Heterobimetallic H2 addition and alkene/alkane elimination reactions related to the mechanism of E-selective alkyne semihydrogenation

Article abstract of DOI:10.1021/acs.organomet.6b00356

Mechanistic aspects of an E-selective alkyne semihydrogenation catalyst are studied through computational modeling and experimental model reactions. We previously communicated the semihydrogenation of diarylalkynes to produce trans-alkenes using the heterobimetallic catalyst (IMes)- AgRuCp(CO)₂. In this report, we disclose further details on the catalyst decomposition products under catalytic conditions, the mechanism of bimetallic H₂ activation, and the factors affecting selectivity for E-alkene generation. Under hydrogenation conditions, the catalyst decomposition product HRuCp(CO)(IMes) was isolated and characterized. Resubmitting this species to the catalytic conditions did not provide useful hydrogenation catalysis, confirming the presence of a bimetallic mechanism under optimal catalytic conditions. The detailed nature of heterobimetallic H₂ activation was probed by calculating internuclear bond orders, atom/fragment charges, and NBO occupancies as functions of reaction coordinate for a model reaction between (IMe)CuRp and H₂. The collected results indicate a late transition state involving deprotonation of a Cu(H₂) σ-complex by the proximal Rp fragment. Late stages of the reaction profile feature H?H dihydrogen bonding between (IMe)CuH and HRuCp(CO)₂, indicative of heterolytic H₂ activation. NBO analysis indicates that the key orbital interactions involved in H₂ activation are (a) donation from the filled H₂ σ-orbital into a Cu 4p acceptor orbital and (b) back-donation from a filled Cu-Ru bonding orbital of predominantly Ru 4d character into the empty H₂ σ?-orbital. Experimental support for the previously proposed cascade alkyne →Z-alkene → E-alkene process was provided by stoichiometric reactions between HRuCp(CO)₂ and isolable (IPr)CuR models of catalytic (IMes)AgR intermediates (R = alkenyl, alkyl). The collected experimental results indicate that selectivity for E-alkene generation is dictated by the relative rates of monometallic β-hydride elimination and bimetallic alkane elimination, which are impacted by several structural features of the catalyst. The mechanistic detail provided by these studies will inform the development of second-generation hydrogenation catalysts.

Full text of DOI:10.1021/acs.organomet.6b00356

Article
pubs.acs.org/Organometallics
Heterobimetallic H Addition and Alkene/Alkane Elimination
2
Reactions Related to the Mechanism of E‑Selective Alkyne
Semihydrogenation
Malkanthi K. Karunananda and Neal P. Mankad*
Department of Chemistry, University of Illinois at Chicago, 845 W. Taylor Street, Chicago, Illinois 60607, United States
*
S Supporting Information
ABSTRACT: Mechanistic aspects of an E-selective alkyne
semihydrogenation catalyst are studied through computational
modeling and experimental model reactions. We previously
communicated the semihydrogenation of diarylalkynes to
produce trans-alkenes using the heterobimetallic catalyst (IMes)-
AgRuCp(CO) . In this report, we disclose further details on the
2
catalyst decomposition products under catalytic conditions, the
mechanism of bimetallic H activation, and the factors affecting selectivity for E-alkene generation. Under hydrogenation
2
conditions, the catalyst decomposition product HRuCp(CO)(IMes) was isolated and characterized. Resubmitting this species to
the catalytic conditions did not provide useful hydrogenation catalysis, confirming the presence of a bimetallic mechanism under
optimal catalytic conditions. The detailed nature of heterobimetallic H activation was probed by calculating internuclear bond
2
orders, atom/fragment charges, and NBO occupancies as functions of reaction coordinate for a model reaction between
(
IMe)CuRp and H . The collected results indicate a late transition state involving deprotonation of a Cu(H ) σ-complex by the
2
2
proximal Rp fragment. Late stages of the reaction profile feature H···H dihydrogen bonding between (IMe)CuH and
HRuCp(CO) , indicative of heterolytic H activation. NBO analysis indicates that the key orbital interactions involved in H
2
2
2
activation are (a) donation from the filled H σ-orbital into a Cu 4p acceptor orbital and (b) back-donation from a filled Cu−Ru
2
bonding orbital of predominantly Ru 4d character into the empty H σ*-orbital. Experimental support for the previously
2
proposed cascade alkyne → Z-alkene → E-alkene process was provided by stoichiometric reactions between HRuCp(CO) and
2
isolable (IPr)CuR models of catalytic (IMes)AgR intermediates (R = alkenyl, alkyl). The collected experimental results indicate
that selectivity for E-alkene generation is dictated by the relative rates of monometallic β-hydride elimination and bimetallic
alkane elimination, which are impacted by several structural features of the catalyst. The mechanistic detail provided by these
studies will inform the development of second-generation hydrogenation catalysts.
INTRODUCTION
H pressure, gave high stereoselectivity for E-alkene products,
2
■
and was chemoselective for alkyne reduction in the presence of
other reducible functional groups such as aldehydes, ketones,
nitriles, and alkyl chlorides (Scheme 1). The observed
Catalytic hydrogenation of C−C multiple bonds, one of the
most studied topics in organometallic chemistry, remains
crucial both for conversion of bulk hydrocarbon feedstocks
and for late-stage manipulations of complex organic molecules.
Advancing new technologies in this area inherently relies on
stereoselective trans-addition of H to alkynes puts this catalyst
2
in a select group of systems exhibiting E-selective semi-
hydrogenation behavior, whether by direct hydrogena-
identifying systems capable of H activation. Many homoge-
2
1
4,21−26
tion
or by transfer hydrogenation or indirect reduc-
neous hydrogenation catalysts rely on single-site oxidative
1
−4
27−31
addition of H2.
Recent advances, some of which have
tion/deprotection routes.
The observed chemoselectivity
enabled unique catalytic selectivity and/or use of nonprecious
catalyst elements, move beyond the single-site paradigm to
for alkyne reduction in the presence of vulnerable functional
groups raises the possibility of using this technology for late-
stage introduction of trans-alkene moieties. While the optimal
system utilized the precious metals Ag and Ru, non-precious
metal analogues pairing Cu and/or Fe did show some activity
and represent opportunities for further development. Under-
32
involve heterolytic H activation by cooperative strategies.
2
Examples include cooperation between a metal site and a basic
5
−9
residue,
residue,
between a metal site and a Lewis acidic
1
0−13
14−17
between two different metal sites,
or between
18,19
nonmetal frustrated Lewis acid−base pairs.
standing the nature of heterobimetallic H activation, the
2
We recently communicated the catalytic activity of
heterobimetallic (NHC)M′-[M] complexes toward alkyne
hydrogenation (NHC = N-heterocyclic carbene, M′ = Cu or
catalytic mechanism, and the selectivity-determining reaction
Special Issue: Hydrocarbon Chemistry: Activation and Beyond
Received: April 29, 2016
20
Ag, [M] = FeCp(CO) or RuCp(CO) ). The optimal
2
2
catalyst, (IMes)AgRp (IMes = N,N′-bis(2,4,6-trimethylphenyl)-
imidazol-2-ylidene, Rp = RuCp(CO) ), was active at 1 atm of
2
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00356
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 1. E-Selective Alkyne Semihydrogenation by
Heterobimetallic Catalysis
with a focus on the Ru-containing compounds produced upon
catalyst decomposition under the reaction conditions. Origi-
nally, we assigned the (IMes)AgRp decomposition mixture
during catalysis as involving precipitation of metallic Ag and
20
1
formation of HRp (observed H NMR resonances in toluene-
d : δ = 4.52 and −12.04). Because a Hg drop did not poison
8
the reaction, metallic Ag was ruled out from being catalytically
relevant, but HRp was not tested as a catalyst due to initial
difficulties in isolation of a pure sample. Now, we report that an
isolated sample of pure HRp was found to catalyze
diphenylacetylene hydrogenation but did not provide high
conversion of diphenylacetylene or high selectivity for E-
stillbene (1) relative to the other reduction products (2 and 3)
(
entry 4).
1
Upon examining the H NMR characterization of isolated
1
HRp, however, we noticed that its observed H NMR
resonances (δ = 4.57 and −10.75 in toluene-d ) did not
8
match the resonances observed under catalytic conditions with
(IMes)AgRp. Furthermore, the IMes resonances observed after
catalysis with (IMes)AgRp did not match those for authentic
samples of free IMes ligand. Filtration of metallic Ag and
removal of organics from a catalytic product mixture provided
an isolated sample of the catalyst decomposition product, which
is now assigned as HRuCp(CO)(IMes) on the basis of NMR
and IR characterization. Isolating this species and re-exposing it
to the catalytic conditions did provide diphenylacetylene
hydrogenation (entry 5), but again the observed conversion
of diphenylacetylene and selectivity for 1 were significantly
lower than those for (IMes)AgRp. These two single-site Ru
catalysts (entries 4, 5) likely mediate trans-hydrogenation of
diphenylacetylene through a Ru-carbene pathway akin to that
steps will aid in the design of second-generation systems
featuring nonprecious metals capable of operating efficiently
and selectively under mild conditions.
In this Article, we present studies on the mechanism of
heterobimetallic H activation and on the heterobimetallic
2
alkene/alkane elimination reactions that contribute to the
observed catalytic selectivity. The H activation step is analyzed
2
by computational modeling of the reaction coordinate,
providing understanding of the key transition states and orbital
interactions involved in H2 cleavage. The alkene/alkane
elimination steps are analyzed by stoichiometric reactivity
studies of isolable models of catalytic intermediates. Addition-
ally, new insights into catalyst decomposition pathways are
disclosed. Collectively, a mechanistic picture emerges that will
aid future catalyst development in this area.
detailed by Fu
pressure. One of the Furstner catalyst mixtures utilized
̈
rstner for [Cp*Ru] catalysts at high H2
33
̈
both Ru and Ag, but does not provide high levels of catalytic
conversion or selectivity under the conditions optimal for
(IMes)AgRp (entry 6). Lastly, to contextualize the Cu/Ru
model studies presented below, catalytic results for (NHC)-
CuRp complexes are shown in entries 7 and 8.
RESULTS AND DISCUSSION
Control Experiments and Catalyst Decomposition. In
our preliminary communication of this system, our evidence
for a bimetallic catalytic mechanism was provided by comparing
results for diphenylacetylene hydrogenation obtained with
■
20
Bimetallic Activation of H . Within the bimetallic
2
(
(
IMes)AgRp (Table 1, entry 1) with those obtained with
mechanism that is operative in this system, we have proposed
IPr)AgOAc and Rp (IPr = N,N′-bis(2,6-diisopropylphenyl)-
that the metal−metal bonded (NHC)M′-[M] complexes react
2
imidazol-2-ylidene), neither of which gave productive catalysis
directly with H to generate (NHC)M′-H and H[M] pairs that
2
(
entries 2, 3). To examine this issue further, we have
subsequently examined a larger array of Ag-free Ru catalysts,
subsequently reduce alkyne substrates. Experimental evidence
for the feasibility of this proposed bimetallic H activation step
2
comes from three observations (Scheme 2): (a) experimental
observation of the microscopic reverse reaction, i.e., bimetallic
Table 1. Ag-Free and Ru-Free Control Experiments
34,35
H elimination from 0.5[(IPr)CuH] + HFp;
(b) H/D
2
2
exchange reactivity of (IPr)CuFp (Fp = FeCp(CO) ) observed
2
34
during dehydrogenative borylation conditions, which likely
involves activation of HD generated in situ; and (c) observation
of HRuCp(CO)(IMes) upon thermolysis of (IMes)AgRp in
a
a
entry
catalyst
(IMes)AgRp
conversion (%)
1:2:3
b
the absence of alkyne under H (1 atm), which is presumably
2
1
95
0
90:4:1
N/A
b
the result of intermediate formation of (IMes)AgH + HRp
followed by thermal decomposition of (IMes)AgH by the
2
(IPr)AgOAc
b
c
3
Rp2
7
4:3:0
36
known pathway and ligand substitution at Ru. However,
4
5
6
HRp
35
47
23
61
60
22:13:0
30:17:0
6:5:12
42:17:2
40:18:3
experimental data on the mechanism of H activation is elusive
2
HRuCp(CO)(IMes)
Cp*Ru(COD)Cl/AgOTf
(IMes)CuRp
d
due to the thermodynamically unfavorable nature of (NHC)-
20
b
M′-H + H[M] relative to (NHC)M′-[M] + H . This
2
7
b
motivated our pursuit of computationally modeling the
8
(IPr)CuRp
heterobimetallic H activation step.
In our initial communication on this system, we presented
2
20
a
1
b
From H NMR integration against an internal standard. From ref
0. Catalyst loading was 10%, i.e., 20% Ru. Catalyst loading was 20%
c
d
2
a computed transition state for H activation by an (IMe)CuRp
2
Ru and 20% Ag.
model (IMe = N,N′-dimethylimidazol-2-ylidene). The reaction
B
DOI: 10.1021/acs.organomet.6b00356
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 2. Experimental Evidence for Heterobimetallic H2
Activation
between (IMe)CuRp and H was calculated to be thermody-
2
namically unfavorable (ΔG
= 20.1 kcal/mol), with an even
smaller driving force than the analogous reaction involving
2
98 K
(
IMe)AgRp (ΔG2
= 14.1 kcal/mol). The calculated
98 K
⧧
activation barrier of ΔG
with the high temperature required for catalysis and was
determined using a computational method that had been
calibrated with experimental data previously. The calculated
activation parameters included a large and negative entropy of
activation (ΔH = 20.1 kcal/mol, ΔS = −29.8 eu), indicating a
highly organized transition-state structure. Here, to develop a
= 29.0 kcal/mol was consistent
2
98 K
37
⧧
⧧
deeper understanding of this H activation reaction, we provide
2
a thorough discussion of the computed reaction pathway
provided by analysis of atom/fragment charges, interatomic
bond orders, and natural bond orbital occupancies, all as
functions of the intrinsic reaction coordinate (IRC). These
results also expand on the discussions of an analogous Cu−Fe
Figure 1. Compuational modeling of the reaction between (IMe)-
2
0
CuRp and H
.
(Top) Relative energies of the structures from the
2
intrinsic reaction scan that were analyzed for this study. Bottom:
Transition-state structure TS1, with key internuclear distances labeled
in units of Å. // = discontinuity.
35
transition state for H elimination and of a Cu−Fe transition
2
37
state for C−Cl activation, both disclosed previously. For all
the results discussed below, the various computed parameters
have been tracked across several points along the computa-
tional IRC scan: the optimized reactant complex (R1), three
points from the IRC scan before the transition state, the
transition state (TS1), three points from the IRC scan after the
transition state, and the optimized product complex (P1). The
energetic profile of these states and the TS1 structure are
shown in Figure 1, and the other structures are provided as
Supporting Information.
Upon examining the calculated trajectory of H approach, it
2
is clear that the reaction proceeds via the H substrate initially
2
forming a σ-complex at the Cu site. As the H molecule
2
approaches Cu, it induces Cu−Ru bond dissociation as well as
NHC−Cu−Ru bending (133° in TS1 vs 179° in R1).
Structural templating of the bimetallic transition state involving
semibridging carbonyl ligands seems not to be operative in this
case, as both Cu···CO distances increase smoothly along the
reaction coordinate without exhibiting a local minimum at the
Figure 2. Selected internuclear distances for the reaction between
37,38
(IMe)CuRp and H , plotted along the calculated intrinsic reaction
transition state that is seen in other cases.
Figure 2 plots
2
coordinate scan. // = discontinuity.
key internuclear distances as functions of reaction coordinate.
The Cu−Ru distances increase smoothly as the Cu−H1
distance decreases smoothly. The Ru−H2 and H1−H2
distances change more abruptly: the Ru−H2 bond is almost
completely associated and the H1−H2 bond nearly dissociated
at the TS1 structure. These observations are consistent with a
late transition state resembling the products and with a model
that resembles deprotonation of a Cu(H ) σ-complex by the Rp
2
fragment. Prior to dissociation of the two metal-hydride
complexes to yield the P1 state, there appears to be a locally
C
DOI: 10.1021/acs.organomet.6b00356
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
preferred H1···H2 contact of 1.76−1.88 Å at later stages of the
reaction profile. These short H···H distances are in the
39,40
dihydrogen-bonding regime
and indicate polarization of
the H molecule. Experimental evidence for M−H···H−M′
2
dihydrogen bonding in a related system has been reported, and
the accompanying computational analysis indicated a dihy-
41
drogen bond strength of 1.7 kcal/mol in that system.
The reaction profile was also analyzed using computed bond
orders, determined by calculation of Wiberg bond indices
(
WBIs) derived from natural bond orbital (NBO) calculations.
The changes in WBI values across the reaction coordinate
Figure 3) reinforce conclusions reached by analyzing bond
(
Figure 3. Selected Wiberg bond index values for the reaction between
IMe)CuRp and H , plotted along the calculated intrinsic reaction
(
2
coordinate scan. // = discontinuity.
Figure 4. Selected natural charge values derived from natural
population analysis for the reaction between (IMe)CuRp and H2,
plotted along the calculated intrinsic reaction coordinate scan. // =
discontinuity.
distances. The Cu−Ru bond in R1 experiences a smooth
dissociation across the reaction coordinate and is partially
broken at the TS1 state. The Cu−H1 and Ru−H2 bonds are
partially formed at TS1, and the covalent H1−H2 bond is
mostly broken at TS1. In the post-TS1 portion of the reaction
coordinate, a locally preferred set of WBI values again indicates
the presence of dihydrogen bonding between (IMe)CuH and
HRp. Only upon complete dissociation at the P1 state are the
Cu−H1 and Ru−H2 bonds completely formed and the Cu−Ru
and H1−H2 bonds completely broken.
the increase in charge of the Rp fragment during the reaction,
indicating that much of the redox activity occurs at the two CO
units. Similar conclusions have been reached previously during
studies of bimetallic reaction pathways involving (NHC)M′-
42
[M] complexes through experimental and computation-
35,37,42
al
analyses, and all of these studies collectively highlight
Analysis of atom/fragment charges across the reaction
coordinate provides insight into the extent to which redox
activity is delocalized throughout the bimetallic reaction center
and its supporting ligands. Natural charge values derived from
natural population analysis are plotted in Figure 4. A traditional,
single-site oxidative addition process is generally thought to
involve homolytic H−H bond dissociation accompanied by
two-electron oxidation of the metal center. In the bimetallic H2
cleavage reaction being examined here, the reaction clearly
the importance of redox-active carbonyl ligands in the
(NHC)M′-[M] catalyst design. On the other hand, the IMe
and Cp ligands have similar charges in R1 and P1 and therefore
are not redox-active with regard to this reaction.
Analysis of NBO occupancies as functions of reaction
coordinate provides insight into the key orbital interactions
12
involved in bimetallic H cleavage. Selected NBO occupancies
2
are plotted in Figure 5 (left axis). As the reaction proceeds from
R1 to TS1, the occupancy of the σHH orbital (labeled BD(H−
H)) decreases, consistent with transfer of charge from this
orbital toward the catalyst upon formation of a σ-complex. The
loss of electron density (0.38 e) from the σHH-orbital is almost
completely matched by an increase in occupancy (0.33 e) of a
involves heterolytic H cleavage. In addition to the computa-
2
tional evidence for dihydrogen bonding disclosed above, the
H1−H2 bond undergoes cleavage to produce hydridic H1 (q =
−
0.42 in P1) and protic H2 (q = 0.18 in P1) as the reaction
0
.05 0.94 0.01
proceeds. Neither the Cu center itself nor the (IMe)Cu
fragment undergoes significant oxidation during the reaction,
indicating that the oxidation behavior is located mainly on the
Rp fragment. Indeed, the Rp charge is calculated to increase
during the reaction. The Ru center maintains constant charge
throughout the reaction, indicating that redox activity is
localized not on Ru but mainly on the ligands supporting Ru.
The cumulative charge of the two carbonyl ligands tracks with
Cu valence orbital (labeled LP*1(Cu)) that has s
p
d
hybridization (Figure 6). This is the only NBO that increases
occupancy significantly as a function of reaction coordinate, and
so it is assigned as the acceptor orbital to which the σHH-orbital
is donating charge. Additionally, the σ*HH-orbital (labeled
BD*(H−H)) gains occupancy (0.43 e) as the reaction
proceeds, consistent with back-donation from the catalyst as
the H molecule coordinates and cleaves. We could not identify
2
D
DOI: 10.1021/acs.organomet.6b00356
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
are accounted for by the increase in non-Lewis occupancy as
the reaction progresses to TS1. The remaining 0.54 e resembles
the buildup of σ*HH occupancy (0.43 e), allowing us to assign
the Cu−Ru bonding orbital (of predominantly Ru 4d
character) as the main back-donating orbital that triggers H−
H cleavage.
The traditional view of single-site H oxidative addition
2
involves formation of a σ-complex followed by concerted H−H
homolysis that is triggered by (a) donation from σ to an
HH
empty metal d-orbital and (b) back-donation from a filled metal
d-orbital into σ*HH (Scheme 3a). For the bimetallic H2
Scheme 3. Transition States and Orbital Interactions
Involved in (a) Monometallic and (b) Bimetallic H2
Addition
Figure 5. Selected NBO occupancy values for the reaction between
(
IMe)CuRp and H , plotted along the calculated intrinsic reaction
2
coordinate scan. Lewis occupancies of the H2 bonding and
antibonding NBOs and of a key Cu acceptor NBO are plotted on
the left axis. Total non-Lewis occupancy of the system is plotted on
the right axis. // = discontinuity.
activation reaction studied here, a complementary mechanistic
scheme has emerged from the computational data above
(
Scheme 3b). Heterobimetallic H addition involves formation
2
of a σ-complex at the electrophilic metal site, followed by H−H
+
heterolysis via H transfer to the proximal nucleophilic metal
site. The key orbital interactions that trigger this heterolytic H2
cleavage are (a) donation from σHH to an empty orbital at the
electrophilic metal site and (b) back-donation into σ*HH from a
filled orbital predominantly located at the nucleophilic metal
site. Such separation of the key orbital interactions across two
reactive sites strongly resembles H2 activation by p-block
frustrated Lewis pairs (FLPs), which involves donation from
Figure 6. Surface plots (0.04 isovalues) of the LP*1(Cu) (top) and
BD(Cu−Ru) (bottom) NBOs in R1 that are crucial acceptor and
43−47
σHH to a Lewis acid and from a Lewis base to σ*HH
.
In this
regard, the (NHC)M′-[M] catalysts can be viewed as d-block
donor orbitals, respectively, for bimetallic H cleavage. For clarity, the
2
analogues of the classical p-block FLPs.
distant H molecule has been omitted from the images.
2
Alkene/Alkane Elimination and Selectivity Effects.
Upon heterobimetallic H activation under catalytic conditions,
2
any NBO that experienced a corresponding decrease in
occupancy over the same part of the reaction coordinate.
Instead, we assign the donor orbital that transfers charge into
the σ*HH-orbital as being the Cu−Ru bonding orbital shown in
Figure 6. In the R1 state, this NBO has 87% Ru character
our proposal is that the alkyne substrate undergoes 1,2-
insertion with the transient (IMes)AgH intermediate to yield
an alkenyl species. On the basis of stoichiometric syn-
48
hydrocupration behavior reported for [(IPr)CuH]2, we
propose that this 1,2-insertion reaction produces the CC
geometry shown in structure A (Scheme 4a). Under catalytic
conditions, bimetallic alkene elimination is proposed to occur
through reaction of A with HRp, producing a Z-alkene and
regenerating the bimetallic catalyst. To test the feasibility of this
proposed step, we prepared alkenylcopper complex 4 by syn-
hydrocupration of diphenylacetylene. Complex 4 was found to
react with HRp to produce (IPr)CuRp and generate stilbene
with high Z-selectivity (Scheme 4b). The fact that both the
hydrocupration and alkene elimination steps proceed at room
0
.13 0.01 0.77
0.87 0.06 0.08
(
s
p
d
) and 13% Cu character (s
p
d
). In
subsequent states in the reaction profile where the Cu−Ru
bond has partially dissociated, this Cu−Ru bonding NBO was
not identified but instead was replaced by enhanced Rp→
Cu(IMe) donor/acceptor interactions and Rp→H donor/
2
acceptor interactions, with the Rp→Cu(IMe) donation
decreasing and the Rp→H donation increasing in energetic
2
contribution with the reaction coordinate. This behavior can be
tracked using the non-Lewis occupancy of each state (Figure 5,
right axis), which does increase as the reaction proceeds. The
Cu−Ru bonding orbital has 1.59-e occupancy in R1, and 1.05 e
temperature implies that H activation is the turnover-limiting
2
step in the catalytic process. The generation of Z-alkene from
E
DOI: 10.1021/acs.organomet.6b00356
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 4. (a) Proposed Alkyne Hydrogenation Mechanism
and (b) a Model Alkene Elimination Reaction
bimetallic alkane elimination from intermediate B is feasible,
consistent with the small amounts of alkane formed in typical
20
catalytic trials. No evidence for ethylene formation was
observed in this experiment, indicating that for model complex
5
the bimetallic alkane elimination reaction outcompetes the
desired β-hydride elimination reaction. We conducted an
analogous experiment with alkylcopper complex 6 (Scheme
5
c), which has been shown to undergo reversible β-hydride
50
elimination under mild conditions. Complex 6 was found to
react with HRp to produce (IPr)CuRp and generate a 2:1
mixture of alkane 7 and E-alkene 8. No evidence for Z-alkene
was observed. These observations show that, in a model system
with an alkyl ligand that more closely resembles the catalytically
relevant intermediate, β-hydride elimination can outcompete
alkane elimination and furthermore that it is selective for
extrusion of E-alkene products.
These results show that the relative rates of β-hydride
elimination and bimetallic alkane elimination are very sensitive
to steric effects, and an optimal steric environment must be
used in the catalyst design in order to produce high selectivity
for E-alkene generation. The real catalytic system, which
features less stable alkyl intermediates B, is thus finely tuned to
favor β-hydride elimination over bimetallic alkane elimination,
unlike the model systems 5 and 6, which are stable enough for
stoichiometric reactivity studies. These observations are
consistent with the dramatic impact that the NHC steric bulk
this model reaction is consistent with our model of a cascade
alkyne → Z-alkene → E-alkene process.
The feasibility of Z → E alkene isomerization was supported
20
by experiments reported in our initial communication,
showing that cis-stilbene was isomerized to trans-stilbene
under catalytic conditions provided that both the catalyst and
H were present. The proposed mechanism of isomerization
2
(Scheme 5a) involves 1,2-insertion of alkene with (IMes)AgH
20
Scheme 5. (a) Proposed Alkene Isomerization Mechanism
and (b and c) Model Reactions for Alkene vs Alkane
Elimination
has on catalytic selectivity. In addition, the use of Ag in place
of Cu in the optimal catalyst may derive, in part, from the more
facile rate of β-hydride elimination from a 4d metal compared
to a 3d metal. Lastly, the use of Rp in place of Fp in the optimal
catalyst may derive, in part, from the fact that HRp is a weaker
51
acid than HFp by ∼2 pK units, which serves to suppress the
a
bimetallic alkane elimination process that is in competition with
productive β-hydride elimination. All of these factors need to be
held in consideration when designing second-generation
catalysts.
CONCLUSIONS
■
In conclusion, in this study we have disclosed further details
regarding catalyst decomposition pathways, the mechanism of
H2 activation, the feasibility of the proposed mechanistic
cascade, and the factors relating to E-alkene selectivity in the
catalytic semihydrogenation of alkynes using heterobimetallic
catalysts of the type (NHC)M′-[M]. The conclusions drawn
from this study will inform future catalyst designs related to
hydrogenation catalysis. Specifically, in order to improve
catalytic activity (possibly with earth-abundant metals) and
operate at milder temperatures, it will be necessary both to (a)
lower the barrier for heterobimetallic H activation and (b)
2
stabilize the vulnerable (NHC)M′H intermediates. Knowledge
gained here about the key transition states and orbital
to produce an alkyl intermediate (B). The key, selectivity-
determining step then involves single-site β-hydride elimination
from B to produce the final E-alkene product. In order to
produce high selectivity for E-alkene relative to Z-alkene and
alkane, this β-hydride elimination reaction must (a) be under
thermodynamic control to suppress reversion to Z-alkene by β-
hydride elimination and (b) outcompete bimetallic alkane
elimination from reaction between B and HRp. To probe these
factors, we conducted model reactions with two known
alkylcopper species that have β-hydrogens. The ethyl complex
interactions involved in heterobimetallic H activation can
2
serve to guide development of second-generation systems that
activate H more readily, and known methods for stabilizing
2
52,53
(NHC)M′H can be pursued in concert.
In order to
maintain high selectivity for E-alkenes, it will be necessary for
such systems to allow for efficient alkene 1,2-insertion and β-
hydride elimination processes while suppressing heterobime-
tallic alkane elimination. The experimental results disclosed
here indicate that the relative rates of these processes will be
highly sensitive not only to the steric demands of the substrate
but also to the steric and electronic properties of the NHC, M′,
and [M] fragments in the catalyst species. Development of new
49
5
was found to react with HRp to produce (IPr)CuRp and
ethane (Scheme 5b). This observation establishes that
F
DOI: 10.1021/acs.organomet.6b00356
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
hydrogenation catalysts, with these factors in mind, is in
progress in our laboratories.
NMR (125 MHz, benzene-d
6
): δ 208.8 (CO), 139.0, 138.5, 136.4,
1
36.4, 136.1, 129.4, 121.8, 82.1 (Cp), 21.3, 18.8, 18.5. IR (solid,
cm ): 1885 (CO). Anal. Calcd for C H N RuO: C, 64.91; H, 6.05;
−1
27
30
2
N, 5.61. Found: C, 64.78; H, 6.20; N, 4.36.
Catalytic Trials. Catalytic hydrogenation trials with HRp and
EXPERIMENTAL SECTION
■
HRuCp(CO)(IMes) were conducted using the procedure we
General Experimental Considerations. Unless otherwise
specified, all reactions and manipulations were performed under
2
0
published previously, under identical reaction conditions, and
1
using the same method of product analysis by H NMR integration.
purified N in a glovebox or using standard Schlenk line techniques.
2
Preparation of 4. This species was prepared by hydrocupration of
Glassware was oven-dried prior to use. Toluene was sparged with
argon and dried using a Glass Contour Solvent System built by Pure
Process Technology, LLC. Xylenes, benzene, and all deuterated
solvents were purified by repeated freeze−pump−thaw cycles followed
by prolonged storage over activated 3 Å molecular sieves. H gas was
purchased from Praxair at a purity of 99.999% (5.0 UHP grade) and
diphenylacetylene using the procedure reported for hydrocupration of
4
8
3
-hexyne. In a scintillation vial, (IPr)CuOtBu (0.0633 g, 0.12 mmol,
1
equiv) and diphenylacetylene (0.0215 g, 0.12 mmol, 1 equiv) were
dissolved in benzene (10 mL). Triethoxysilane (40.5 μL, 0.23 mmol,
.9 equiv) was added to the solution, which was then stirred overnight
2
1
at room temperature. The solution was concentrated in vacuo to afford
an off-white powder. Yield: 0.0574 g, 0.091 mmol, 76%. The solid was
purified further by running through an O -removing catalyst column
2
(
RCI GetterMax 133T) and a drying column (Drierite). Literature
49 50 20
1
stored in a glovebox freezer at −36 °C. H NMR (400 MHz, benzene-
methods were used to synthesize 5, 6, (IMes)AgRp, (IPr)-
48
54
3
3
d ): δ 7.26 (t, J
4
= 4.0 Hz, 2H, p-H), 7.19 (s, 2H), 7.08 (d, J
=
CuOtBu, and NaRp. Unless otherwise specified, all other chemicals
were purchased from commercial sources and used without further
purification.
Physical Measurements. NMR spectra were recorded at ambient
temperature using Bruker Avance DPX-400 and Bruker Avance DRX-
6
H−H
H−H
3
.0 Hz, 4H, m-H), 7.05 (d, J
= 4.0 Hz, 2H), 6.99−6.87 (m, 3H),
H−H
3
6
.86−6.74 (m, 4H), 2.56 (sept, J
= 4.0 Hz, 4H, CH(CH ) ), 1.32
3 2
H−H
3
3
(d, J
= 4.0 Hz, 12H, CH(CH ) ), 1.08 (d, J
= 4.0 Hz, 12H,
H−H
3
2
H−H
1
3
1
CH(CH ) ). C{ H} NMR (125 MHz, benzene-d ): δ 185.6
3
2
6
1
13
1
(
1
NCCu), 177.1, 154.7, 145.8, 141.3, 135.3, 133.9, 130.5, 129.6,
26.0, 124.3, 122.3, 121.4, 29.0, 25.0, 23.7. Anal. Calcd for
5
00 spectrometers. H NMR and C{ H} NMR chemical shifts were
referenced to residual solvent peaks. FT-IR spectra were recorded on
solid samples in a glovebox using a Bruker ALPHA spectrometer fitted
with a diamond-ATR detection unit. GC-MS data were obtained using
an Agilent Technologies 7890A GC system interfaced to an Agilent
Technologies 5975C VL mass selective detector. Elemental analyses
were performed by Midwest Microlab, LLC (Indianapolis, IN, USA).
Computational Methods. Optimized structures for the reactants,
C H N Cu: C, 78.00; H, 7.50; N, 4.44. Found: C, 77.70; H, 7.61;
41 47
2
N, 4.06.
Alkene/Alkane Elimination Reactions with HRp. In a nitrogen-
filled glovebox, HRp (∼5 mg, 1 equiv) and either 4, 5, or 6 (1 equiv)
were dissolved separately in either toluene-d or benzene-d (0.5 mL
8
6
each), transferred to the same J. Young NMR tube, and sealed. The
tube was inverted several times to mix the solutions. The reactions
products, and transition state for (IMe)CuRp + H were obtained
from our previous study, along with the IRC scan data. All
calculations were performed using Gaussian09, revision B.01.
Density functional theory calculations were carried out using a hybrid
functional, BVP86, consisting of Becke’s 1988 gradient-corrected
Slater exchange functional combined with the VWNS local electron
2
20
1
were monitored by H NMR until all HRp was consumed. For the
5
5
1
reactions with 4 and 5, the organic products were detected by H
NMR. For the reaction with 6, the product mixture was transferred to
a vial under air, pipet-filtered through silica gel, diluted with THF, and
analyzed by calibrated GC-MS (with dodecane internal standard).
56
correlation functional and Perdew’s 1986 nonlocal electron correlation
57
functional.5 Mixed basis sets were employed: the LANL2TZ(f) triple-
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.6b00356.
■
8−60
60−62
ζ basis set
with effective core potential
was used for Cu and
*
S
6
3,64
Ru and the 6-311+G(d) basis set
was used for C, H, N, and O.
The sum of electronic and thermal free energies was used to calculate
ΔG values at 298 K. Natural population analysis was used to
determine atom/fragment charges, and Wiberg bond indices were
used to determine bond orders. Both were obtained from NBO v.
Spectral data (PDF)
65
3
.1 calculations within Gaussian09.
Preparation of HRp. A modified version of the literature
procedure was used. NaRp (0.0380 g, 0.155 mmol, 1 equiv) was
mixed with toluene (5 mL), and 4.0 M HCl in dioxane (0.038 mL,
.155 mmol, 1 equiv) was added. The resulting mixture was stirred
overnight at room temperature. The solution was filtered through a
51
AUTHOR INFORMATION
Corresponding Author
*E-mail (N. P. Mankad): npm@uic.edu.
Notes
■
0
Celite pad, and the filtrate was evaporated to dryness. Yield: 0.0131 g,
The authors declare no competing financial interest.
0
.59 mmol, 38%. The solid was stored in a glovebox freezer at −36 °C.
1
H NMR (400 MHz, toluene-d ): δ 4.57 (s, 5H, Cp), −10.75 (s, 1H,
8
ACKNOWLEDGMENTS
Ru−H).
■
Preparation of HRuCp(CO)(IMes). In a nitrogen-filled glovebox,
Funding was provided by the NSF (CHE-1362294) and the
Alfred P. Sloan Foundation (research fellowship to N.P.M.).
Computations were performed at the UIC Extreme Computing
facility. Prof. Vladimir Gevorgyan and his group provided
access to a GC-MS instrument.
(
IMes)AgRp (0.0051 g, 0.00788 mmol) was dissolved in xylenes (1
mL), transferred to a J. Young NMR tube, and sealed. (The transfer
was done in two steps: First, the maximum amount of catalyst was
dissolved in 0.7 mL of xylenes and transferred. The vial was washed
with another 0.3 mL of xylenes to transfer remaining solids.) The J.
Young tube was inverted multiple times to make sure that all solids
dissolved; then the tube was connected to a Schlenk line containing H2
gas (1 atm). After degassing the solution using three 5 min freeze−
REFERENCES
■
(1) Osborn, J. A.; Jardine, F. H.; Young, J. F.; Wilkinson, G. J. Chem.
Soc. A 1966, 1711−1722.
pump−thaw cycles, the sample was frozen again, exposed to H , and
2
allowed to thaw and equilibrate for 30 min. The J. Young tube was
then resealed and heated in an oil bath at 150 °C for 24 h. The tube
was then transferred back into the glovebox, filtered through Celite,
(2) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1976, 98, 4450−
4455.
(3) Crabtree, R. H.; Felkin, H.; Morris, G. E. J. Organomet. Chem.
1977, 141, 205−215.
(4) Friedfeld, M. R.; Shevlin, M.; Hoyt, J. M.; Krska, S. W.; Tudge,
M. T.; Chirik, P. J. Science 2013, 342, 1076−1080.
1
and dried in vacuo. Yield: 0.0026 g, 0.00520 mmol, 66%. H NMR
(
4
400 MHz, benzene-d ): δ 7.08 (s), 7.04 (s, 4H), 6.85 (s), 6.28 (s),
6
13
1
.58 (s, Cp), 2.21 (s), 2.18 (s), 2.14 (s), −11.98 (s, Ru−H). C{ H}
G
DOI: 10.1021/acs.organomet.6b00356
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(
5) Shvo, Y.; Czarkie, D.; Rahamim, Y. J. Am. Chem. Soc. 1986, 108,
(41) Levina, V. A.; Rossin, A.; Belkova, N. V.; Chierotti, M. R.;
7
(
(
400−7402.
́
Epstein, L. M.; Filippov, O. A.; Gobetto, R.; Gonsalvi, L.; Lledos, A.;
6) Noyori, R.; Ohkuma, T. Angew. Chem., Int. Ed. 2001, 40, 40−73.
Shubina, E. S.; Zanobini, F.; Peruzzini, M. Angew. Chem., Int. Ed. 2011,
50, 1367−1370.
7) Langer, R.; Leitus, G.; Ben-David, Y.; Milstein, D. Angew. Chem.,
Int. Ed. 2011, 50, 2120−2124.
8) Xu, X.; Kehr, G.; Daniliuc, C. G.; Erker, G. J. Am. Chem. Soc.
015, 137, 4550−4557.
9) Zuo, W.; Lough, A. J.; Li, Y. F.; Morris, R. H. Science 2013, 342,
080−1083.
10) Lin, T.-P.; Peters, J. C. J. Am. Chem. Soc. 2013, 135, 15310−
5313.
11) Lin, T.-P.; Peters, J. C. J. Am. Chem. Soc. 2014, 136, 13672−
3683.
12) Li, Y.; Hou, C.; Jiang, J.; Zhang, Z.; Zhao, C.; Page, A. J.; Ke, Z.
ACS Catal. 2016, 6, 1655−1662.
13) Cammarota, R. C.; Lu, C. C. J. Am. Chem. Soc. 2015, 137,
2486−12489.
14) Burch, R. R.; Muetterties, E. L.; Teller, R. G.; Williams, J. M. J.
Am. Chem. Soc. 1982, 104, 4257−4258.
15) Zhang, Y.; Roberts, S. P.; Bergman, R. G.; Ess, D. H. ACS Catal.
015, 5, 1840−1849.
16) Baranger, A. M.; Bergman, R. G. J. Am. Chem. Soc. 1994, 116,
822−3835.
17) Buchwalter, P.; Rose,
8−126.
18) Stephan, D. W.; Greenberg, S.; Graham, T. W.; Chase, P.;
́
(42) Karunananda, M. K.; Vazquez, F. X.; Alp, E. E.; Bi, W.;
Chattopadhyay, S.; Shibata, T.; Mankad, N. P. Dalton Trans. 2014, 43,
13661.
(
2
(
1
(
1
(
1
(43) Rokob, T. A.; Hamza, A.; Stirling, A.; Soos
́ ́
, T.; Papai, I. Angew.
Chem., Int. Ed. 2008, 47, 2435−2438.
(44) Guo, Y.; Li, S. Inorg. Chem. 2008, 47, 6212−6219.
(45) Grimme, S.; Kruse, H.; Goerigk, L.; Erker, G. Angew. Chem., Int.
Ed. 2010, 49, 1402−1405.
(46) Mu
9111−9118.
(47) Rokob, T. A.; Bako,
Chem. Soc. 2013, 135, 4425−4437.
̈
ck-Lichtenfeld, C.; Grimme, S. Dalton Trans. 2012, 41,
(
́
́
I.; Stirling, A.; Hamza, A.; Papai, I. J. Am.
(
1
(
(48) Mankad, N. P.; Laitar, D. S.; Sadighi, J. P. Organometallics 2004,
23, 3369−3371.
(49) Goj, L. A.; Blue, E. D.; Munro-Leighton, C.; Gunnoe, T. B.;
Petersen, J. L. Inorg. Chem. 2005, 44, 8647−8649.
50) Laitar, D. S.; Tsui, E. Y.; Sadighi, J. P. Organometallics 2006, 25,
405−2408.
51) Estes, D. P.; Vannucci, A. K.; Hall, A. R.; Lichtenberger, D. L.;
Norton, J. R. Organometallics 2011, 30, 3444−3447.
52) Frey, G. D.; Donnadieu, B.; Soleilhavoup, M.; Bertrand, G.
Chem. - Asian J. 2011, 6, 402−405.
53) Jordan, A. J.; Wyss, C. M.; Bacsa, J.; Sadighi, J. P.
(
2
(
3
(
2
(
(
2
(
́
J.; Braunstein, P. Chem. Rev. 2015, 115,
(
(
Hastie, J. J.; Geier, S. J.; Farrell, J. M.; Brown, C. C.; Heiden, Z. M.;
Welch, G. C.; Ullrich, M. Inorg. Chem. 2011, 50, 12338−12348.
(
M.; Repo, T. Nat. Chem. 2013, 5, 718−723.
(
Organometallics 2016, 35, 613−616.
(54) Banerjee, S.; Karunananda, M. K.; Bagherzadeh, S.; Jayarathne,
U.; Parmelee, S. R.; Waldhart, G. W.; Mankad, N. P. Inorg. Chem.
́
́ ́
19) Chernichenko, K.; Madarasz, A.; Papai, I.; Nieger, M.; Leskela,
̈
2
(
014, 53, 11307−11315.
20) Karunananda, M. K.; Mankad, N. P. J. Am. Chem. Soc. 2015, 137,
55) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
1
(
4598−14601.
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.;
Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.;
Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador,
P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision B.01; Gaussian, Inc.: Wallingford, CT, 2010.
21) Abley, P.; McQuillin, F. J. J. Chem. Soc. D 1969, 1503−1504.
22) Schleyer, D.; Niessen, H. G.; Bargon, J. New J. Chem. 2001, 25,
(
4
(
23−426.
23) Radkowski, K.; Sundararaju, B.; Fu
Ed. 2013, 52, 355−360.
24) Srimani, D.; Diskin-Posner, Y.; Ben-David, Y.; Milstein, D.
Angew. Chem., Int. Ed. 2013, 52, 14131−14134.
25) Liu, Y.; Hu, L.; Chen, H.; Du, H. Chem. - Eur. J. 2015, 21,
̈
rstner, A. Angew. Chem., Int.
(
(
3
(
(
495−3501.
26) Furukawa, S.; Komatsu, T. ACS Catal. 2016, 6, 2121−2125.
27) Tani, K.; Iseki, A.; Yamagata, T. Chem. Commun. 1999, 1821−
1
822.
28) Trost, B. M.; Ball, Z. T.; Jo
922−7923.
29) Shirakawa, E.; Otsuka, H.; Hayashi, T. Chem. Commun. 2005,
885−5886.
30) Luo, F.; Pan, C.; Wang, W.; Ye, Z.; Cheng, J. Tetrahedron 2010,
6, 1399−1403.
31) Shen, R.; Chen, T.; Zhao, Y.; Qiu, R.; Zhou, Y.; Yin, S.; Wang,
X.; Goto, M.; Han, L.-B. J. Am. Chem. Soc. 2011, 133, 17037−17044.
32) Fuchs, M.; Furstner, A. Angew. Chem., Int. Ed. 2015, 54, 3978−
982.
33) Leutzsch, M.; Wolf, L. M.; Gupta, P.; Fuchs, M.; Thiel, W.;
Fares, C.; Furstner, A. Angew. Chem., Int. Ed. 2015, 54, 12431−12436.
34) Mazzacano, T. J.; Mankad, N. P. J. Am. Chem. Soc. 2013, 135,
7258−17261.
35) Parmelee, S. R.; Mazzacano, T. J.; Zhu, Y.; Mankad, N. P.; Keith,
J. A. ACS Catal. 2015, 5, 3689−3699.
36) Tate, B. K.; Wyss, C. M.; Bacsa, J.; Kluge, K.; Gelbaum, L.;
Sadighi, J. P. Chem. Sci. 2013, 4, 3068−3068.
37) Karunananda, M. K.; Parmelee, S. R.; Waldhart, G. W.; Mankad,
N. P. Organometallics 2015, 34, 3857−3864.
38) Parmelee, S. R.; Mankad, N. P. Dalton Trans. 2015, 44, 17007−
7014.
39) Crabtree, R. H.; Siegbahn, P. E.; Eisenstein, O.; Rheingold, A.
L.; Koetzle, T. F. Acc. Chem. Res. 1996, 29, 348−354.
40) Custelcean, R.; Jackson, J. E. Chem. Rev. 2001, 101, 1963−1980.
(
7
(
5
(
6
(
̈
ge, T. J. Am. Chem. Soc. 2002, 124,
(
56) Becke, A. D. Phys. Rev. A: At., Mol., Opt. Phys. 1988, 38, 3098−
100.
57) Perdew, J. Phys. Rev. B: Condens. Matter Mater. Phys. 1986, 33,
3
(
8
(
(
822−8824.
58) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299−310.
59) Roy, L. E.; Hay, P. J.; Martin, R. L. J. Chem. Theory Comput.
(
3
(
̈
2
008, 4, 1029−1031.
60) Ehlers, A. W.; Bohme, M.; Dapprich, S.; Gobbi, A.; Ho
A.; Jonas, V.; Kohler, K. F.; Stegmann, R.; Veldkamp, A.; Frenking, G.
(
̈
llwarth,
̈
̀
̈
Chem. Phys. Lett. 1993, 208, 111−114.
(
1
(
(
(
(
61) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270−283.
62) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284−298.
63) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys.
1
(
980, 72, 650−656.
64) Clark, T.; Chandrasekhar, J.; Spitznagel, G. W.; Schleyer, P. V.
R. J. Comput. Chem. 1983, 4, 294−301.
65) Glendening, E. D.; Reed, A. E.; Carpenter, J. E.; Weinhold, F.
NBO Version 3.1.
(
(
(
(
1
(
(
H
DOI: 10.1021/acs.organomet.6b00356
Organometallics XXXX, XXX, XXX−XXX