Scalable and Chemoselective Synthesis of ?-Keto Esters and Acids via Pd-Catalyzed Carbonylation of Cyclic β-Chloro Enones

Article abstract of DOI:10.1021/acs.organomet.8b00468

The Pd-catalyzed carbonylation of cyclic β-chloro enones using simple phosphine ligands is described. Screening identified P(Me)(t-Bu)₂ as the most general ligand for an array of chloro enone electrophiles. The reaction scope has been evaluated on a milligram scale across 80 examples, with excellent reactivity observed in nearly every case. Carbonylation can be achieved even in the presence of potentially sensitive or inhibitory functional groups, including basic nitrogens as well as aryl chlorides or bromides. Twenty examples have been run on a gram scale, demonstrating scalability and practical utility. Using P(Me)(t-Bu)₂, the reaction rate depends on both nucleophile and electrophile identity, with completion times varying between 3 and >18 h under a standard set of conditions. Switching to P(t-Bu)₃ for the carbonylation of 3-chlorocyclohex-2-enone with methanol results in a dramatic rate increase, enabling effective catalysis with kinetics consistent with rate-limiting mass transfer. Stoichiometric oxidative addition of 3-chlorocyclohex-2-enone and 3-oxocyclohex-1-enecarbonyl chloride to both Pd[P(t-Bu)₃]₂ and Pd(PCy₃)₂ has enabled characterization and isolation of several potential catalytic intermediates, including Pd-vinyl and Pd-acyl species supported by P(t-Bu)₃ and PCy₃ ligands. Monitoring the oxidative addition of 3-chlorocyclohex-2-enone to Pd(PCy₃)₂ by NMR spectroscopy indicates that coordination of the alkene precedes oxidative addition. As a result of these studies, methyl 3-oxocyclohex-1-enecarboxylate has been synthesized via Pd-catalyzed carbonylation of 3-chlorocyclohex-2-enone in 90% yield on a 60 g scale with only 0.5 mol % catalyst loading.

Full text of DOI:10.1021/acs.organomet.8b00468

Article
pubs.acs.org/Organometallics
Cite This: Organometallics XXXX, XXX, XXX−XXX
Scalable and Chemoselective Synthesis of γ‑Keto Esters and Acids
via Pd-Catalyzed Carbonylation of Cyclic β‑Chloro Enones
†
†
†
‡
Justin M. Kaplan, Damian P. Hruszkewycz, Iulia I. Strambeanu, Christopher J. Nunn,
†
†
§
§
Kelsey F. VanGelder, Anna L. Dunn, Derek I. Wozniak, Graham E. Dobereiner,
and David C. Leitch*
,
†
†
API Chemistry, GlaxoSmithKline, King of Prussia, Pennsylvania, United States
‡
Product and Process Engineering, GlaxoSmithKline, King of Prussia, Pennsylvania, United States
Department of Chemistry, Temple University, Philadelphia, Pennsylvania, United States
§
*
S Supporting Information
ABSTRACT: The Pd-catalyzed carbonylation of cyclic
β-chloro enones using simple phosphine ligands is described.
Screening identified P(Me)(t-Bu) as the most general ligand
2
for an array of chloro enone electrophiles. The reaction scope
has been evaluated on a milligram scale across 80 examples,
with excellent reactivity observed in nearly every case. Car-
bonylation can be achieved even in the presence of potentially
sensitive or inhibitory functional groups, including basic nitro-
gens as well as aryl chlorides or bromides. Twenty examples
have been run on a gram scale, demonstrating scalability and
practical utility. Using P(Me)(t-Bu) , the reaction rate
2
depends on both nucleophile and electrophile identity, with
completion times varying between 3 and >18 h under a standard set of conditions. Switching to P(t-Bu) for the carbonylation
3
of 3-chlorocyclohex-2-enone with methanol results in a dramatic rate increase, enabling effective catalysis with kinetics
consistent with rate-limiting mass transfer. Stoichiometric oxidative addition of 3-chlorocyclohex-2-enone and 3-oxocyclohex-1-
enecarbonyl chloride to both Pd[P(t-Bu) ] and Pd(PCy ) has enabled characterization and isolation of several potential
3
2
3 2
catalytic intermediates, including Pd−vinyl and Pd−acyl species supported by P(t-Bu) and PCy ligands. Monitoring the
3
3
oxidative addition of 3-chlorocyclohex-2-enone to Pd(PCy ) by NMR spectroscopy indicates that coordination of the alkene
3
2
precedes oxidative addition. As a result of these studies, methyl 3-oxocyclohex-1-enecarboxylate has been synthesized via
Pd-catalyzed carbonylation of 3-chlorocyclohex-2-enone in 90% yield on a 60 g scale with only 0.5 mol % catalyst loading.
INTRODUCTION
use substrates with weaker aryl or vinyl C−I bonds; effective
■
use of unactivated aryl or vinyl chlorides is considerably
Metal-catalyzed carbonylation is a powerful and extensively
5
,7
rarer. Unfortunately, the stepwise installation and subse-
quent elimination of iodide is not mass efficient, making
Pd-catalyzed carbonylation unattractive from an economic and
environmental perspective. Even in the few reported carbon-
applied method for the introduction of CO as a C unit into
1
1
organic compounds. The current abundance of CO available
from methane, the promise of carbon-neutral reductions of
2
CO to CO, and the excellent atom economy of many carbon-
7
2
ylations of unactivated aryl or vinyl chlorides, the generally
ylation reactions make CO the ideal ambiphilic C source. It is
1
high reaction temperatures and relatively narrow substrate
scope limit practical applications in the preparation of func-
tionally complex molecules.
During chemical development for a clinical candidate API,
we required rapid access to large quantities of several cyclic
γ-keto ester derivatives. A survey of existing methods to prepare
therefore no surprise that carbonylative processes are among
the highest volume applications of homogeneous catalysis in
3
,4
industry. In the fine chemical, agrochemical, and pharma-
ceutical sectors, Pd-catalyzed carbonylation is an efficient and
direct method to construct aldehyde, ketone, and carboxylic
5
acid derivatives in a single multicomponent reaction.
these compounds revealed that allylic oxidation using CrO is
3
Many Pd-catalyzed cross-couplings can be modified to per-
form as carbonylations; however, CO forms very strong bonds
with Pd(0), occupying a coordination site and stabilizing the low
oxidation state through synergistic bonding. This raises energy
barriers for the oxidative addition of C−X bonds, often making
8−13
nearly universally preferred,
despite the environmental and
Special Issue: The Roles of Organometallic Chemistry in Pharma-
ceutical Research and Development
5
−7
this step turnover limiting in carbonylation catalysis.
As a
Received: July 5, 2018
result, the vast majority of reported Pd-catalyzed carbonylations
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.8b00468
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 1. Approach to γ-Keto Esters via Carbonylation and
Examples of the Synthetic Versatility of These Products
potassium carbonate) and two solvents (acetonitrile and
toluene); the other reaction parameters (time, temperature,
and CO pressure) were fixed. These reactions were conducted
in plates using an HEL Cat96 high-throughput pressure
reactor, enabling rapid execution of nearly 200 experiments.
Each run was performed using an internal standard (biphenyl),
which doubles as both an HPLC and an NMR handle for
quantification of the remaining starting material and the
19
desired product. The output of this parallel-in-parallel screen
is depicted graphically in Figure 1.
While it is clear from the data that each substrate behaves
differently, there are options for a general set of conditions.
Both 1a and 2a work well with nearly every ligand tested under
multiple base/solvent combinations, while α-substituted chloro
enones (3a−8a) are less reactive. For the most sterically
7k,l
demanding substrates, dcpp, and P(Me)(t-Bu) are more
2
effective. In addition, there are clear solvent and base effects on
reactivity: CH CN and K CO generally give higher conver-
3
2
3
sion but also greater byproduct formation. In particular, forma-
tion of the corresponding carboxylic acid is enhanced using
CH CN and/or K CO , undoubtedly due to adventitious
3
2
3
water originating from these hygroscopic materials.
Across this 192-experiment array, the most general system is
safety issues with use of Cr(VI) reagents (Scheme 1A). This
method would also be incompatible with substrates containing
P(Me)(t-Bu) with NEt in either CH CN or toluene. These
2
2
3
3
other oxidizable sites, such as where R = alkyl. An alternative
conditions are tolerant of all the assessed functional groups and
are even able to achieve chemoselective carbonylation of 8a.
14
approach involves Birch reduction of m-anisic acid derivatives,
which also carries significant safety, scalability, and chemo-
selectivity concerns (Scheme 1B). We therefore sought both a
versatile, redox-neutral synthetic method in the near term and
a viable long-term solution to access these compounds on
scale.
While dcpp, dppp, DPEphos, and P(t-Bu) all result in mix-
3
tures of carbonylation at both Cl and Br (observed by LCMS),
P(Me)(t-Bu) gives good selectivity for carbonylation at only
2
19
Cl versus at both Cl and Br (∼5:1 desired to overreacted).
This is, to the best of our knowledge, the first example of a syn-
Herein we describe the development of the Pd-catalyzed
carbonylation of readily accessible cyclic β-chloro enones to
2
thetically useful chemoselective carbonylation of a C(sp )−Cl
2
17d,20
8
−12
over a C(sp )−Br bond.
These results highlight the power
access highly versatile
α,β-unsaturated γ-keto esters and
of multivariate screening for reaction optimization: screening
only one of these substrates in a univariate manner would not
have revealed a general system or yield alternatives specific to
acids in high yield with a broad substrate scope (Scheme 1C).
We hypothesized that the activated nature of the chloro enone
moiety (a vinylogous acid chloride) would enable effective
oxidative addition even in the presence of CO. Accordingly,
these carbonylations give good to excellent yields without the
each substrate (e.g., dcpp for 3a, PCy for 7a and 8a).
3
Carbonylation Scope and Preparative-Scale Synthe-
ses. After identifying a general set of conditions for carbon-
ylation of 1a−8a with methanol, we tested these conditions
against an array of 80 substrate combinations on a 0.1 mmol
scale. Ten functionalized alcohols with varying steric and elec-
tronic properties (b−k) were paired with 1a−8a (Figure 2).
This set includes a number of challenging alcohol substrates
5
need for the alcohol substrate to be used as the solvent. Car-
bonylation of α-substituted β-chloro enones (easily accessed
via alkylation or arylation of the precursor diones) generates
highly functionalized tetrasubstituted alkenes. Chemoselective
carbonylation at the vinyl chloride position is possible in the
presence of a variety of functional groups, including aryl
chlorides or bromides. Finally, these reactions proceed at
practical rates with catalyst loadings as low as 0.5 mol %.
(
tert-butyl alcohol, trifluoroethanol, phenol, 2-(2-pyridyl)-
ethanol), as well as water, which could hydrolyze the chloro
enone substrates back to the 1,3-diones. This array was run in a
plate-based format analogous to the screen from Figure 1, with
products identified by LCMS and assessed semiquantiatively
RESULTS AND DISCUSSION
High-Throughput Reaction Screening. Given the large
■
19
using LC area percent values; 16 examples were further
number of palladium catalysts reported for carbonylation reac-
1
5
analyzed by H NMR spectroscopy to confirm solution yields,
tions and the anticipated challenges associated with achieving
1
5
which are in good agreement with the LC area percent values.
Across these 80 examples, there is remarkable consistency in
product formation using different alcohols, regardless of nucle-
ophilicity or sterics. Water does not seem to hydrolyze the
starting materials, allowing access to carboxylic acids in addi-
tion to esters. Only substrates 5a and 8a exhibit low conver-
sions in some cases, especially with sterically hindered alcohols.
Menthol, one such hindered alcohol, can be used to install an
chemoselective carbonylation of β-chloro enones, we began
by evaluating eight electrophiles with various functional groups
1
6
(
1a−8a) in a “parallel-in-parallel” screening format (Figure 1).
We screened six simple phosphine ligands (PCy , P(t-Bu) ,
3
3
P(Me)(t-Bu) , dppp, DPEphos, and dcpp) that have either been
2
reported for carbonylation reactivity or is a close analogue (such
1
7
as P(Me)(t-Bu) as a replacement for CataCXium A). We also
2
1
8
chose to use the phosphonium tetrafluoroborate salts for all
trialkylphosphines in order to ensure reproducibility on a small
scale and to simplify automated solid dispensing. In addition to
ligand diversity, we also evaluated two bases (triethylamine and
1
0b
inexpensive chiral auxiliary.
Even potentially chelating
alcohols such as h and k are effective coupling partners, giving
highly functionalized ester products in good yields. Finally,
B
DOI: 10.1021/acs.organomet.8b00468
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 1. “Parallel-in-parallel” screening design and results. For data visualization, increasing size indicates higher substrate conversion; the color
gradient indicates the HPLC solution yield (red, 0%; yellow, 50%; green, 100%). Numbers correspond to solution yields (>60%) determined by
1
HPLC or H NMR spectroscopy (bold values) versus internal standard. See the Supporting Information for a full table of results.
both 7a and 8a exhibit chemoselectivity for carbonylation at
the vinyl−Cl over the other Ar−X positions.
the reactivity order 1a > 3a ≈ 6a > 5a. Similarly, increasing
nucleophile steric bulk leads to diminished rates, as observed
for the formation of 1c,f,g.
Twenty examples have been prepared on a gram scale to
establish the practical utility of this methodology for a number
of challenging substrate combinations (Figure 3). These
reactions were conducted at higher concentration with intense
mixing, enabling a reduction in both catalyst loading (2 mol %)
and temperature (80 °C). Reactions were run under identical
conditions (except for carbonylation of 8a) to enable direct
comparisons for the most general set of reaction conditions.
Despite being unoptimized, the isolated yields range from 44
to 88%, correlating well with solution yields from screening
reactions. In some cases, we observe small amounts of alkene
While the conditions from Figure 3 are a valuable starting
point for exploratory synthesis, we emphasize that the listed
yields are unoptimized and can certainly be improved with fur-
ther development. For example, during process optimization for
the production of 1b, which is commercially available in small
quantities with inconsistent quality, we re-evaluated P(t-Bu)3
as a ligand due to its greater availability on a large scale. Reac-
tion monitoring using the OSR revealed that the formation of
1b proceeds to 80−90% in ∼30 min, even at catalyst loadings
≤1 mol %. In fact, these results revealed an apparent zero-
order dependence on catalyst concentration, leading us to
1
isomerization in the crude H NMR spectra; however, these
19
isomeric byproducts are removed during chromatographic
suspect mass transfer as contributing to the rate. Obtaining
meaningful kinetic data under synthetically relevant conditions
proved extremely difficult, which prompted an exploration of
stoichiometric reactivity as an alternative means to study the
potential reaction mechanisms for the formation of 1b using
P(t-Bu)3.
1
9
purification. One exception is the acetophenone-substituted
ester 4b, which was isolated as an inseparable 2.3:1 mixture of
alkene isomers. For reactions with 8a, we observed only ∼10%
carbonylation at Br to give the diester byproduct; no evidence
for carbonylation at only Br to give a monoester byproduct was
1
9
observed.
Stoichiometric Reactivity and Isolation of Pd(II) Vinyl
and Acyl Species. The high reactivity of β-chloro enones
toward carbonylation is in stark contrast to difficult carbon-
We have also monitored several of these reactions over time
by liquid sampling using an Unchained Laboratories OSR
2
1
2
7,17d
apparatus, revealing a wide range of rates for different sub-
strate combinations (Figure 4). Comparing the rates of carbox-
ylic acid formation for 1c, 3c, 5c, or 6c indicates that the steric
properties of the electrophile have a large effect on rate, with
ylations of other C(sp )−Cl substrates.
As slow oxidative
addition of the C−Cl bond is often blamed for poor reactivity
of aryl and vinyl chlorides in Pd-catalyzed carbonylation, we
7
targeted our stoichiometric studies toward this reaction
C
DOI: 10.1021/acs.organomet.8b00468
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 2. Plate-based reaction scope evaluation between 1a−8a and b−k under one set of reaction conditions. For data visualization, increasing
size corresponds higher substrate conversion; the color gradient indicates the solution yield of the desired product (red, 0%; yellow, 50%; green,
1
1
00%). Numbers correspond to solution yields (>30%) determined by HPLC or H NMR spectroscopy (boldface values) versus internal standard.
See the Supporting Information for a full table of results.
Figure 3. Gram-scale synthesis of 20 γ-keto esters and acids with solution (NMR) and isolated yields.
(
Scheme 2). Monitoring the reaction of a 1:1 mixture of
along with ∼10% of free P(t-Bu) (Scheme 2A). Unfortu-
3
Pd[P(t-Bu) ] and 1a at 60 °C for 1 h reveals ∼10% of a new
species, which we propose is the oxidative addition complex 9,
nately, prolonged heating does not give an increase in 9 but
3
2
3
1
rather formation of multiple new Pd species (observed by P
D
DOI: 10.1021/acs.organomet.8b00468
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 2. Formation of Vinyl and Acyl Pd Species and
Observation of Byproducts during Catalysis with 0−1 atm
of CO
Figure 4. Reaction progress plots for selected examples.
NMR spectroscopy) and concomitant reductive homocoupling
of the cyclohexenone moiety to give 1l. After heating for 17 h
at 60−70 °C, there is an approximately 2:1 ratio of 1l to 1a,
with only minimal 9 present.
We attribute the minimal formation of 9 under these
conditions to the fact that oxidative addition to Pd[P(t-Bu)3]2
proceeds via the monophosphine complex after dissociation
22
of 1 equiv of P(t-Bu)3. As the reaction progresses, buildup of
free P(t-Bu) will disfavor phosphine dissociation from Pd[P(t-
3
Bu) ] , limiting the rate of oxidative addition. Subsequent
3
2
decomposition to 1l may occur via migratory insertion of the
Pd−C bond onto the alkene of another equivalent of 1a,
followed by β-chloride elimination (effectively a reductive
Heck coupling) or via disproportionation of 9 to generate a
bis(vinyl) palladium species and Pd[P(t-Bu) ]Cl , followed by
3
2
reductive elimination of 1l.
Generation of the corresponding acyl complex (10) was
accomplished by reacting Pd[P(t-Bu) ] with the acid chloride
3
2
1m. Rapid conversion to 10 (89% solution yield vs internal
standard) and 1 equiv of free P(t-Bu) is observed at room
3
1
temperature. The H NMR spectrum of complex 10 has a
diagnostic downfield singlet (7.5 ppm) for the vinyl C−H; a
13
broad downfield C NMR signal at 208.5 ppm is indicative of
a Pd−acyl unit. Unfortunately, attempts to isolate 10 have not
been successful. While 10 can be precipitated from toluene/
pentane in 50−60% yield, this material is not pure and com-
pletely decomposes in solution after 24 h at room temperature,
forming a complex mixture that includes 1l. In this case, forma-
23
tion of 1l must occur via decarbonylation to give 9, followed
by disproportionation; a Heck-type pathway cannot be oper-
ative, as 1a is not present. Thus, we favor disproportionation as
the likely mechanism for formation of 1l in our attempts to
prepare 9 from 1a and Pd[P(t-Bu) ] but cannot completely
rule out the Heck-type pathway. Finally, treatment of 10 with
excess MeOH and DIPEA directly generates 1b in 82%
solution yield at room temperature in <15 min.
transformation with only 1 atm of CO, leading to an 85% or
72% solution yield of 1b in toluene or acetonitrile, respectively.
While 3% of 1n is observed using acetonitrile, we do not
observe evidence for 1l in either solvent. Despite the absence
of 1l from these catalytic runs at 1 atm of CO, formation of 1l
and/or 1n is a possible failure mode if insufficient CO is
present. As a result, we have maintained at least 60 psi of CO
in our preparative-scale carbonylation processes.
3
2
Observation of reductive homocoupling in stoichimetric
reactions with P(t-Bu) led us to investigate its relevance under
3
catalytic conditions (Scheme 2B). In the absence of CO, both
1l and the direct substitution product 1n are formed, with the
product distribution dependent on the reaction solvent. This
indicates that MeOH can act both as a nucleophile and as a
reductant to effect catalytic turnover in the formation of 1l. To
assess whether low pCO could result in either 1l or 1n as
byproducts in the carbonylation of 1a, we performed this
In an effort to cleanly isolate a Pd−vinyl species derived
from oxidative addition of 1a, we explored the reactivity of
Pd(PCy ) under analogous conditions. While PCy is by no
means the most reactive ligand in the carbonylation chemistry,
it is active for carbonylation of 1a to 1b (Figure 1). Pd(PCy3)2
3
2
3
E
DOI: 10.1021/acs.organomet.8b00468
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
Figure 5. (top) Stack plot of H NMR spectra (2.4−6.9 ppm) acquired over time during the oxidative addition of 1a to Pd(PCy ) (time increases
3
2
front to back). (bottom) Reaction progress plot (concentrations determined by integration versus biphenyl as an internal standard or, in the case of
Pd(PCy ) , calculated assuming mass balance).
3
2
is known to undergo direct oxidative addition without phos-
phine dissociation, which we anticipated would mitigate the
been isolated in 64% yield by precipitation from heptane.
Unlike 10, this acyl species is stable upon isolation, undoubt-
edly because it is coordinatively and electronically saturated.
The NMR spectroscopic features of 12 are consistent with
those observed for 10, with downfield resonances in both the
24
conversion problem observed for Pd[P(t-Bu) ] . Finally,
3
2
Pd(PCy ) is readily available in pure form (unlike Pd[P-
3
2
[
(Me)(t-Bu) ] , which cannot be crystallized and is typically
2 2
2
5
1
13
generated in situ ).
H (8.0 ppm for the vinyl C−H) and C (240.5 ppm for the
Treating Pd(PCy ) with 1a results in smooth formation of
Pd-acyl carbon) NMR spectra. This species can also be gener-
ated by subjecting 11 to a CO atmosphere at 80 °C for 1 h,
giving a 1.4:1 mixture of 11 and 12; prolonged heating results
in decomposition. As for complex 10, treatment of 12 with
excess MeOH and DIPEA generates 1b (88% solution yield);
however, rather than being complete in minutes at room
temperature, ester formation from 11 requires more forcing
conditions (80 °C, 1 h).
3
2
the desired complex (11) (Scheme 2C), which has been
isolated (77% yield) and fully characterized. Monitoring the
1
31
oxidative addition reaction progress by both H and P NMR
spectrocopy reveals the presence of an intermediate that we
have assigned as the alkene complex (11-INT) on the basis of
1
31
19,26
diagnostic signals in the H and P NMR spectra.
This
intermediate forms upon mixing and then completely converts
to complex 11 over 6 h (Figure 5). In addition, we observe
broad peaks for the vinyl C−H signal of 1a and 11-INT; a 2D
ROESY NMR spectrum confirms exchange between free and
Single crystals of complexes 11 and 12 have been analyzed
by X-ray crystallography to establish the solid-state molecular
structures; the ORTEP drawings are shown in Figure 6, along
with key bond lengths and angles in Table 1. Complex 11 is
19
coordinated 1a on the NMR time scale. Calculating Keq for
alkene coordination from the concentrations of 1a, Pd(PCy ) ,
7a
essentially isostructural with Pd(PCy ) (Ph)Cl, with the
3
2
3 2
and 11-INT determined in the first 3 h of the reaction gives a
phosphine ligands situated trans in a slightly distorted square
planar geometry. The cyclohexenyl group is orthogonal to the
square plane (P1−Pd1−C1−C2 torsion angle: 95°), with a
Pd−C distance of slightly less than 2.0 Å. Complex 12 has
value of 27(2) M− at room temperature.
1
Reacting Pd(PCy ) with the acid chloride 1m results in
3
2
rapid oxidative addition to the Pd−acyl complex 12, which has
F
DOI: 10.1021/acs.organomet.8b00468
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 6. ORTEP representations of the solid-state molecular structures of complexes 11 (left) and 12 (right) determined by X-ray crystallography.
Thermal ellipsoids are plotted at 50% probability; phosphine substituents are displayed as wireframe for clarity. Complex 11 has a plane of
symmetry through the Cl1−Pd1−C1 plane, rendering both PCy ligands equivalent, and C5 is disordered over two positions (one shown for
3
clarity). Complex 12 crystallizes with two inequivalent molecules in the asymmetric unit; the image and metrical parameters correspond to one of
these molecules.
Table 1. Bond Lengths (Å), Bond Angles (deg), and
Torsion Angles (deg) for Complexes 11 and 12
requires 1 h at elevated temperature. These rate differences are
almost certainly due to the coordination environments of the
two acyl species. Complex 12 is coordinatively and electroni-
cally saturated, meaning that PCy3 dissociation is likely
required for methanolysis to occur. In contrast, complex 10
has an open coordination site to accommodate the incoming
nucleophile. Of course, the presence of CO will lead to changes
in Pd speciation and the rates of elementary steps in the catalytic
cycle; kinetic studies to probe these aspects are ongoing.
Large-Scale Preparation of 1b. All of the information
gathered during screening, gram-scale preparations, and stoi-
chiometric organometallic studies has informed our develop-
ment of a multigram preparation of 1b. Given the very fast
complex 11
complex 12
Pd1−C1
Pd1−P1
1.988(7)
2.3540(11)
Pd1−C1
Pd1−P1
Pd1−P2
Pd1−Cl1
C1−O1
C1−C2
C2−C7
C6−C7
1.9827(18)
2.3812(5)
2.3581(5)
2.4302(5)
1.212(2)
1.526(3)
1.335(3)
1.474(3)
1.219(3)
1.508(3)
178.02(6)
169.846(18)
94.26(5)
88.03(5)
86.050(16)
92.008(16)
88.6(2)
Pd1−Cl1
2.4147(18)
C1−C2
C2−C3
C3−O1
C3−C4
C1−Pd1−Cl1
P1−Pd1−P1
C1−Pd1−P1
1.380(9)
1.455(10)
1.230(10)
1.525(12)
172.3(2)
169.71(7)
91.31(4)
C6−O2
C5−C6
rates observed for the formation of 1b using P(t-Bu) even at
3
low catalyst loadings, we chose this ligand over P(Me)(t-Bu)2
for further development; large-scale availability was also a
consideration. On the basis of the results shown in Scheme 2B,
we chose toluene as the reaction solvent, limited the MeOH
charge to 1 equiv, and maintained an elevated CO pressure
C1−Pd1−Cl1
P1−Pd1−P2
C1−Pd1−P1
C1−Pd1−P2
Cl1−Pd1−P1
Cl1−Pd1−P2
P1−Pd1−C1−O1
Cl1−Pd1−P1
89.37(4)
95.0(2)
(
eq 1). Another key aspect of the reaction setup is to charge
P1−Pd1−C1−C2
essentially the same structural features, with the acyl ligand ori-
ented orthogonal to the square plane and a similar Pd−C
distance of 1.98 Å.
While further kinetic and spectroscopic studies under the
carbonylation reaction conditions are required to fully eluci-
date the catalytic mechanism, the stoichoimetric chemistry from
Scheme 2 points to key differences in reactivity depending on
ligand identity. As documented for oxidative addition of aryl
halides, Pd/P(t-Bu) and Pd/PCy systems undergo oxidative
CO at room temperature, followed by heating to 80 °C, rather
than the more typical reverse order of operations. All of these
choices were made to minimize or eliminate the formation of
1l and 1n. Furthermore, moving to a single solvent system
rather than the mixed toluene/acetonitrile conditions from
Figure 3 simpifies the process considerably and makes solvent
recycling on a commercial scale more feasible. Finally, we
increased the reaction concentration to >1 M in 1a to
maximize volume efficiency.
3
3
22
24
addition with 1a via dissociative or associative mechanisms,
respectively. Precoordination of the alkene to Pd(PCy ) is
3
2
rapid and favorable (on/off rates on the NMR time scale, with
−
1
Keq = 27 M ), whereas the need for phosphine dissociation
from Pd[P(t-Bu) ] limits the extent of oxidative addition
3
2
under stoichiometric conditions.
On the other hand, methanolysis of the P(t-Bu) -ligated acyl
complex 10 is rapid, reaching complete conversion in minutes
even at room temperature. This is in stark contrast to the
Our initial 60 g scale run with 1.3 mol % of Pd reached
completion in only 20 min as judged by CO uptake, enabling a
further reduction in catalyst loading. Using only 0.5 mol % of
Pd gives 1b in 90% isolated yield (eq 1). Analysis of the crude
product by both NMR and LCMS indicates a very high level of
3
reactivity of PCy -ligated acyl complex 12, where methanolysis
3
G
DOI: 10.1021/acs.organomet.8b00468
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
purity after only a simple aqueous workup, with residual
catalyst as the only significant contaminant. We believe that
carbonylation of β-chloro enones will now be a preferred
method to prepare structurally varied cyclic γ-keto ester and
acid derivatives on scale.
temperature, 45 °C; flow rate, 1.3 mL/min; solvent gradient, ACN
(0.05% TFA v/v)/H O (0.05% TFA v/v), from 97/3 to 2/98 over
2
1
.9 min.
SAFETY NOTE. Pressurized carbonylation reactions should always
be carried out in well-maintained, leak-checked pressure equipment
located in a well-ventilated environment, preferrably a dedicated
pressure laboratory. Carbon monoxide detectors should always be in
use and a buddy system employed to ensure safe operation. Before
any large-scale reaction (>100 g input) is performed, a thorough
safety assessment should be conducted, including reaction calorimetry
studies.
General Procedures for High-Throughput Screening. Plate
Setup. Small-scale HTS reactions were carried out on 0.10 mmol
scale with a total reaction volume of ∼0.3−0.4 mL. Solid bases,
palladium sources, and ligands were automatically dosed using a
Quantos QX96 (Mettler Toledo) instrument housed inside a nitrogen
glovebox. These materials were dispensed into a 48-well array of
1.5 mL crimp-cap HPLC vials containing micro stirbars. The plate
was removed from the glovebox, and all liquid reagents and stock
solutions of substrates/biphenyl (internal standard) in reaction sol-
vent were dispensed via micropipet to the appropriate vials according
to the experimental design on the benchtop.
Reaction. Once all materials were charged, each vial was sealed
with an aluminum crimp-cap containing a PTFE/silicone/PTFE
septum. The septa were pierced with a wide-bore needle to allow gas
ingress/egress. The vials (arrayed in a metal plate) were heated to the
desired reaction temperature with magnetic stirring under the desired
pressure of carbon monoxide (∼60 psig) inside a Cat96 pressure
reactor (HEL). Prior to the commencement of heating/stirring, the
following purge cycles were conducted: 3 × 75 psig nitrogen, 3 × 70
psig carbon monoxide. At the end of the desired reaction time, stirring
was stopped, the reactor was cooled to room temperature, and the
atmosphere was replaced with nitrogen following 3 × 5 bar purge
cycles.
CONCLUSIONS
■
By employing high-throughput screening of multiple substrates
simultaneously, we have developed a general, operationally
simple, and scalable synthesis of cyclic α,β-unsaturated γ-keto
esters and acids via the Pd-catalyzed carbonylation of easily
accessed β-chloro enones. Using an appropriate catalyst sys-
tem, centered on simple trialkylphosphine ligands such as
P(Me)(t-Bu) and P(t-Bu) , enables carbonylative coupling
2
3
between a wide array of functionalized electrophiles and
oxygen-based nucleophiles with exceptional chemoselectivity.
These reactions can be carried out at high reaction concen-
trations on a multigram scale, providing rapid and efficient
8
−12
access to these versatile compounds.
We believe a key
element of the carbonylation reactivity observed is coordina-
tion of the enone CC bond to Pd prior to oxidative addition,
combined with the high electrophilicity of the β-carbon. This
propensity of β-chloro enones to add to Pd(0) will likely find
application in many other cross-coupling reactions under mild
2
7
conditions, especially given that the chloro enone moiety can
be functionalized in the presence of many other potentially
sensitive groups. Evidently, a chloro enone can be more acti-
vated toward oxidative addition in comparison to an aryl
bromidewhen an appropriate ligand is usedenabling
chemoselective functionalization. Further studies on elucidat-
ing the reaction scope and mechanism are currently underway
and will be reported in due course.
Analysis. Once the system returned to ambient pressure, the plate
was removed and the caps were removed. Each vial was diluted with
0
.7 mL of acetonitrile. The plate was placed on a stirplate to ensure
EXPERIMENTAL SECTION
complete dissolution of any organic products. The resulting sus-
pensions were centrifuged prior to withdrawing 20 uL aliquots via
micropipet. These aliquots were diluted into an analysis plate with
wells containing 1 mL of acetonitrile. The analysis plate was then
analyzed by either HPLC or LCMS to assay for desired product and
the remaining starting material. Select reactions were also analyzed by
■
General Information. All common reagents and solvents were
purchased from commercial sources and used as received. Palladium
catalysts and ligands were stored and handled under dry nitrogen in
gloveboxes unless otherwise noted. Pd[P(t-Bu)3]2 and Pd(PCy3)2
were obtained from Strem and stored at −35 °C in a glovebox. All
chloroenone substrates were stored at −20 °C to prevent
decomposition, which appears to occur via hydrolysis back to the
corresponding dione.
1
H NMR to obtain solution yields versus internal standard
19
(biphenyl).
General Procedures for Gram-Scale Carbonylation Reac-
tions. Gram-scale runs were performed using either an HEL
AutoMATE parallel pressure reactor with 50 mL vessels and
magnetically driven stirring with a fixed magnetic impeller or using
an Unchained Laboratories OSR unit with 30 mL vessels and
overhead mechanical stirring. General procedures for the reaction
setup in each reactor type, as well as the general workup and isolation,
are described below.
Analytical Details. All NMR spectra were acquired at ambient
temperature on a Bruker 400 MHz spectrometer. Solvents and
frequencies for specific data acquisitions are noted for each case in the
following sections. Chemical shifts were calibrated relative to residual
1
13
31
19
protio solvent ( H and C) or to external standards ( P and F).
Data were processed using TopSpin and reports generated using ACD
SpecManager.
HPLC analysis was performed on Agilent 1260 or 1290 series
instruments with diode array detectors, though analysis was typically
done with traces from a single wavelength. Two HPLC methods were
utilized during the course of this work. Method A (1260 series):
column, Zorbax SB-C18, 1.8 μm, 3 × 50 mm; column temperature,
HEL AutoMATE. The entirety of this preparation was carried out on
the benchtop. The desired electrophile (1 g) and biphenyl (internal
standard) were dissolved in 5 mL of anhydrous toluene in a
scintillation vial, and the desired nucleophile (2 equiv, except for runs
with h, where 1.2 equiv was used) was dissolved in anhydrous aceto-
nitrile (5 mL) in a second vial. A 50 mL Hastelloy reactor bottom was
6
0 °C; flow rate, 1.5 mL/min; solvent gradient, ACN (0.05% TFA v/v)/
H O (0.05% TFA v/v), from 100/0 to 5/95 over 2.7 min; detection
staged in the AutoMATE deck and charged with solid PdCl
(ACN)
2 2
2
wavelength, 220 nm. Method B (1290 series): column, Waters X-Select
(2 mol %, or 10 mol % with 8a) and P(Me)(t-Bu) -HBF (4 mol %,
2
4
CSH, 2.5 μm, 2.1 × 30 mm; column temperature, 45 °C; flow rate,
or 20 mol % with 8a). In quick succession, the toluene solution of
substrate/standard, triethylamine (1.5 equiv), and the acetonitrile
solution of the nucleophile were added to the catalyst solids via
syringe. Note that this order of addition is critical to reaction success:
adding the nucleophile prior to the triethylamine can result in
substrate and/or catalyst decomposition. Once the solutions were
charged, the vessel was screwed into the reactor top and lowered into
the heat well. Stirring was commenced, and the reactor was pres-
surized with 60 psig of nitrogen and vented three times. The reactor
1
.6 mL/min; solvent gradient, ACN (0.05% TFA v/v)/H O (0.05%
2
TFA v/v), from 97/3 to 5/95 over 1.9 min; detection wavelength,
2
20 nm.
LCMS analysis was performed on a Waters Acquity system
equipped with UV (Waters Acquity PDA, 210−360 nm), ELS
(
Waters Acquity ELSD, 50 °C), and MS (ESI, Waters Acquity SQD,
positive ion mode, scan time 0.1 s) detectors. Chromatography
method: column, Waters CSH (C18), 1.7 μm, 2.1 × 30 mm; column
H
DOI: 10.1021/acs.organomet.8b00468
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
was then pressurized with 60 psig of carbon monoxide, vented three
times, and pressurized back to 60 psig of carbon monoxide. The
reactor was heated to 80 °C for 18 h with stirring at 450 rpm and was
set to maintain a minimum pressure of 70 psig throughout the reac-
tion. After 18 h, stirring was stopped and the vessel cooled to room
temperature. After 1 h, the reactor was vented and the headspace
replaced with 60 psig of nitrogen. Finally, before the reaction mixture
was retrieved, the reactor was vented.
Large-Scale Preparation of 1b. Procedure. A 1 L Buchiglas
Ecoclave pressure reactor equipped with an overhead stirrer and a gas
inlet controlled by a mass-flow controller was flushed with nitrogen.
The nitrogen stream was maintained throughout the reaction setup.
The reactor was rinsed with anhydrous toluene (3×) prior to charging
reagents. 3-Chlorocyclohex-2-enone (1a; 60.0 g, 460 mmol) was
dissolved in anhydrous toluene (250 mL) and added to the reactor.
Stirring was commenced at 700 rpm. Solid PdCl (ACN) (0.600 g,
2
2
Unchained Laboratories OSR. This procedure was used to prepare
compounds 1c,f,g, 3c, 5c, and 6c. The entirety of this preparation was
carried out inside a nitrogen glovebox. A 30 mL OSR glass liner was
charged with solid PdCl (ACN) (2 mol %) and P(Me)(tBu) -HBF
2.31 mmol) was charged through an addition port using a powder
funnel. The funnel was rinsed with a small amount of toluene to
ensure a quantitative transfer. The addition port was fitted with a
rubber septum, and the system was vented through a second port. Tri-
tert-butylphosphine (1.00 g, 4.94 mmol) was dissolved in anhydrous
toluene (50 mL) inside a nitrogen glovebox. This solution was drawn
into a large syringe, which was brought out of the glovebox to the
pressure reactor. The ligand solution was added to the reactor
through the septum. DIPEA (89.0 g, 120 mL, 689 mmol) was added
via syringe through the septum, followed by methanol (15.8 g, 20 mL,
494 mmol). The septum was removed and the reactor sealed and the
headspace purged with 30 psig of nitrogen three times. Stirring was
then stopped, and the headspace was purged with 30 psig of CO three
times. The reactor was then pressurized with 68 psig of CO, and the
uptake monitor on the mass-flow controller was reset to zero. Stirring
was restarted at 1000 rpm, and the reaction mixture was heated to
80 °C over 35 min. Gas uptake commenced once the temperature had
surpassed 55 °C. The initial rate of CO uptake was monitored
extensively for 20 min after the reaction temperature was reached and
2
2
2
4
(
4 mol %). The desired electrophile (1 g) and biphenyl (internal
standard) were dissolved in 5 mL of anhydrous toluene in a scintil-
lation vial and then transferred to the catalyst solids. Triethylamine
(
1.5 equiv) was added, followed by the desired nucleophile (2 equiv)
dissolved in anhydrous acetonitrile (5 mL). Note that this order of
addition is critical to reaction success: adding the nucleophile prior to
the triethylamine can result in substrate and/or catalyst decom-
position. Once the solutions were charged, the liner was placed in the
OSR module and the reactor top was fixed in place. Each step in the
process was predefined in a library design made with Unchained
Laboratories Library Studio software and executed with Automation
Studio software. The stirring rate was set to 500 rpm followed by
purging of the reactors with ∼120 psig of nitrogen and subsequently
1
00 psig of carbon monoxide. The pressure was then set to the
operating pressure of 60 psig, which the system automatically refills
after the pressure level falls under 55 psig. An initial sample was then
taken (corresponding to t = 0). All samples taken were deposited
directly in 1.5 mL HPLC sample vials in a 48-well plate orientation.
The reactor top, which also contains the stirring belt, was set to
19
then periodically until 180 min. The reaction mixture was stirred
under constant CO pressure at 80 °C for 18 h. The next morning, the
total CO uptake was observed to be ∼460 mmol, indicating reaction
completion. The reaction mixture was an orange solution with a
suspended gray/white solid. The reaction mixture was cooled to
20 °C over 35 min, followed by depressurization and purging with
nitrogen to ensure all CO was removed. An aliquot was analyzed by
LCMS, showing >95% area for the desired product and no remaining 1a.
Workup and Purification. The reaction mixture was drained into a
2 L Erleyenmeyer flask via the bottom valve, and the reactor was
rinsed three times with a 1/1 toluene/water mixture (300 mL total
volume) with stirring to ensure a quantitative transfer. The mixed
organic/aqueous phases were transferred to a separatory funnel, and
the layers were separated. The organic phase was dried over MgSO4,
and the solvent was removed after filtration to give an orange oil.
95 °C; this temperature is above the reactor operating temperature to
prevent reflux in the vessel. Once the reactor top was at temperature,
each reactor was brought to temperature (80 °C) with a 3 min delay
to account for sampling time (each sampling cycle is 3 min). Sampling
then occurred every 15 min for the first 1 h, every 1 h for hours 1−8,
and every 2 h for hours 8−18 (17 total samples from each reactor)
followed by dilution with acetonitrile. After the final time point was
sampled at 18 h, the reactors were vented and cooled to room
temperature. After the reactors cooled, the reactor top was removed
and glass liners removed for further isolation and characterization.
Samples taken were moved directly to HPLC for analysis.
1
Workup and Isolation. The crude reaction mixture was diluted
with TBME (10 mL) or ethyl acetate (10 mL, for carboxylic acid
products), brine (8 mL), and 1 M HCl (8 mL) to completely dissolve
A H NMR spectrum of the crude product indicated that all DIPEA
was removed in the workup. The only other discernible components
were a small amount of P(O)(t-Bu) and residual toluene. In order to
3
all solids (HNEt Cl), followed by transfer to a separatory funnel. The
remove the residual palladium, the material was purified by passage
through a silica plug as follows. A 3 L medium-porosity glass frit was
loaded with 750 g of silica gel (approximate dimensions of the plug:
12 in. diameter, 3 in. depth). The silica was slurried with heptane with
stirring using a glass rod to remove air bubbles and inhomogenaities.
The heptane was allowed to drain via gravity. Once the solvent level
reached the top of the silica bed, the neat orange oil was loaded onto
the plug, using heptane to ensure a quantitative transfer. Once the com-
pound was loaded, a 2 in. depth bed of sand was added to the top of the
plug, and the desired compound was eluted with a heptane/ethyl acetate
mixture (3 L of 10/1 heptane/ethyl acetate, 2 L of 7.5/1 heptane/
ethyl acetate, 2 L of 5/1 heptane/ethyl acetate). The eluent solutions
were concentrated in vacuo to give pure 1b (63.8 g, 90% yield).
Synthesis of 11. In a nitrogen-filled glovebox, a tared 4 mL screw-
cap vial was charged with Pd(PCy ) (120 mg, 0.180 mmol) and a
3
phases were well mixed, and then the aqueous layer was removed. The
aqueous layer was back-extracted with TBME (10 mL) or ethyl
acetate (10 mL, for carboxylic acid products), and the combined
organic phases were dried over MgSO . After filtration, the solvent
4
1
was removed in vacuo, and a H NMR spectrum was obtained of the
crude material to determine an NMR yield. For all ester products
(
except 3k), the residue was redissolved in 10−20 mL of hexanes
(using a minimum amount of EtOAc if required) and purified by
column chromatography with either a 0−20% or 0−40% ethyl acetate
in hexanes gradient, depending on the R value for the desired product
f
(
determined by TLC). All preparative column chromatography was
performed using a Teledyne ISCO Combiflash R system with
f
prepacked RediSep columns.
Carboxylic acid products were not subject to chromatography but
were purified in the following manner. The crude residue was
dissolved in a mixture of TBME (20 mL) and 1 M NaOH (20 mL)
with vigorous mixing. After mixing, the layers were allowed to settle
and then separated. The aqueous layer was back-extracted with
TBME (10 mL). The basic aqueous layer was then acidified with 3 M
HCl to pH <2 before extraction with ethyl acetate (3 × 10 mL). The
3
2
stirbar. A second 4 mL vial was charged with 1a (25.8 mg,
0.198 mmol) and toluene (3 mL). The solution of 1a was transferred
to the vial containing Pd(PCy ) , and the mixture was stirred until it
3
2
was homogeneous. The stirbar was then removed, and the mixture
was left to stand at room temperature over several days to crystallize
the product. The mother liquor was then decanted, and the white
solid was washed with pentane (2 × 1 mL) and dried in vacuo.
The solid was again triturated with pentane to remove residual toluene
solvent and dried extensively in vacuo to give compound 11 (110 mg,
combined ethyl acetate layers were dried over MgSO , filtered, and
4
concentrated in vacuo to give the carboxylic acid products as white/
off-white solids.
Compound 3k was purified by selective extractions.
19
1
77% yield). H NMR (400 MHz, CD Cl ): δ (ppm) 1.10−1.39
2
2
I
DOI: 10.1021/acs.organomet.8b00468
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(
(
1
5
m, 18 H, 18 × Cy-H), 1.51−1.77 (m, 18 H, 18 × Cy-H), 1.77−1.90
ORCID
m, 12 H, 12 × Cy-H), 1.90−2.16 (m, 14 H, −CH CH CCH−,
2
2
Anna L. Dunn: 0000-0003-2852-7755
Graham E. Dobereiner: 0000-0001-6885-2021
David C. Leitch: 0000-0002-8726-3318
2 × Cy-H), 2.32 (m, 8 H, 6 × PCHR , −CH CCH−), 2.90 (t, J =
2
2
1
3
1
.5 Hz, 2 H, −CH CO), 6.40 (s, 1 H, −CCH−). C{ H} NMR
2
(101 MHz, CD Cl ): δ (ppm) 24.53, 26.53, 27.76 (d, J = 9.80 Hz),
2
2
7.81 (d, J = 9.50 Hz), 27.86 (d, J = 9.80 Hz), 30.17 (d, J =
Notes
4.50 Hz), 33.82 (d, J = 9.40 Hz), 33.91 (d, J = 9.40 Hz), 37.95,
The authors declare no competing financial interest.
3
1
1
9.53, 136.84 (t, J = 2.80 Hz), 191.25, 199.15 (br). P{ H} NMR
2 2
ACKNOWLEDGMENTS
■
The authors thank Dr. Ian Andrews, Dr. Alexander O’Brien,
Dr. Praew Thansandote, and Dr. Antonella Carangio for
insightful discussions and Dr. Bernie O’Hare for assistance
with NMR experiments. Analytical data were obtained from
the CENTC Elemental Analysis Facility at the University of
Rochester, funded by NSF CHE-0650456. This material is
based upon work supported by the National Science Foun-
dation under Grant No. CHE-1565721.
+
3
2
3
+
3
+
3
4
2
73
2
Synthesis of 12. In a nitrogen-filled glovebox, a tared 4 mL screw-
cap vial was charged with Pd(PCy ) (100 mg, 0.150 mmol) and a
3
2
REFERENCES
stirbar. A second 4 mL vial was charged with 1l (25.0 mg, 0.158 mmol)
and heptane (1 mL). The solution of 1l was transferred to the vial
containing Pd(PCy ) , using two additional portions of heptane
■
(
1) (a) Catalytic Carbonylation Reactions; Beller, M., Ed.; Springer-
Verlag: Berlin, 2006. (b) Beller, M.; Wu, X.-F. Transition Metal
Catalyzed Carbonylation Reactions; Springer-Verlag: Berlin, 2013.
3
2
(0.5 mL each), to effect a quantiative transfer. The mixture changed
(
2) Appel, A. M.; Bercaw, J. E.; Bocarsly, A. B.; Dobbek, H.; DuBois,
color to brown/orange immediately upon addition of the acid chloride.
The suspension was stirred and became homogeneous after
approximately 10 min. At this point, an orange solid began to
precipitate; stirring was continued for a further 2 h, after which the vial
was placed in a −35 °C freezer for several days. After standing, the
mother liquor was decanted and the orange solid washed with pentane
D. L.; Dupuis, M.; Ferry, J. G.; Fujita, E.; Hille, R.; Kenis, P. J. A.;
Kerfeld, C. A.; Morris, R. H.; Peden, C. H. F.; Portis, A. R.; Ragsdale,
S. W.; Rauchfuss, T. B.; Reek, J. N. H.; Seefeldt, L. C.; Thauer, R. K.;
Waldropa, G. L. Frontiers, Opportunities, and Challenges in
Biochemical and Chemical Catalysis of CO
2013, 113, 6621−6658.
̈
(3) (a) Franke, R.; Selent, D.; Borner, A. Applied Hydroformylation.
Fixation. Chem. Rev.
2
(
2 × 1 mL) and dried in vacuo. The solid was again triturated with pen-
tane to remove residual toluene solvent and dried extensively in vacuo
to give compound 12 (79.3 mg, 64% yield). H NMR (400 MHz,
1
Chem. Rev. 2012, 112, 5675−5732. (b) de Vries, J. G. Hydro-
formylation of Alkenes: Industrial Applications. In Science of Synthesis:
C-1 Building Blocks in Organic Synthesis; van Leeuwen, P. W. N. M.,
Ed.; Georg Thieme Verlag: Stuttgart, 2014; Vol. 1, pp 193−228.
(4) (a) Howard, M. J.; Jones, M. D.; Roberts, M. S.; Taylor, S. A. C1
to acetyls: catalysis and process. Catal. Today 1993, 18, 325−354.
(b) Sunley, G. J.; Watson, D. J. High productivity methanol
carbonylation catalysis using iridium: The Cativa process for the
manufacture of acetic acid. Catal. Today 2000, 58, 293−307.
(c) Zoeller, J. R. 2015 Paul N. Rylander Award Address: Selected
Advances in Catalysis and Reactor Systems for the Generation of
Acetates and Propionates Using Carbonylation. Org. Process Res. Dev.
2016, 20, 1016−1025.
d -toluene): δ (ppm) 1.04−1.34 (m, 18 H, 18 × Cy-H), 1.53−1.83
8
(
m, 32 H, −CH CH CCH−, 30 × Cy-H), 1.90−2.04 (m, 6 H, 6 ×
2
2
PCHR ), 2.13−2.36 (m, 14 H, −CH CCH−, 12 × Cy-H), 2.45 (br t,
2
2
1
3
1
J = 5.30 Hz, 2 H, −CH CO), 7.99 (s, 1 H, −CCH−). C{ H}
2
NMR (101 MHz, d -toluene): δ (ppm) 27.60, 30.05, 31.60, 32.60−
8
3
3.05 (m, 3 C overlapping), 35.18 (d, J = 22.6 Hz), 39.60 (t, J =
8
.90 Hz), 43.60, 146.74 (br), 165.17 (t, J = 11.60 Hz), 204.12, 240.53.
3
1
1
P{ H} NMR (162 MHz, d -toluene) δ (ppm) 21.44. LRMS (ESI):
8
obtained using LCMS under standard conditions; fragment observed
between m/z 787 and 795 in diagnostic isotope pattern, with most
+
intense peak at m/z 789.8, corresponding to [Pd(PCy ) (COR)] ; most
3
2
+
intense fragment peak observed at m/z 281 for [HPCy ] . These
3
isotope patterns are pictured in the Supporting Information. Anal. Calcd
for C H ClO P Pd: C, 62.54; H, 8.91. Found: C, 62.80; H, 9.19.
(5) (a) Barnard, C. F. J. Palladium-Catalyzed CarbonylationA
Reaction Come of Age. Organometallics 2008, 27, 5402−5422.
43
73
2 2
(
b) Grigg, R.; Mutton, S. P. Pd-catalysed carbonylations: versatile
technology for discovery and process chemists. Tetrahedron 2010, 66,
515−5548. (c) Magano, J.; Dunetz, J. R. Large-Scale Applications of
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the ACS
Publications website at DOI: 10.1021/acs.organomet.8b00468.
■
5
*
S
Transition Metal-Catalyzed Couplings for the Synthesis of Pharma-
ceuticals. Chem. Rev. 2011, 111, 2177−2250. (d) Wu, L.; Fang, X.;
Liu, Q.; Jackstell, R.; Beller, M.; Wu, X.-F. Palladium-Catalyzed
Carbonylative Transformation of C(sp3)−X Bonds. ACS Catal. 2014,
Detailed experimental procedures, full tables of screen-
ing data, characterization data, reaction progress plots,
and crystallographic details (PDF)
4
, 2977−2989. (e) Neumann, H.; Schranck, J. Carbonylation of Aryl
and Vinyl Halides. In Science of Synthesis: C-1 Building Blocks in
Organic Synthesis; van Leeuwen, P. W. N. M., Ed.; Georg Thieme
Verlag: Stuttgart, 2014; Vol. 2, pp 67−93. (f) Bai, Y.; Davis, D. C.;
Dai, M. Natural Product Synthesis via Palladium-Catalyzed Carbon-
ylation. J. Org. Chem. 2017, 82, 2319−2328.
Accession Codes
CCDC 1852940−1852941 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
(6) Labinger, J. Tutorial on Oxidative Addition. Organometallics
2015, 34, 4784−4795.
(7) (a) Huser, M.; Youinou, M.-T.; Osborn, J. A. Chlorocarbon
Activation: Catalytic Carbonylation of Dichloromethane and
Chlorobenzene. Angew. Chem., Int. Ed. Engl. 1989, 28, 1386−1388.
(b) Ben-David, Y.; Portnoy, M.; Milstein, D. Chelate-assisted,
palladium-catalyzed efficient carbonylation of aryl chlorides. J. Am.
Chem. Soc. 1989, 111, 8742−8744. (c) Ben-David, Y.; Portnoy, M.;
Milstein, D. Formylation of aryl chlorides catalysed by a palladium
complex. J. Chem. Soc., Chem. Commun. 1989, 1816−1817.
AUTHOR INFORMATION
Corresponding Author
E-mail for D.C.L.: david.c.leitch@gsk.com; dcleitch@uvic.ca.
■
*
J
DOI: 10.1021/acs.organomet.8b00468
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(
d) Grushin, V. V.; Alper, H. An exceptionally simple biphasic
(c) Webster, F. X.; Silverstein, R. M. Synthesis of Optically Pure
Enantiomers of Grandisol. J. Org. Chem. 1986, 51, 5226−5231.
(d) Herzog, H.; Koch, H.; Scharf, H. D.; Runsink, J. Chiral induction
in photochemical reactions. V. Regio- and diastereoselectivity in the
photochemical [2 + 2] cycloaddition of chiral cyclenone-3-
carboxylates with 1,1′-diethoxyethene. Tetrahedron 1986, 42, 3547−
3558. (e) Herzog, H.; Koch, H.; Scharf, H. D.; Runsink, J. Chiral
induction in photochemical reactions. VII. Double stereodifferentia-
tion in the [2 + 2] photocycloaddition. Chem. Ber. 1987, 120, 1737−
1740. (f) Audenaert, F.; De Keukeleire, D.; Vandewalle, M. A new
and facile perhydroazulene formation: the total synthesis of the
carotane (+)-daucene. Tetrahedron 1987, 43, 5593−5604. (g) Lange,
G. L.; Decicco, C. Asymmetric induction in mixed photoadditions
employing α,β-unsaturated homochiral ketals. Tetrahedron Lett. 1988,
method for the metal catalysed carbonylation of chloroarenes. J.
Chem. Soc., Chem. Commun. 1992, 611−612. (e) Portnoy, M.;
Milstein, D. Chelate effect on the structure and reactivity of electron-
rich palladium complexes and its relevance to catalysis. Organo-
̈
metallics 1993, 12, 1655−1664. (f) Magerlein, W.; Indolese, A. F.;
Beller, M. A More Efficient Catalyst for the Carbonylation of
Chloroarenes. Angew. Chem., Int. Ed. 2001, 40, 2856−2859. (g) Beller,
M.; Magerlein, W.; Indolese, A. F.; Fischer, C. Efficient Palladium-
̈
Catalyzed Alkoxycarbonylation of N-Heteroaryl Chlorides - A
Practical Synthesis of Building Blocks for Pharmaceuticals and
Herbicides. Synthesis 2001, 2001, 1098−1109. (h) Magerlein, W.;
̈
Indolese, A. F.; Beller, M. Development of new palladium catalysts for
the alkoxycarbonylation of aryl chlorides. J. Organomet. Chem. 2002,
6
41, 30−40. (i) Jimenez-Rodriguez, C.; Eastham, G. R.; Cole-
29, 2613−2614. (h) Lange, G. L.; Decicco, C. P.; Willson, J.;
Hamilton, D. J. The methoxycarbonylation of aryl chlorides catalysed
by palladium complexes of bis(di-tert-butylphosphinomethyl)-
benzene. Dalton Trans. 2005, 1826−1830. (j) Cai, C.; Rivera, N.
R.; Balsells, J.; Sidler, R. R.; McWilliams, J. C.; Shultz, C. S.; Sun, Y.
An Efficient Catalyst for Pd-Catalyzed Carbonylation of Aryl
Arenesulfonates. Org. Lett. 2006, 8, 5161−5164. (k) Watson, D. A.;
Fan, X.; Buchwald, S. L. Carbonylation of Aryl Chlorides with Oxygen
Nucleophiles at Atmospheric Pressure. Preparation of Phenyl Esters
as Acyl Transfer Agents and the Direct Preparation of Alkyl Esters
and Carboxylic Acids. J. Org. Chem. 2008, 73, 7096−7101.
Strickland, L. A. Ring expansions of [2 + 2] photoadducts. Potential
applications in the synthesis of triquinane and taxane skeletons. J. Org.
Chem. 1989, 54, 1805−1810. (i) Lange, G. L.; Organ, M. G.; Lee, M.
Reversal of regioselectivity with increasing ring size of alkene
component in [2 + 2] photoadditions. Tetrahedron Lett. 1990, 31,
4
689−4692. (j) Lange, G. L.; Gottardo, C. Free radical fragmentation
of derivatives of [2 + 2] photoadducts. Tetrahedron Lett. 1990, 31,
985−5988. (k) Lange, G. L.; Gottardo, C. Free Radical
5
Fragmentation of [2 + 2] Photoadduct Derivatives. Formal Synthesis
of Pentalenene. J. Org. Chem. 1995, 60, 2183−2187. (l) Omar, H. I.;
Odo, Y.; Shigemitsu, Y.; Shimo, T.; Somekawa, K. Transition state
analysis on regioselectivity in [2 + 2] photocycloaddition reactions of
substituted 2-cyclohexenone with cycloalkenecarboxylates. Tetrahe-
dron 2003, 59, 8099−8105. (m) Zhao, G.; Yang, C.; Sun, H.; Lin, R.;
Xia, W. (+)-Camphor Derivative Induced Asymmetric [2 + 2]
Photoaddition Reaction. Org. Lett. 2012, 14, 776−779.
(l) Munday, R. H.; Martinelli, J. R.; Buchwald, S. L. Palladium-
Catalyzed Carbonylation of Aryl Tosylates and Mesylates. J. Am.
Chem. Soc. 2008, 130, 2754−2755. (m) Burhardt, M. N.; Taaning, R.;
Nielsen, N. C.; Skrydstrup, T. Isotope-Labeling of the Fibril Binding
Compound FSB via a Pd-Catalyzed Double Alkoxycarbonylation. J.
Org. Chem. 2012, 77, 5357−5363. (n) Chung, S.; Sach, N.; Choi, C.;
Yang, X.; Drozda, S. E.; Singer, R. A.; Wright, S. W. Amino-
carbonylation of Aryl Tosylates to Carboxamides. Org. Lett. 2015, 17,
(11) (a) Brown, M. K.; May, T. L.; Baxter, C. A.; Hoveyda, A. H.
All-Carbon Quaternary Stereogenic Centers by Enantioselective Cu-
Catalyzed Conjugate Additions Promoted by a Chiral N-Heterocyclic
Carbene. Angew. Chem., Int. Ed. 2007, 46, 1097−1100. (b) Shintani,
R.; Tsutsumi, Y.; Nagaosa, M.; Nishimura, T.; Hayashi, T. Sodium
Tetraarylborates as Effective Nucleophiles in Rhodium/Diene-
Catalyzed 1,4-Addition to β,β-Disubstituted α,β-Unsaturated Ke-
tones: Catalytic Asymmetric Construction of Quaternary Carbon
Stereocenters. J. Am. Chem. Soc. 2009, 131, 13588−13589.
2
848−2851. (o) Wu, X.-F. Palladium-catalyzed carbonylative trans-
formation of aryl chlorides and aryl tosylates. RSC Adv. 2016, 6,
3831−83837. (p) Wang, J. Y.; Strom, A. E.; Hartwig, J. F.
8
Mechanistic Studies of Palladium-Catalyzed Aminocarbonylation of
Aryl Chlorides with Carbon Monoxide and Ammonia. J. Am. Chem.
Soc. 2018, 140, 7979−7993.
(8) (a) Bougioukou, D. J.; Kille, S.; Taglieber, A.; Reetz, M. T.
Directed Evolution of an Enantioselective Enoate-Reductase: Testing
the Utility of Iterative Saturation Mutagenesis. Adv. Synth. Catal.
(12) Izawa, Y.; Zheng, C.; Stahl, S. Aerobic Oxidative Heck/
Dehydrogenation Reactions of Cyclohexenones: Efficient Access to
2009, 351, 3287−3305. (b) Turrini, N. G.; Cioc, R. C.; van der Niet,
meta-Substituted Phenols. Angew. Chem., Int. Ed. 2013, 52, 3672−
D. J. H.; Ruijter, E.; Orru, R. V. A.; Hall, M.; Faber, K. Biocatalytic
access to nonracemic γ-oxo esters via stereoselective reduction using
ene-reductases. Green Chem. 2017, 19, 511−518. (c) Hadi, T.; Diaz-
Rodriguez, A.; Khan, D.; Morrison, J. P.; Kaplan, J. M.; Gallagher, K.
T.; Schober, M.; Webb, M. R.; Brown, K.; Fuerst, D.; Snajdrova, R.;
Roiban, G.-D. Identification and Implementation of Biocatalytic
Transformations in Route Discovery: Synthesis of Chiral 1,3-
Substituted Cyclohexanone Building Blocks. Org. Process Res. Dev.
3
(
675.
13) (a) Lange, G. L.; Otulakowski, J. A. Improved preparation of
methyl 3-oxo-1-cyclohexene-1-carboxylate and its use in the synthesis
of substituted 1,5-cyclodecadienes. J. Org. Chem. 1982, 47, 5093−
5096. (b) For a Rh-catalyzed allylic oxidation, see: Catino, A. J.;
Forslund, R. E.; Doyle, M. P. Dirhodium(II) Caprolactamate: An
Exceptional Catalyst for Allylic Oxidation. J. Am. Chem. Soc. 2004,
1
(
26, 13622−13623.
2
(
018, 22, 871.
14) Webster, F. X.; Silverstein, R. M. Control of the Birch
9) (a) Zhang, H.-J.; Hu, L.; Ma, Z.; Li, R.; Zhang, Z.; Tao, C.;
Reduction of m-Anisic Acid to Produce Specific 3-Oxocyclohex-
Cheng, B.; Li, Y.; Wang, H.; Zhai, H. Total Synthesis of the
Diterpenoid (+)-Harringtonolide. Angew. Chem., Int. Ed. 2016, 55,
enecarboxylic Acids. Synthesis 1987, 1987, 922−924.
(15) The only previously reported examples of β-halo enone
11638−11641. (b) Liu, Y.; Cheng, L.-J.; Yue, H.-T.; Che, W.; Xie, J.-
carbonylation utilize either iodo or bromo leaving groups:
a) Baillargeon, V. P.; Stille, J. K. Direct conversion of organic
H.; Zhou, Q.-L. Divergent enantioselective synthesis of hapalindole-
type alkaloids using catalytic asymmetric hydrogenation of a ketone to
construct the chiral core structure. Chem. Sci. 2016, 7, 4725−4729.
(
halides to aldehydes with carbon monoxide and tin hydride catalyzed
by palladium. J. Am. Chem. Soc. 1983, 105, 7175−7176.
(b) Baillargeon, V. P.; Stille, J. K. Palladium-catalyzed formylation
of organic halides with carbon monoxide and tin hydride. J. Am.
Chem. Soc. 1986, 108, 452−461. (c) Kuethe, J. T.; Wong, A.; Wu, J.;
Davies, I. W.; Dormer, P. G.; Welch, C. J.; Hillier, M. C.; Hughes, D.
L.; Reider, P. J. Asymmetric Synthesis of 1,2,3-Trisubstituted
Cyclopentanes and Cyclohexanes as Key Components of Substance
P Antagonists. J. Org. Chem. 2002, 67, 5993−6000. (d) Ebner, D. C.;
Bagdanoff, J. T.; Ferreira, E. M.; McFadden, R. M.; Caspi, D. D.;
Trend, R. M.; Stoltz, B. M. The Palladium-Catalyzed Aerobic Kinetic
(c) Nicolaou, K. C.; Liu, G.; Beabout, K.; McCurry, M. D.; Shamoo,
Y. Asymmetric Alkylation of Anthrones, Enantioselective Total
Synthesis of (−)- and (+)-Viridicatumtoxins B and Analogues
Thereof: Absolute Configuration and Potent Antibacterial Agents. J.
Am. Chem. Soc. 2017, 139, 3736−3746.
(10) (a) Agosta, W. C.; Lowrance, W. W., Jr. Synthesis and
Photocycloadditions of Compounds Related to 3-Carboxycyclohex-
enone. J. Org. Chem. 1970, 35, 3851−3856. (b) Lange, G. L.; Decicco,
C.; Tan, S. L.; Chamberlain, G. Asymmetric Induction in Simple [2 +
2
] Photoadditions. Tetrahedron Lett. 1985, 26, 4707−4710.
K
DOI: 10.1021/acs.organomet.8b00468
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Resolution of Secondary Alcohols: Reaction Development, Scope, and
Observations of Mechanistic Complexity with Iodobenzene as
Reactant. Chem. - Eur. J. 2014, 20, 1116−1125.
Applications. Chem. - Eur. J. 2009, 15, 12978−12992.
(16) Buitrago Santanilla, A.; Regalado, E. L.; Pereira, T.; Shevlin, M.;
(25) (a) Kirchhoff, J. H.; Netherton, M. R.; Hills, I. D.; Fu, G. C.
Boronic Acids: New Coupling Partners in Room-Temperature Suzuki
Reactions of Alkyl Bromides. Crystallographic Characterization of an
Oxidative-Addition Adduct Generated under Remarkably Mild
Conditions. J. Am. Chem. Soc. 2002, 124, 13662−13663. (b) Mitchell,
E. A.; Baird, M. C. Optimization of Procedures for the Syntheses of
Bisphosphinepalladium(0) Precursors for Suzuki-Miyaura and Similar
Cross-Coupling Catalysis: Identification of 3:1 Coordination
Bateman, K.; Campeau, L.-C.; Schneeweis, J.; Berrit, S.; Shi, Z.-C.;
Nantermet, P.; Liu, Y.; Helmy, R.; Welch, C. J.; Vachal, P.; Davies, I.
W.; Cernak, T.; Dreher, S. D. Nanomole-scale high-throughput
chemistry for the synthesis of complex molecules. Science 2015, 347,
4
(
9−53.
17) (a) Klaus, S.; Neumann, H.; Zapf, A.; Strubing, D.; Hubner, S.;
̈
̈
Almena, J.; Riermeier, T.; Gross, P.; Sarich, M.; Krahnert, W.-R.;
Rossen, K.; Beller, M. A General and Efficient Method for the
Formylation of Aryl and Heteroaryl Bromides. Angew. Chem., Int. Ed.
Compounds in Catalyst Mixtures Containing Pd(0), PCy , and/or
3
t
PMeBu . Organometallics 2007, 26, 5230−5238.
2
2
006, 45, 154−158. (b) Barnard, C. F. J. Carbonylation of Aryl
Halides: Extending the Scope of the Reaction. Org. Process Res. Dev.
008, 12, 566−574. (c) Martinelli, J. R.; Watson, D. A.; Freckmann,
(26) Majchrzak, M.; Kostera, S.; Kubicki, M.; Kownacki, I. Synthesis
of new styrylarenes via Suzuki−Miyaura coupling catalysed by highly
active, well-defined palladium catalysts. Dalton Trans. 2013, 42,
15535−15539.
2
D. M. M.; Barder, T. E.; Buchwald, S. L. Palladium-Catalyzed
Carbonylation Reactions of Aryl Bromides at Atmospheric Pressure:
A General System Based on Xantphos. J. Org. Chem. 2008, 73, 7102−
(27) Wang, Z.; Orellana, A. Convenient Access to meta-Substituted
Phenols by Palladium-Catalyzed Suzuki−Miyaura Cross-Coupling
and Oxidation. Chem. - Eur. J. 2017, 23, 11445−11449.
7107. (d) Arthuis, M.; Lecup, A.; Roulland, E. Pd0-Catalyzed
carbonylation of 1,1-dichloro-1-alkenes, a new selective access to Z-α-
chloroacrylates. Chem. Commun. 2010, 46, 7810−7812. (e) Quesnel,
J. S.; Arndtsen, B. A. A Palladium-Catalyzed Carbonylation Approach
to Acid Chloride Synthesis. J. Am. Chem. Soc. 2013, 135, 16841−
1
(
6844.
18) Netherton, N. R.; Fu, G. C. Air-Stable Trialkylphosphonium
Salts: Simple, Practical, and Versatile Replacements for Air-Sensitive
Trialkylphosphines. Applications in Stoichiometric and Catalytic
Processes. Org. Lett. 2001, 3, 4295−4298.
(
(
19) See the Supporting Information for more details.
20) For examples of chemoselective Pd-catalyzed functionalizations
of R−Cl over R−Br groups in noncarbonylative cross couplings, see:
(a) Almond-Thynne, J.; Blakemore, D. C.; Pryde, D. C.; Spivey, A. C.
Site-selective Suzuki−Miyaura coupling of heteroaryl halides −
understanding the trends for pharmaceutically important classes.
Chem. Sci. 2017, 8, 40−62. (b) Keylor, M. H.; Niemeyer, Z. L.;
Sigman, M. S.; Tan, K. L. Inverting Conventional Chemoselectivity in
Pd-Catalyzed Amine Arylations with Multiply Halogenated Pyridines.
J. Am. Chem. Soc. 2017, 139, 10613−10616.
(21) Nunn, C.; DiPietro, A.; Hodnett, N.; Sun, P.; Wells, K. M.
High-Throughput Automated Design of Experiment (DoE) and
Kinetic Modeling to Aid in Process Development of an API. Org.
Process Res. Dev. 2018, 22, 54−61.
(22) (a) Littke, A. F.; Dai, C.; Fu, G. C. Versatile Catalysts for the
Suzuki Cross-Coupling of Arylboronic Acids with Aryl and Vinyl
Halides and Triflates under Mild Conditions. J. Am. Chem. Soc. 2000,
1
22, 4020−4028. (b) Christmann, U.; Vilar, R. Monoligated
palladium species as catalysts in cross-coupling reactions. Angew.
Chem., Int. Ed. 2005, 44, 366−374.
(
23) Meng, G.; Szostak, M. General Olefin Synthesis by the
Palladium-Catalyzed Heck Reaction of Amides: Sterically Controlled
Chemoselective N-C Activation. Angew. Chem., Int. Ed. 2015, 54,
1
(
4518−14522 and references therein.
24) (a) Galardon, E.; Ramdeehul, S.; Brown, J. M.; Cowley, A.; Hii,
K. K. M.; Jutand, A. Profound Steric Control of Reactivity in Aryl
Halide Addition to Bisphosphane Palladium(0) Complexes. Angew.
Chem., Int. Ed. 2002, 41, 1760−1763. (b) Barrios-Landeros, F.;
Hartwig, J. F. Distinct Mechanisms for the Oxidative Addition of
Chloro-, Bromo-, and Iodoarenes to a Bisphosphine Palladium(0)
Complex with Hindered Ligands. J. Am. Chem. Soc. 2005, 127, 6944−
6945. (c) Barrios-Landeros, F.; Carrow, B. P.; Hartwig, J. F. Effect of
Ligand Steric Properties and Halide Identity on the Mechanism for
Oxidative Addition of Haloarenes to Trialkylphosphine Pd(0)
Complexes. J. Am. Chem. Soc. 2009, 131, 8141−8154. (d) Mitchell,
E. A.; Jessop, P. G.; Baird, M. C. A Kinetics Study of the Oxidative
Addition of Bromobenzene to Pd(PCy ) (Cy = cyclohexyl) in a
3
2
Nonpolar Medium: The Influence on Rates of Added PCy and
3
Bromide Ion. Organometallics 2009, 28, 6732−6738. (e) Kurbangalie-
va, A.; Carmichael, D.; Hii, K. K. M.; Jutand, A.; Brown, J. M.
Oxidative Addition to Palladium(0) Diphosphine Complexes:
L
DOI: 10.1021/acs.organomet.8b00468
Organometallics XXXX, XXX, XXX−XXX