Organometallics
Article
bound to the phosphine group (Figures 4 and 5). No transition
centers possessing distinctly different coordination environ-
ments. In particular, in comparison to the tetra(pyridyl)-
naphthyridine complexes studied earlier, the new dicopper
state was found between this intermediate and 2b . In contrast
C
with TS1βC, both 2b and the 4 feature full coordination of
C
B
the DPEOPN ligand donors to the copper centers, and the
counteranion settles into different locations suggesting a
complex interplay between these rearrangements along the
reaction pathway.
Since many potential transition states converged, the output
data from the DFT calculations were analyzed to find
correlations between the energy barriers and 15 different
species incorporate a bulky, soft donor in one position. As
t
shown by these investigations, installation of this Bu P− group
2
via −CH − or −O− linkers opens new pathways for ligand
2
degradation upon reaction with nucleophilic reagents.
Interestingly, DPEOPN allowed observation of redox behavior,
with reduction to a species that undergoes ligand−ligand,
radical-like coupling. Despite the differentiated copper
environments, the dicopper(I) center appears to favor binding
of σ-type hydrocarbyl ligands in a symmetrical σ,σ-binding
mode, as found for analogous symmetrical complexes of
DPFN. However, as suggested by the structure of [5]NTf2,
bimodal hard−soft binding modes for a substrate may in some
cases be slightly favored. In addition, C−H bond activations of
parameters with a significant influence on the energy of the
transition states. The first is the position of the alkyne
coordinated to the copper centers; the transition states have a
lower energy when the alkyne is bound to the copper on the
di(pyridyl) side. The second important factor appears to be the
location of the counterion. In the lower-energy transition
states, the alkyne is the closest fragment to the counterion. The
third parameter is the Cu···Cu interatomic distance, which
terminal alkynes readily occur at the new Cu sites, and DFT
2
studies indicate that these occur preferentially by dissociation
of a pyridyl (rather than the phosphino) group to activate the
alkyne. Variations of the electronic and steric properties of
such N,N−P binucleating ligands should point the way to a
wider array of related transformations.
2
=
0.71; Figure S48). This correlation likely stems from the
sterically bulky DPEOPN side-arms shielding the active site of
The μ-phenyl organocopper complexes reported here
activate aromatic and alkynyl C−H bonds. Experimental and
computational mechanistic studies of the stoichiometric C−H
bond activation of an alkyne indicate a concerted proton
transfer mechanism that occurs on a complex energy surface,
which is very sensitive to the coordination mode of the ligand
arms, as well as the presence and position of the counterion. In
this way, this study provides a basis for understanding the role
of metal−metal cooperativity in the bond activation. A similar
mechanism may be operative in the product-forming step of
the catalytic azide−alkyne cycloaddition (CuAAC) reaction
+
2b . To allow the alkyne and the phenyl fragments to come
close enough for proton transfer to occur, without creating
additional steric repulsion, the bimetallic binding site must be
expanded by separating the two copper centers. The Cu···Cu
distance of 2.95 Å at TS1βC (Figure 4) allows the system to
accommodate the alkyne while maintaining coordination of the
DPEOPN ligand.
The alkyne activation is associated with a moderate kinetic
isotope effect (KIE) of k /k = 1.8 ± 0.2, as determined by
H
D
monitoring the conversion of [2b]NTf in the presence of a
2
+
21
1
0:10 mixture of p-tolylacetylene and p-tolylacetylene-d (in
demonstrated for [Cu (DPFN)] complexes. Further inves-
1
2
dichloromethane). This indicates that the C−H bond
activation influences the reaction rate, but perhaps not as
strongly as suggested by the specific pathway of Figure 5,
which has a computed KIE value of k /k = 9.3, determined
tigations of the structural and electronic influences on related
bond activation chemistry is expected to inform development
of additional dicopper-based catalytic transformations.
H
D
by comparing H- and D-transfer in TS1βC (Figure 5) with p-
tolylacetylene and dichloromethane solvent. However, in terms
of energy, the difference between the experimental and
computational KIEs is only 1.0 kcal/mol. Furthermore, the
ASSOCIATED CONTENT
sı Supporting Information
■
*
2
'
Additional experimental and computational details,
order conditions (1:10 equiv; Figure S28) reveals a first-order
dependence on complex, while the observed rates determined
latter result suggests a somewhat complex reaction mechanism,
as might be expected for a bimetallic reaction center with a
multidentate ligand and variable cation−anion interactions, as
detailed by the DFT calculations. Nevertheless, these
investigations provide useful insights into how cooperative
behavior between two interacting copper centers might be
employed to promote C−H bond activation processes. Thus,
the DFT results suggest a viable pathway for C−H bond
activation at the dicopper center, but a more detailed
description of the reaction mechanism, including competing
pathways, will require more in-depth experimental and
computational investigations.
supplementary figures and tables as described in the
text, crystallographic figures and data (PDF)
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
Corresponding Authors
■
David Balcells − Hylleraas Centre for Quantum Molecular
CONCLUSIONS
The DPEPN and DPEOPN ligands described here allow
formation of dicopper species with closely interacting copper
■
1
871
Organometallics 2021, 40, 1866−1873