laboratory (CaRLa), an Industrial-Academic
Partnership. J. Org. Chem. 2019, 84, 4604−4614.
(2) (a) Hoberg, H.; Shaefer, D. NICKEL(0)-INDUZIERTE C−C-
̈
VERKNUPFUNG ZWISCHEN KOHLENDIOXID UND ETHYL-
(5) (a) Stieber, S. C. E.; Huguet, N.; Kageyama, T.; Jevtovikj, I.;
Ariyananda, P.; Gordillo, A.; Schunk, S. A.; Rominger, F.; Hofmann,
P.; Limbach, M. Acrylate formation from CO2 and ethylene: catalysis
with palladium and mechanistic insight. Chem. Commun. 2015, 51,
10907−10909. (b) Manzini, S.; Huguet, N.; Trapp, O.; Paciello, R.
A.; Schaub, T. Synthesis of acrylates from olefins and CO2 using
sodium alkoxides as bases. Catal. Today 2017, 281, 379−386.
(c) Manzini, S.; Cadu, A.; Schmidt, A.-C.; Huguet, N.; Trapp, O.;
Paciello, R.; Schaub, T. Enhanced Activity and Recyclability of
Palladium Complexes in the Catalytic Synthesis of Sodium Acrylate
EN SOWIE MONO- ODER DI-SUBSTITUIERTEN ALKENEN. J.
Organomet. Chem. 1983, 251, C51−C53. (b) Hoberg, H.; Schaefer,
D.; Burkhart, G.; Kru
̈
ger, C.; Romao, M. J. NICKEL(0)-
̈
INDUZIERTE C−C-VERKNUPFUNG ZWISCHEN KOHLEN-
DIOXID UND ALKINEN SOWIE ALKENEN. J. Organomet.
Chem. 1984, 266, 203−224. (c) Hoberg, H.; Peres, Y.; Milchereit,
̈
A. C−C-VERKNUPFUNG VON ALKENEN MIT CO2 AN
̈
NICKEL(0); n-PENTENSAUREN AUS ETHEN. J. Organomet.
Chem. 1986, 307, C41−C43. (d) Hoberg, H.; Peres, Y.; Kru
̈
ger, C.;
Tsay, Y.-H. A 1-Oxa-2-nickela-5-cyclopentanone from Ethene and
J
https://dx.doi.org/10.1021/acs.organomet.9b00659
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
from Carbon Dioxide and Ethylene. ChemCatChem 2017, 9, 2269−
molecules (CH2O, CO2) and NaBPh4. J. Organomet. Chem. 1993,
463, 215−221. (b) Knopf, I.; Courtemanche, M.-A.; Cummins, C. C.
Cobalt Complexes Supported by cis-Macrocyclic Diphosphines:
Synthesis, Reactivity, and Activity toward Coupling Carbon Dioxide
and Ethylene. Organometallics 2017, 36, 4834−4843.
(13) Ito, T.; Takahashi, K.; Iwasawa, N. Reactivity of a
Ruthenium(0) Complex Bearing a Tetradentate Phosphine Ligand:
Applications to Catalytic Acrylate Salt Synthesis from Ethylene and
CO2. Organometallics 2019, 38, 205−209.
2274.
́
(6) (a) Papai, I.; Schubert, G.; Mayer, I.; Besenyei, G.; Aresta, M.
Mechanistic Details of Nickel(0)-Assisted Oxidative Coupling of CO2
with C2H4. Organometallics 2004, 23, 5252−5259. (b) Graham, D. C.;
Mitchell, C.; Bruce, M. I.; Metha, G. F.; Bowie, J. H.; Buntine, M. A.
Production of Acrylic Acid through Nickel-Mediated Coupling of
Ethylene and Carbon Dioxide−A DFT Study. Organometallics 2007,
̈
26, 6784−6792. (c) Plessow, P. N.; Schafer, A.; Limbach, M.;
Hofmann, P. Acrylate Formation from CO2 and Ethylene Mediated
by Nickel Complexes: A Theoretical Study. Organometallics 2014, 33,
3657−3668. (d) Guo, W.; Michel, C.; Schwiedernoch, R.; Wischert,
R.; Xu, X.; Sautet, P. Formation of Acrylates from Ethylene and CO2
on Ni Complexes: A Mechanistic Viewpoint from a Hybrid DFT
Approach. Organometallics 2014, 33, 6369−6380. (e) Yang, G.;
(14) The ruthenalactone underwent a retro-oxidative cyclization
reaction slowly to give the ethylene-coordinated complex accom-
panied by a carbonate-type complex at 100 °C. See ref 13.
(15) Jia, G.; Drouin, S. D.; Jessop, P. G.; Lough, A. J.; Morris, R. H.
Use of the New Ligand P(CH2CH2PCy2)3 in the Synthesis of
Dihydrogen Complexes of Iron(II) and Ruthenium(II). Organo-
metallics 1993, 12, 906−916.
̈
Schaffner, B.; Blug, M.; Hensen, E. J. M.; Pidko, E. A. A Mechanistic
Study of Ni-catalyzed Carbon Dioxide Coupling with Ethylene
towards the Manufacture of Acrylic Acid. ChemCatChem 2014, 6,
800−807. (f) Li, Y.; Liu, Z.; Cheng, R.; Liu, B. Mechanistic Aspects of
Acrylic Acid Formation from CO2-Ethylene Coupling over Palladium-
and Nickel-based Catalysts. ChemCatChem 2018, 10, 1420−1430.
(g) Li, Y.; Liu, Z.; Zhang, J.; Cheng, R.; Liu, B. Insights into the Base-
Assisted Acrylate Formation from CO2/C2H4 Coupling by Pd- and
Ni-catalyst: A DFT Mechanistic Study. ChemCatChem 2018, 10,
5669−5678.
(16) King, R. B.; Kapoor, P. N. A New Synthesis of Polytertiary
Phosphines and Arsines. J. Am. Chem. Soc. 1969, 91, 5191−5192.
(17) Ziebart, C.; Federsel, C.; Anbarasan, P.; Jackstell, R.; Baumann,
W.; Spannenberg, A.; Beller, M. Well-Defined Iron Catalyst for
Improved Hydrogenation of Carbon Dioxide and Bicarbonate. J. Am.
Chem. Soc. 2012, 134, 20701−20704.
(18) (a) Wong, W.-K.; Chiu, K. W.; Statler, J. A.; Wilkinson, G.;
Motevalli, M.; Hursthouse, M. B. ALKYL, HYDRIDO AND
RELATED COMPOUNDS OF RUTHENIUM WITH TRIME-
THYLPHOSPHINE AND BIS(1,2-DIMETHYLPHOSPHINO)-
ETHANE. X-RAY CRYSTAL STRUCTURES OF CIS-HYDRIDO-
ETHYLTETRAKIS(TRIMETHYLPHOSPHINE)-RUTHENIUM-
(II) AND ETHYLENE TETRAKIS-(TRIMETHYLPHOSPHINE)-
RUTHENIUM(0). Polyhedron 1984, 3, 1255−1265. (b) Burn, M. J.;
Fickes, M. G.; Hollander, F. J.; Bergman, R. G. Reactions of
(PMe3)4Ru(C2H4) and (DMPE)2Ru(C2H4) with Weak Proton-
Donating Electrophiles HX (X = OAr, SAr, NHPh, PHPh). Synthesis
of Complexes with Metal-Heteroatom Single Bonds. Organometallics
1995, 14, 137−150. (c) Dahlenburg, L.; Kerstan, S.; Werner, D.
Oligophosphan-Liganden XXXIII. Inter- und intramolekulare Spal-
(7) (a) Alvarez, R.; Carmona, E.; Cole-Hamilton, D. J.; Galindo, A.;
́
Gutierrez-Puebla, E.; Monge, A.; Poveda, M. L.; Ruiz, C. Formation
of Acrylic Acid Derivatives from the Reaction of CO2 with Ethylene
Complexes of Molybdenum and Tungsten. J. Am. Chem. Soc. 1985,
107, 5529−5531. (b) Alvarez, R.; Carmona, E.; Galindo, A.;
́
Gutierrez, E.; Marín, J. M.; Monge, A.; Poveda, M. L.; Ruiz, C.;
Savariault, J. M. Formation of Carboxylate Complexes from the
Reactions of CO2 with Ethylene Complexes of Molybdenum and
Tungsten. X-ray and Neutron Diffraction Studies. Organometallics
1989, 8, 2430−2439. (c) Schubert, G.; Papai, I. Acrylate Formation
via Metal-Assisted C−C Coupling between CO2 and C2H4: Reaction
Mechanism as Revealed from Density Functional Calculations. J. Am.
Chem. Soc. 2003, 125, 14847−14858. (d) Collazo, C.; del Mar
Conejo, M.; Pastor, A.; Galindo, A. Synthesis and reactivity of
bis(ethylene)-phosphite complexes of molybdenum(0). Inorg. Chim.
Acta 1998, 272, 125−130. (e) Wolfe, J. M.; Bernskoetter, W. H.
Reductive functionalization of carbon dioxide to methyl acrylate at
zerovalent tungsten. Dalton Trans. 2012, 41, 10763−10768.
(f) Bernskoetter, W. H.; Tyler, B. T. Kinetics and Mechanism of
Molybdenum-Mediated Acrylate Formation from Carbon Dioxide
and Ethylene. Organometallics 2011, 30, 520−527. (g) Hanna, B. S.;
MacIntosh, A. D.; Ahn, S.; Tyler, B. T.; Palmore, G. T. R.; Williard, P.
G.; Bernskoetter, W. H. Ancillary Ligand Effects on Carbon Dioxide-
Ethylene Coupling at Zerovalent Molybdenum. Organometallics 2014,
33, 3425−3432.
tungen von sp2- und sp3-C−H-Bindungen an L4Ru0-Komplexru
̈
mpfen
m i t P ( o - C 6 H 4 P M e 2 ) 3 , N ( C H 2 C H 2 P M e 2 ) 3 , M e P -
(CH2CH2CH2PMe2)2/PR3 und MeSi(CH2PMe2)3/PR3 (R = Me,
OMe) als Stutzliganden. J. Organomet. Chem. 1991, 411, 457−469.
̈
(d) Bianchini, C.; Perez, P. J.; Peruzzini, M.; Zanobini, F.; Vacca, A.
Classical and Nonclassical Polyhydride Ruthenium(II) Complexes
Stabilized by the Tetraphosphine P(CH2CH2PPh2)3. Inorg. Chem.
1991, 30, 279−287.
(19) A similar correlation between the basicity of phosphine ligands
and 13C NMR chemical shifts of coordinated olefinic carbons was
reported for Ni(PR3)2(ethyl methacrylate) (PR3 = PPh3, PPh2Et,
PPhMe2, PPhEt2, PEt3) complexes: Yamamoto, T.; Ishizu, J.; Komiya,
S.; Nakamura, Y.; Yamamoto, A. NMR STUDIES OF PHOSPHINE-
NICKEL(0) COMPLEXES OF ETHYL METHACRYLATE Ni-
(PR3)2(CH2C(CH3)COOC2H5). J. Organomet. Chem. 1979, 171,
103−112.
(8) Cohen, S. A.; Bercaw, J. E. Titanacycles Derived from Reductive
Coupling of Nitriles, Alkynes, Acetaldehyde, and Carbon Dioxide
with Bis(pentamethylcyclopentadienyl)(ethylene)titanium(II). Orga-
nometallics 1985, 4, 1006−1014.
(9) Alt, H. G.; Denner, C. E. Metallacyclen des Zirkonocens. J.
Organomet. Chem. 1990, 390, 53−60.
i
(20) The effect of the steric bulkiness of Cy and Pr groups on the
phosphorus atoms of Ru(PP3) type complexes has been described:
Field, L. D.; Guest, R. W.; Vuong, K. Q.; Dalgarno, S. J.; Jensen, P.
Iron(0) and Ruthenium(0) Complexes of Dinitrogen. Inorg. Chem.
2009, 48, 2246−2253. See also ref 15.
(10) Hessen, B.; Meetsma, A.; van Bolhuis, F.; Teuben, J. H.;
Helgesson, G.; Jagner, S. Chemistry of Carbon Monoxide Free
Cyclopentadienylvanadium(I) Alkene and Alkyne Complexes. Orga-
nometallics 1990, 9, 1925−1936.
(21) Dissociation of one of the phosphorus atoms might occur in the
case of tetradentate phosphine ligands bearing an ethylene tether due
to their flexibility. As an example, tridentate coordination of
P(CH2CH2PtBu2)3 was reported: (a) Gilbert-Wilson, R.; Field, L.
D.; Bhadbhade, M. M. New Superhindered Polydentate Polyphos-
phine Ligands P(CH2CH2PtBu2)3, PhP(CH2CH2PtBu2)2, P-
(CH2CH2CH2PtBu2)3, and their Ruthenium(II) Chloride Complexes.
Inorg. Chem. 2012, 51, 3239−3246. See also ref 15 and: (b) Field, L.
D.; Li, H. L.; Dalgarno, S. J.; McIntosh, R. D. Base-Induced
Dehydrogenation of Ruthenium Hydrazine Complexes. Inorg. Chem.
2013, 52, 1570−1583.
(11) (a) Hoberg, H.; Jenni, K.; Angermund, K.; Kruger, C. CC-
̈
Linkages of Ethene with CO2 on an Iron(0) Complex−Synthesis and
Crystal Structure Analysis of [(PEt3)2Fe(C2H4)2]. Angew. Chem., Int.
Ed. Engl. 1987, 26, 153−155. (b) Rummelt, S. M.; Zhong, H.;
Korobkov, I.; Chirik, P. J. Iron-Mediated Coupling of Carbon Dioxide
and Ethylene: Macrocyclic Metallalactones Enable Access to Various
Carboxylates. J. Am. Chem. Soc. 2018, 140, 11589−11593.
(12) (a) Aresta, M.; Quaranta, E. Synthesis, characterization and
reactivity of [Rh(bpy)(C2H4)Cl]. A study on the reaction with C1
K
https://dx.doi.org/10.1021/acs.organomet.9b00659
Organometallics XXXX, XXX, XXX−XXX

Organometallics
pubs.acs.org/Organometallics
Article
t
̌
(22) This mixed solvent system was employed for RuL1_E, RuL2_E
and RuL4_E to dissolve the complexes. For RuL3_E, only DMA was
used because use of THF/DMA solvent system significantly reduced
,
aggregate of KO Bu has been reported. See: (a) Halaska, V.;
Lochmann, L.; Lím, D. ASSOCIATION DEGREE OF t-BUT-
OXIDES OF ALKALI METALS IN APROTIC SOLVENTS. Collect.
Czech. Chem. Commun. 1968, 33, 3245−3253. (b) Schmidt, P.;
Lochmann, L.; Schneider, B. STRUCTURE AND VIBRATIONAL
SPECTRA OF THE SODIUM AND POTASSIUM tert-BUTOX-
IDES. J. Mol. Struct. 1971, 9, 403−411. (c) Chisholm, M. H.; Drake,
S. R.; Naiini, A. A.; Streib, W. E. SYNTHESIS AND X-RAY
CRYSTAL STRUCTURES OF THE ONE-DIMENSIONAL RIB-
BON CHAINS [MOBut• ButOH]∞ AND THE CUBANE SPECIES
[MOBut]4 (M = K AND Rb). Polyhedron 1991, 10, 337−345.
(33) As another pathway which can be first order on the amount of
KOtBu, deprotonation of the β-agostic complex formed by
dissociation of carboxylate is also conceivable. An increase in the
acidity of the sp3 C−H bond via a β-agostic interaction was reported:
Bezdek, M. J.; Chirik, P. J. Proton-Coupled Electron Transfer to a
Molybdenum Ethylene Complex Yields a β−agostic Ethyl: Structure,
Dynamics and Mechanism. J. Am. Chem. Soc. 2018, 140, 13817−
13826.
the yield of RuL3_lac
.
(23) The yield of ruthenalactone RuL3_lac after 17 h was 48%, which
was lower than that in our previous report (71%), and the major side
product was carbonate complex Ru(L3)(CO3) (27%). Such a
difference could be derived from the difference in the reaction
conditions (this work, under ethylene 0.5 atm, CO2 0.5 atm; previous
work, under CO2 1.0 atm). The formation of Ru(L3)(CO3) was also
noted in our previous work,13 and it is thought to be generated via
retro-oxidative cyclization of ruthenalactone RuL3_lac to give ethylene-
coordinated Ru(0) complex RuL3_E, followed by reductive dispro-
portionation of CO2 to CO and CO32−. For selected examples for
reductive disproportionation of CO2 on zerovalent ruthenium
complexes, see: (a) Allen, D. L.; Green, M. L. H.; Bandy, J. A.
Reactions of Carbon Dioxide with Electron-rich Trimethylphosphine
Compounds of Rhenium and Tungsten: Crystal Structure of
[W(PMe3)4H2(CO3)]. J. Chem. Soc., Dalton Trans. 1990, 541−549.
(b) Nakajima, H.; Kushi, Y.; Nagao, H.; Tanaka, K. Multistep CO2
Reduction Catalyzed by [Ru(bpy)2(qu)(CO)]2+ (bpy = 2,2′-
Bipyridine, qu = Quinoline). Double Methylation of the Carbonyl
Moiety Resulting from Reductive Disproportionation of CO2.
Organometallics 1995, 14, 5093−5098.
(24) For the complexes with L1 and L2, ruthenalactone RuL_lac was
not observed during the conversion of RuL_E to RuL_acr, which
indicated that the conversion of RuL_E to RuL_lac is the rate-
determining step. Therefore, the reactivity for the oxidative cyclization
was estimated on the basis of the rate of the generation of RuL_lac for
L3 and RuL_acr for L1 and L2.
(25) Another explanation that the rapid β-H elimination on the RuL1
complex is due to the dissociation of one of the PCy2 groups from the
ruthenium center to give an open coordination site cannot be ruled
out.
(34) Hydrido acetato complexes were used as precursors because
they are easily prepared and are stable.
(35) The TON at 140 °C (TON = 1.5) was lower than that of our
previous report employing RuL3_E (TON = 5) under similar
conditions. The difference might be derived from the difference in
the precursor complex (in this work, RuL3_OAc was employed).
(36) There has been only one example of achieving catalytic acrylate
salt synthesis by employing carbonate (K2CO3) as a base with TON =
2.4b
(26) The initial reaction rate was evaluated by the amount of
RuL3_acr formed due to the presence of a retro-oxidative cyclization
pathway, although it is thought to be minor.
(37) The details of the screening of the reaction conditions are
described in the Supporting Information.
(38) The stability of the ruthenium complexes, RuL1_acr, RuL2_acr
,
(27) The reaction was slightly accelerated with an increased amount
of KOtBu. One possible explanation is that the potassium cation
works as a weak Lewis acid to facilitate the dissociation of the acrylate
anion (step A, Scheme 5c). Such an acceleration effect of the alkaline-
metal cation was observed for the cleavage of nickelalactone. See ref
3d. When the reaction was conducted with 250 equiv of KOtBu, the
reaction was not accelerated very much (about 5 times faster than the
reaction using 10 equiv), which indicated the rate dependence was
similar even in the presence of a large excess of KOtBu (See Figure
S16f, SI-4).
RuL3_acr, and RuL4_E were examined by heating them at 140 or 180 °C
in THF/DMA (1/1) or in DMA under an atmospheric pressure of
ethylene/CO2 (1/1) in the absence of base, and the reaction progress
was followed by 31P NMR. Although they became mixtures of several
complexes, free ligands were not detected and formation of Ru black
was not observed. To obtain information on the catalyst resting state,
a solution after a catalytic reaction using RuL3_OAc at 140 °C was
analyzed by ESI-MS, which suggested the presence of Ru(L3)(CO)
(m/z 799, Ru(L3)(CO) − H−) and RuL3_lac or RuL3_acr (m/z 843, m/
z 843, RuL3_lac or RuL3_acr − H−), whereas a 31P NMR spectrum of a
solution obtained under similar conditions showed many unidentified
signals. Details are described in SI-8 and SI-9.
́
(28) (a) Jimenez-Tenorio, M.; Puerta, M. C.; Valerga, P. Synthesis
and Characterization of Ruthenium Hydride Complexes Containing
the Bulky Diphosphine l,2-Bis(diisopropylphosphino)ethane (dippe).
Crystal Structure of [RuH(η2-O2)(dippe)2][BPh4]. Inorg. Chem.
̈
(39) Belyaev, A.; Dau, T. M.; Janis, J.; Grachova, E. V.; Tunik, S. P.;
Koshevoy, I. O. Low-Nuclearity Alkynyl d10 Clusters Supported by
Chelating Multidentate Phosphines. Organometallics 2016, 35, 3763−
3774.
1994, 33, 3515−3520. (b) Rocchini, E.; Mezzetti, A.; Ruegger, H.;
̈
Burckhardt, U.; Gramlich, V.; Zotto, A. D.; Martinuzzi, P.; Rigo, P.
Heterolytic H2 Activation by Dihydrogen Complexes. Effects of the
Ligand X in [M(X)H2{Ph2P(CH2)3PPh2}2]n+ (M = Ru, Os; X = CO,
Cl, H). Inorg. Chem. 1997, 36, 711−720.
(40) Mitchell, R. W.; Spencer, A.; Wilkinson, G. Carboxylato-
triphenylphosphine Complexes of Ruthenium, Cationic Triphenyl-
phosphine Complexes derived from them, and their Behaviour as
Homogeneous Hydrogenation Catalysts for Alkenes. J. Chem. Soc.,
Dalton Trans. 1973, 846−854.
(29) Osman, R.; Pattison, D. I.; Perutz, R. N.; Bianchini, C.; Casares,
J. A.; Peruzzini, M. Photochemistry of M(PP3)H2 (M = Ru, Os; PP3 =
P(CH2CH2PPh2)3): Preparative, NMR, and Time-Resolved Studies.
J. Am. Chem. Soc. 1997, 119, 8459−8473.
(30) In the regeneration of RuL_E, the bulkiness of the ligand may
not be significant because different reaction rates were observed (L1 >
L2 > L3 > L4) in spite of the similarity of their steric hindrance
around ruthenium center (L1 and L2 have sec-alkyl groups on the
tethered phosphorus atoms and L3 and L4 have Ph groups).
(31) The reaction under an Ar atmosphere was not clean and
afforded several unidentified complexes.
̈
(41) Noel-Duchesneau, L.; Lugan, N.; Lavigne, G.; Labande, A.;
́
Cesar, V. Tailoring Buchwald-Type Phosphines with Pyrimidinium
Betaines as Versatile Aryl Group Surrogates. Organometallics 2014, 33,
5085−5088.
(32) This downward curvature was probably due to the formation of
an aggregate of KOtBu at higher concentration. Formation of an
L
https://dx.doi.org/10.1021/acs.organomet.9b00659
Organometallics XXXX, XXX, XXX−XXX