Divergent reactivity of a new dinuclear xanthene-bridged bis(iminopyridine) di-nickel complex with alkynes

Article abstract of DOI:10.1039/c6dt04532d

The reaction of a dinucleating bis(iminopyridine) ligand L bearing a xanthene linker (L = N,N′-(2,7-di-tert-butyl-9,9-dimethyl-9H-xanthene-4,5-diyl)bis(1-(pyridin-2-yl)methanimine)) with Ni₂(COD)₂(DPA) (COD = cyclooctadiene, DPA = diphenylacetylene) leads to the formation of a new dinuclear complex Ni₂(L)(DPA). Ni₂(L)(DPA) can also be obtained in a one-pot reaction involving Ni(COD)₂, DPA and L. The X-ray structure of Ni₂(L)(DPA) reveals two square-planar Ni centers bridged by a DPA ligand. DFT calculations suggest that this species features Ni^I centers antiferromagnetically coupled to each other and their iminopyridine ligand radicals. Treatment of Ni₂(L)(DPA) with one equivalent of ethyl propiolate (HCCCO₂Et) forms the Ni₂(L)(HCCCO₂Et) complex. Addition of the second equivalent of ethyl propiolate leads to the observation of cyclotrimerised products by ¹H NMR spectroscopy. Carrying out the reaction under catalytic conditions (1 mol% of Ni₂(L)(DPA), 24 h, room temperature) transforms 89% of the substrate, forming primarily benzene products (triethyl benzene-1,2,4-tricarboxylate and triethyl benzene-1,3,5-tricarboxylate) in 68% yield, in a ca. 5:1 relative ratio. Increasing catalyst loading to 5 mol% leads to the full conversion of ethyl propiolate to benzene products; no cyclotetramerisation products were observed. In contrast, the reaction is significantly more sluggish with methyl propargyl ether. Using 1 mol% of the catalyst, only 25% conversion of methyl propargyl ether was observed within 24 h at room temperature. Furthermore, methyl propargyl ether demonstrates the formation of cyclooctatetraenes in significant amounts at a low catalyst concentration, whereas a higher catalyst concentration (5 mol%) leads to benzene products exclusively. Density functional theory was used to provide insight into the reaction mechanism, including structures of putative dinuclear metallocyclopentadiene and metallocycloheptatriene intermediates.

Full text of DOI:10.1039/c6dt04532d

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: R. L.
Hollingsworth, A. Bheemaraju, N. Lenca, R. L. Lord and S. Groysman, Dalton Trans., 2017, DOI:
1
0.1039/C6DT04532D.
Volume 45 Number
1
7
January 2016 Pages 1–398
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Dalton
Transactions
An international journal of inorganic chemistry
www.rsc.org/dalton
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-9226
PAPER
Joseph T. Hupp, Omar K. Farha et al.
Efficient extraction of sulfate from water using
framework
a Zr-metal–organic
rsc.li/dalton

Page 1 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
Divergent Reactivity of a New Dinuclear Xanthene-Bridged
Bis(iminopyridine) Di-Nickel Complex with Alkynes
1
1,2
3
4
Ryan L. Hollingsworth, Amarnath Bheemaraju, Nicole Lenca, Richard L. Lord,* Stanislav
Groysman*
1
1
Department of Chemistry, Wayne State University, 5101 Cass Ave, Detroit, MI, 48202.
Current Address: BML Munjal University, 7th KM Stone, NHꢀ8, Gurgaon, Haryana, India
2
3
Lumigen Instrument Center, Department of Chemistry, Wayne State University, 5101 Cass Ave, Detroit, MI,
2
4
8202. Department of Chemistry, Grand Valley State University, Allendale, MI, 49401.
*Eꢀmail: groysman@chem.wayne.edu, lordri@gvsu.edu

Dalton Transactions
Page 2 of 32
View Article Online
DOI: 10.1039/C6DT04532D
Abstract
The reaction of a dinucleating bis(iminopyridine) ligand L bearing a xanthene linker (L = N,N'ꢀ
2,7ꢀdiꢀtertꢀbutylꢀ9,9ꢀdimethylꢀ9Hꢀxantheneꢀ4,5ꢀdiyl)bis(1ꢀ(pyridinꢀ2ꢀyl)methanimine)) with
Ni (COD) (DPA) (COD = cyclooctadiene, DPA = diphenylacetylene) leads to the formation of a
(
2
2
new dinuclear complex Ni (L)(DPA). Ni (L)(DPA) can be also obtained in a oneꢀpot reaction
2
2
involving Ni(COD) , DPA and L. The Xꢀray structure of Ni (L)(DPA) reveals two squareꢀ
2
2
pyramidal Ni centers bridged by a DPA ligand. DFT calculations suggest this species features
I
Ni centers antiferromagnetically coupled to each other and their iminopyridine ligand radicals.
Treatment of Ni (L)(DPA) with one equivalent of ethyl propiolate (HCCCO Et) forms the
2
2
Ni (L)(HCCCO Et) complex. Addition of the second equivalent of ethyl propiolate leads to the
2
2
1
observation of cyclotrimerised products by H NMR spectroscopy. Carrying out the reaction
under catalytic conditions (1 mol% of Ni (L)(DPA), 24 h, room temperature) transforms 89% of
2
the substrate, forming primarily benzene products (triethyl benzeneꢀ1,3,5ꢀtricarboxylate and
triethyl benzeneꢀ1,2,4ꢀtricarboxylate) in 68% yield, in ca. 5:1 relative ratio. Increasing catalyst
loading to 5 mol% leads to the full conversion of ethyl propiolate to benzene products; no
cyclotetramerisation products were observed. In contrast, the reaction is significantly more
sluggish with methyl propargyl ether. Using 1 mol% of the catalyst, only 25% conversion of
methyl propargyl ether was observed within 24 h at room temperature. Furthermore, methyl
propargyl ether demonstrates formation of cyclooctatetraenes in significant amounts at low
catalyst concentration, whereas higher catalyst concentration (5 mol%) leads to benzene products
exclusively. Density functional theory was used to provide insight into the reaction mechanism,
including structures of putative dinuclear metallocyclopentadiene and metallocycloheptatriene
intermediates.

Page 3 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
Introduction
Transition metalꢀcatalyzed cyclotrimerisation of monoꢀsubstituted alkynes is a wellꢀ
1
, 2
studied transformation that yields a variety of aromatic compounds.
Several factors are of
importance in cyclotrimerisation, including regioselectivity (ratio of 1,3,5ꢀ to 1,2,4ꢀisomers), and
the overall selectivity of the cyclotrimerised product versus largerꢀsize rings (generally
1
cyclooctatetraenes) and/or oligomers. Cyclotrimerisation is generally proposed to involve
1
,2
oxidative coupling as the first step to give metallacyclopentadiene.
Subsequent [4+2]
2
cycloaddition yields the resulting substituted benzene. An alternative mechanism for the second
step involves alkyne insertion into a metalꢀcarbon bond to give metallacycloheptatriene, which
may undergo reductive elimination/cyclization to give benzene; this mechanism may explain
3
formation of observed byꢀproducts such as largeꢀsize rings. Many different monometallic
precursors serve as precatalysts for cycloaddition of alkynes, including various nickel
4
complexes. In recent years, there has been a growing interest in preꢀdesigned bimetallics for
5
cyclotrimerisation. Similar to the case of mononuclear catalysts, metallacyclopentadiene is
5
proposed to serve as the key intermediate in cyclotrimerisation. In contrast to the mononuclear
complexes, such metallacyclopentadiene is likely to bind to both metals, which may have a
5
significant impact on reactivity and selectivity of the catalyst. However, the bridging
metallocyclopentadiene structure is far from uniform in these complexes (Figure 1, aꢀd),
5
depending on the distance between the metals and the nature of the metal centers. Most of the
5
ꢀ10
previously reported bimetallics for cyclotrimerisation involved early transition metals,
whereas the use of late transition metal dinuclear complexes for alkyne trimerisation is relatively
5
, 11
uncommon.
In most of the previously reported cases, dinuclear complexes under investigation
were assembled with an aid of a bridging monodentate ligand (OMe, Cl) or by a direct metalꢀ

Dalton Transactions
Page 4 of 32
View Article Online
DOI: 10.1039/C6DT04532D
metal bond (W=W) prior to the reaction with alkynes. The lack of a stabilising dinucleating
ligand that secures a relatively fixed metalꢀmetal distance may lead to lower stability within the
catalyst, enabling decomposition into mononuclear species. Uyeda and coworkers have recently
reported a diꢀnickel complex, supported by a dinucleating ligand, that serves as highly effective
1
3
and selective catalyst for cyclotrimerisation of terminal alkynes (Figure 1, d). It is possible that
the key to the outstanding performance of Uyeda’s catalyst lies in the fact that the ligand utilized
was highly rigid, fixing the nickels at a specific distance from each other, in a sharp contrast to
1
3
the related mononuclear complexes that form mixtures of benzenes and cyclooctatetraenes. Our
group is investigating cooperative reactivity of late base transition metal complexes in the
1
4
activation of small molecules and catalysis. Following Uyeda’s report, we became interested in
the effect of nickelꢀnickel distance and overall ligand flexibility on cyclotrimerisation reactivity.
Herein we report synthesis, structure, and cyclotrimerisation reactivity of a new diꢀnickel
complex ligated by a semiꢀrigid dinucleating ligand, featuring two iminopyridine chelates
brought together by a xanthene bridge (Scheme 1).

Page 5 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
Figure 1. The structures of bridging metallocyclopentadiene intermediate in selected bimetallics.
Results and Discussion
Synthesis and Characterization Dinickel-Alkyne Complex
Ou initial attempt to synthesize lowꢀvalent dinickel complex with L involved treatment of
1
two equivalents of Ni(COD) with one equivalent of L. H NMR of the crude product suggested
2
formation of the bis(homoleptic) complex NiL , along with unreacted Ni(COD) . Next, we
2
2
turned our attention to the preformed diꢀnickel precursor Ni (COD) (DPA) (1) (COD =
2
2
1
5
cyclooctadiene, DPA = diphenylacetylene) originally reported by Muetterties and coworkers.
Treatment of 1 with one equivalent of L generates a dinuclear complex Ni (L)(DPA) (2).
2
Complex 2 is isolated in 62% yield as black crystalline solid. Prior isolation of precursor 1
turned out to be unnecessary: a combination of two equivalents of Ni(COD) , one equivalent of
2
DPA and one equivalent of L forms complex 2 in 90% isolated yield.
Scheme 1. Reaction pathways leading to compounds 2 and 3.

Dalton Transactions
Page 6 of 32
View Article Online
DOI: 10.1039/C6DT04532D
1
13
Complex 2 was characterized by H and C NMR spectroscopy, Xꢀray crystallography
and elemental analysis. The NMR spectrum of 2 indicated a C ꢀsymmetrical structure. Thus,
s
imino protons give rise to one singlet at 9.50 ppm, while the tꢀbutyl groups appear as a sharp
singlet at 1.37 ppm. In contrast, the xanthene methyl groups give rise to two singlets integrating
as three protons each, at 1.62 ppm and 2.02 ppm. The solidꢀstate structure of compound 2 is
given in Figure 2. The structure demonstrates two squareꢀplanar nickel centers bridged by a
diphenylacetylene ligand. The CꢀC bond distance of 1.369(4) Å in the bridging
diphenylacetylene ligand indicates strong activation as a result of interaction with two reducing
metal centers. This bond distance is comparable to the bridging diphenylacetylene bond distance
1
5
of 1.386(11) Å in Ni (DPA)(COD) . For comparison, we have recently reported a structure of a
2
2
squareꢀplanar mononuclear Ni(iminopyridine)(DPA) complex, in which the acetylene CC bond
1
6
of diphenylacetylene was determined to be 1.292(3) Å. Bond metrics of the redoxꢀactive
iminopyridine chelates also indicate a significant degree of reduction (see Figure 2 for selected
bond distances). Furthermore, NiꢀꢀꢀNi separation of 2.451(1) Å in 2 is shorter than the postulated
14
single NiꢀNi bond (2.617(2) Å) in Ni (DPA)(COD) . The iminopyridine bond lengths suggest a
2
2
1
4a, b
Ni(I) center,
but the approximate square planar geometry at each metal suggests lowꢀspin
Ni(II). To better understand the electronic structure of this complex, we interrogated these
species with density functional theory.

Page 7 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
Figure 2. Xꢀray structure (50% ellipsoids) of complex 2. H atoms and coꢀcrystallized solvents
hexane and ether) are omitted for clarity. Selected bond distances: Ni1ꢀꢀꢀNi2 2.451(1) Å, Ni1 –
C5 1.913(3) Å, Ni1 – C6 1.907(3) Å, Ni2 – C5 1.906(3) Å, Ni2 – C6 1.895(3) Å, C5 – C6
(
1.369(4) Å, N2 ꢀ C2 1.310(4) Å, N4 ꢀ C4 1.318(4) Å, C1 – C2 1.431(4) Å, C3 – C4 1.427(4) Å.
The geometry of a model monoꢀacetylene adduct (4) was optimized at the B3LYP/6ꢀ
1
7
3
1G(d) level of theory. tꢀBu and Me substituents on the xanthene were truncated to H. The
electronic structure of this optimized species is best described as an openꢀshell singlet with two
I
Ni ions antiferromagnetically coupled to both their iminopyridine ligands and the adjacent Ni
1
8
center, as shown in the corresponding orbital analysis in Figure 3. A simpler metric that further
supports this interpretation is Mulliken spins of 0.80 and –0.80 for the two Ni centers. We find
no evidence for covalent bonding between the two Ni centers; each bonding interaction is
nullified by the appropriate antibonding combination (Figure 3).

Dalton Transactions
Page 8 of 32
View Article Online
DOI: 10.1039/C6DT04532D
Figure 3. Corresponding orbitals with significant dꢀorbital character for 4. Overlaps are listed for
1
9
magnetically coupled orbitals. Images generated with GaussView.
I
Consistent with a Ni ion are iminopyridine bond lengths of 1.339, 1.420, and 1.380 Å for N –
im
1
5a, 15b, 20
C , C –C , and C –N , respectively.
Significant activation of the alkyne is predicted
im
im
py
py
py
with C–C elongation from 1.205 Å in free acetylene to 1.370 Å and deviation of the H–C–C
o
angles from linear to ~140 . The optimized structure of 4 is shown in Figure 4 along with monoꢀ
alkyne adducts (with the truncated xanthene backbone) of diphenylacetylene (2’), methyl
propiolate (5), and methyl propargyl ether (6) (vide infra). The bond lengths in 2’ are in good
agreement with those in Figure 2, deviating by only 0.02 Å.

Page 9 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
Figure 4. Comparison of the optimized geometries of 4 (top left), 2’ (top right), 5 (bottom left),
2
1
and 6 (bottom right). Important bond lengths are listed in Å. Images generated in CYLview.
Species 2’, 5, and 6 have similar electronic structures as 4, as demonstrated by Mulliken
spins of 0.74/–0.74, 0.66/–0.66, and 0.76/–0.76 for the Ni ions in these species. The
iminopyridine bonding is also consistent with an iminosemiquinone form, as shown in Figure 2,
with N –C bond lengths of 1.33ꢀ1.34 Å, C –C bond lengths of 1.42ꢀ1.43 Å, and C –N
py
im
im
im
py
py
bond lengths of ~1.38 Å. Significant C–C activation is observed for each species with elongation
from 1.216 to 1.387 Å for 2’, 1.207 to 1.384 Å for 5, and 1.206 to 1.375 Å for 6. For the latter
two species, structural data is only presented for the lowest energy regioisomer (see Figures S46

Dalton Transactions
Page 10 of 32
View Article Online
DOI: 10.1039/C6DT04532D
and S47). Collectively, these data suggest that regardless of the alkyne’s nature, these species are
all dinickel(I) and poised to undergo oxidative cyclization.
Reactivity of Complex 2 with Alkynes
To determine whether complex 2 can serve as an appropriate preꢀcatalyst for
cyclotrimerisation, its stoichiometric reactivity with two different terminal alkynes, ethyl
propiolate and methyl propargyl ether, was evaluated. Addition of one equivalent of ethyl
propiolate to 2 at room temperature forms complex 3 quantitatively, along with free
diphenylacetylene. Complex 3 features similar spectroscopic features to these of 2, indicating a
similar structure. We were not able to obtain Xꢀray quality crystals of compound 3, however, the
DFTꢀoptimized model 5 (with truncated xanthene backbone, vide supra) also indicates structure
similar to that of 2. Addition of the second equivalent of ethyl propiolate to 3 leads to full
turnover: proton NMR of the reaction mixture demonstrates formation of (mostly) the respective
1,2,4ꢀbenzene product, along with complex 3. Conducting the reaction of 2 with 2.5 equivalents
of ethyl propiolate in the presence of an internal standard indicates reaction stoichiometry as
shown in eq. 1. These findings indicate that “metallacyclopentadiene” intermediate is not stable
5
, 13
in our system, unlike previously reported systems,
and cannot be isolated. However, the
restoration of alkyneꢀbridged 3 following catalytic cycle supports the notion that the dinuclear
nature of [Ni (L)] remains intact throughout catalysis. For methyl propargyl ether, addition of
2
one or two equivalents of the alkyne to the C D solution of 2 produces a mixture of compounds,
6
6
which includes small amount of cyclotrimerisation products, along with additional unidentified
1
4f
compounds (see SI). It is important to emphasize that free ligand is not observed in any of
these experiments, suggesting that [Ni (L)] complexes do not decompose under these conditions.
2

Page 11 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
Ni (L)(DPA) (2) + 2.5 HC≡CCO Et → Ni (L)(HC≡CCO Et) (3) +
2
2
2
2
0
.39 (1,2,4ꢀisomer) + 0.11 (1,3,5ꢀisomer)
(1)
Next, we turned to investigate catalytic performance of complex 2 in cyclotrimerisation
of various terminal alkynes. The results are summarized in Table 1. Varying reaction conditions
were evaluated, including different catalyst loadings (1 or 5 mol%), different solvents
(
C D /ether/THF/CH CN), and/or different reaction temperatures (RT or 50 °C); all experiments
6 6 3
1
were run for 24 h. The degree of conversion of the alkyne substrates was determined by H NMR
spectroscopy using an internal standard, hexamethylbenzene. The respective amounts of
1
cyclotrimerised and cyclotetramerised products were elucidated using H NMR spectroscopy (by
1
3, 22, 23
comparison to previously reported spectra),
or by GCꢀMS. As the first five entries in
Table 1 indicate, complex 2 catalyzes efficiently cyclotrimerisation of two different propiolates,
ethyl and tꢀbutyl. At low catalyst concentrations (1 mol%) in nonꢀcoordinating solvent (C D ),
6
6
high conversion (≥90%) of the starting alkyne is obtained. Under these reaction conditions,
cyclotrimerisation is not the sole reaction outcome: approximately 25ꢀ30% of the observed
product is cyclooctatetraenes. Coordinating solvents (ethers or acetonitrile) have a negative
effect on the overall cycloaddition reactivity. Increasing catalyst loading to 5 mol% leads to
cyclotrimerisation exclusively for both propiolate substrates. Among the two possible benzene
products, this system demonstrates good selectivity for the 1,2,4ꢀisomer in C D .
6
6
With phenylacetylene, complex 2 forms benzene products exclusively, albeit less
efficiently (compared to propiolates). At 1 mol%, approximately 30% conversion is observed at
RT within 24 h, with 3:1 selectivity for 1,2,4ꢀtriphenylbenzene. Conversion can be increased by

Dalton Transactions
Page 12 of 32
View Article Online
DOI: 10.1039/C6DT04532D
increasing the amount of catalyst (from 1 to 5 mol%) and further by increasing reaction
temperature (to 92% total). Significantly, increasing catalyst loading and reaction temperature
maintains 3:1 ratio of the obtained 1,2,4/1,3,5ꢀbenzene products.
Methyl propargyl ether demonstrates a somewhat different reactivity from propiolates or
phenylacetylene. Conversion of methyl propargyl ether also depends on the reaction conditions:
only 25% of the starting material reacts after 24 h at RT with 1 mol% catalyst present, while
increasing the reaction temperature to 50 °C increases conversion rate to 49%. Consistent with
our previous observations (for ethyl propiolate), catalysis at 1 mol% produces mixtures of
benzenes and cylooctatetraenes. However, in contrast to the reactivity of propiolates and
phenylacetylene, cyclooctatetraenes are the dominant product under these conditions for methyl
propargyl ether. At 1 mol% catalyst and 50 °C, approximately 75% of the product constitutes
1
cyclooctatetraenes, as determined by H NMR spectroscopy and GCꢀMS. Interestingly, and in
agreement with the results observed for propiolates and phenyl acetylene, the reaction at 5 mol%
catalyst and room temperature produces benzene products exclusively. For cyclopropylacetylene,
only relatively low conversion was detected; no reactivity with trimethylsilylacetylene was
observed.

Page 13 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
Table 1. Reactivity of complex 2 in cyclotrimerisation reactions with terminal alkynes. All
1
reactions were carried out for 24 h, after which the nature of the products was determined by H
NMR spectroscopy and GCꢀMS.
R
solvent mol
%
T
%
Benzenes Combined
conversion % 1,2,4 / cyclotetraenes
%
1,3,5
(%)
21
C D
1
5
1
1
1
1
RT
RT
RT
RT
RT
RT
89
56 / 12
75 / 25
28 / 7
6
6
6
C D
6
100
0
a
Ether
THF
nd
nd
nd
0
a
nd
27 / 5
a
CH CN
nd
22 / 23
48 / 10
3
C D
6
94
96
36
6
6
C D
6
5
RT
82 / 14
<1
C D
1
1
5
5
1
5
5
1
1
1
RT
50 °C
RT
27
49
44
10 / 4
9 / 4
13
36
0
6
6
6
6
C D
6
C D
6
29 / 15
9 / 4
a
CH CN
RT
nd
nd
0
3
C D
6
RT
32
44
92
8
25 / 8
33 / 11
70 / 22
6 / 2
6
6
6
6
6
C D
6
RT
0
C D
6
50 °C
RT
0
C D
6
0
C D
6
50 °C
18
<1
15 / 3
ꢀ / ꢀ
0
C D
6
RT
6
-
a
nd = Not determined. We were not able to determine % conversion as volatiles had to be
removed prior to dissolving the reaction mixture in deuterated solvents.
We note that cycloaddition reactions of various terminal alkynes using related
mononuclear diazadiene and iminopyridine nickel(0) precursors has been reported by several

Dalton Transactions
Page 14 of 32
View Article Online
DOI: 10.1039/C6DT04532D
1
3, 24
groups.
For these complexes, the cycloaddition reaction often takes the cyclotetramerisation
route preferably. tom Dieck and coworkers demonstrated selective cyclotetramerisation of
2
4
several different monoꢀsubstutited alkynes by a diazadiene nickel precursors, among them ethyl
2
4d
and methyl propiolates. Uyeda and coworkers demonstrated that a mononuclear iminopyridine
nickel(0) complex (14, Figure 5) catalyses alkyne cycloaddition reactions of ethyl propiolate and
1
3
phenylacetylene. Specifically for ethyl propiolate substrate, mononuclear iminopyridine nickel
catalyst forms cyclooctatetraenes primarily (>70%); the combined yield of benzene products
(both 1,2,4ꢀ and 1,3,5 regioisomers) was only around 20%, and the total substrate conversion
was slightly higher than 30%. A related mononuclear bipyridineꢀnickel catalyst (15, Figure 5)
showed similar reactivity: ca. 30% total conversion, of which cyclooctatetraenes constituted the
majority of the product. Under the same reaction conditions, Uyeda’s dinuclear catalyst (13,
Figure 5) formed mostly benzenes (in ca. 90% yield).
To evaluate the effect of metalꢀmetal cooperativity engendered by our [Ni (L)] system,
2
we decided to directly compare its cycloaddition reactivity with the reactivity of mononuclear
and dinuclear nickel iminopyridine/napthyridine complexes reported by Uyeda and coworkers
(
Figure 5), using identical reaction conditions. Accordingly, we carried out catalytic experiments
under reaction conditions identical to these used by Uyeda and coworkers for ethyl propiolate
22 °C, 11 min, 1 mol % catalyst), and phenylacetylene (60 °C, 40 min, 5 mol% catalyst). The
(
results are described in Table 2 below. For ethyl propiolate, our catalyst enables 46% conversion
after 11 min; benzenes constitute nearly 80% of the product (1,2,4:1,3,5 ratio of 4:1). Similar
results were obtained for phenylacetylene: our system forms benzenes exclusively. These
experiments clearly support metalꢀmetal cooperativity in our catalyst, as all related mononuclear

Page 15 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
catalysts (with iminopyridine, bipyridine or diazadiene ligands) form primarily
cyclooctatetraenes.
Table 2. Comparison between the catalytic reactivity of complex 2 with the previously reported
mononuclear and dinuclear compounds. Reaction conditions for ethyl propiolate: 22 °C, 11 min,
1
mol% catalyst. Reaction conditions for phenylacetylene: 60 °C, 40 min, 5 mol% catalyst.
Structures of the mononuclear precatalysts (other than complex 2) is given in Figure 5 below;
their synthesis and reactivity is reported in ref. 13.
R
catalyst
%
Benzenes Combined
conversion % 1,2,4 / cyclotetraenes
%
1,3,5
(%)
11
10
25
18
0
0
25
24
2
46
100
33
28 / 7
68/ 22
6 / 2
12 / 2
59 / 5
97 / 3
24 / 4
12 / 4
1
1
1
3
4
5
32
2
64
1
1
1
3
4
5
100
53
40
Figure 5. Structures of preꢀcatalysts in Table 2.

Dalton Transactions
Page 16 of 32
View Article Online
DOI: 10.1039/C6DT04532D
DFT Analysis of Reaction Mechanism
To understand the cyclotrimerisation observed experimentally, we mapped the reaction
path for 4 with additional equivalents of acetylene. This model was chosen to minimize the
numerous regioisomers and rotamers involved in the mechanism of the substituted alkynes. Free
energies are based on B3LYP/6ꢀ311+G(d,p) single point refinements at the B3LYP/6ꢀ31G(d)
optimized geometries (see Computational Methods for full details). Binding of a second
acetylene results in a 90ꢀdegree twist of the first alkyne relative to the monoꢀalkyne adduct
(Figure 6). This dialkyne adduct, 7, is computed to be 23.78 kcal/mol higher in energy and
shows similar intraligand bonding in the iminopyridine as the monoalkyne adducts. However, the
Ni centers no longer have unpaired spins; the species is best described as a closedꢀshell singlet.
Figure 6. Optimized structure of the diꢀalkyne adduct 7 with important bond lengths (Å) labeled.
2
1
Images generated in CYLview.
Bond elongation in both alkynes of 1.33 Å (vs. 1.21 Å for free acetylene) is only slightly
less than the value of 1.37 Å for 1. Coupling of these activated alkynes to afford a
nickelacyclopentadiene (8) has a barrier of only 3.49 kcal/mol relative to 7, or 27.27 kcal/mol
relative to 4. The structure of this transition state (7-8-TS) is shown in Figure 7 (left). After
surmounting this barrier, an extremely stable nickelacyclopentadiene is formed (–27.54 kcal/mol

Page 17 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
4
vs. 4 and free acetylene). As shown in Figure 7 (right), the nickelacycle is η ꢀbound to the
8
second nickel center. This computed structure is similar to previously described ditungsten,
1
2
9
2
dicobalt,
and ditantalum species , and contrasts with the η coordination of the
13
nickelacyclopentadiene in Uyeda’s dinickel system.
Figure 7. Optimized structure of 7-8-TS (left) and 8 (right) with important bond lengths (Å)
2
1
labeled. Images generated in CYLview.
From intermediate 8, we explored the hypothesized [4 + 2] and alkyne insertion
pathways. No transition states or intermediates corresponding to a [4 + 2] cycloaddition or
transition state for direct insertion were located. [2 + 2] addition products to both Ni–C bonds
afforded intermediates 9a and 9b (Figure 8) with relative energies of 6.35 and 7.38 kcal/mol,
respectively. 9a and 9b differ in whether the newly formed nickelacyclobutene is syn or anti to
the xanthene in the dinucleating ligand backbone. In principle, addition could occur from the
bottom face of the complex as well, however, both of these regioisomers, 9c and 9d, were
computed to be much more endergonic at 16.21 and 47.63 kcal/mol (see SI Figure S48).

Dalton Transactions
Page 18 of 32
View Article Online
DOI: 10.1039/C6DT04532D
Figure 8. Topꢀdown views of the fused nickelabicycle optimized structures of 9a and 9b with
2
1
important bond lengths (Å) labeled. Images generated in CYLview.
As Figure 8 shows, bonding within the fused nickelabicycles of 9a and 9b is quite similar, with
the largest deviations in the Ni–C contacts (~0.005 Å). In both structures, the fiveꢀmembered
3
ring remains η ꢀbound to the second Ni ion. Despite the slight thermodynamic preference for 9a
vs. 9b, the corresponding cycloaddition barriers are 28.75 and 22.53 kcal/mol, respectively (see
SI Figure S49for 8-9a-TS and 8-9b-TS structures). Thus, we focused further reaction analysis on
regioisomer 9b, based on this kinetic preference.
Elongation of the Ni–C bond in this intermediate relieves ring strain and affords 10, a
nickelacycloheptatriene. The transition state for this ring expansion, 9b–10–TS (Figure 9, left),
has similar bonding within the metallacycle other than the significantly elongated Ni–C bond
from 1.989 to 2.317 Å. Similar to 8 that featured two short and one long C–C bond lengths, 10
has alternating short and long bond lengths suggesting three π bonds in this metallacycle (Figure

Page 19 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
3
9
, right). Importantly, the ring maintains η ꢀcoordination, seen in both 9b and 9b–10–TS, within
this intermediate. Because of this, the ring is concave toward the two Ni centers, causing the
hydrogens not involved in haptic binding to the second Ni to pucker above the ring.
Figure 9. Optimized structure of 9b-10-TS (left) and 10 (right) with important bond lengths (Å)
2
1
labeled. Images generated in CYLview.
The barrier for 9b-10-TS is only 0.41 kcal/mol higher in energy than 9b, suggesting the ring
expansion is effectively barrierless. 10 is exergonic by 27.35 kcal/mol vs. 8 and acetylene.
Reductive elimination of benzene from 10 is facile with a barrier of only 1.64 kcal/mol;
however, this reductive elimination requires a conformational change to the xanthene backbone
(10’, see Supporting Information Figure S50). This conformer is computed to be slightly more
stable (–0.63 kcal/mol) than 10.

Dalton Transactions
Page 20 of 32
View Article Online
DOI: 10.1039/C6DT04532D
To explore the origin of the COT isomers observed experimentally,
nickelacyclononatetraene 11 was optimized (Figure 10). All attempts to find the barrier for [2 +
2] addition of acetylene to 10 resulted in COTꢀcoordinated species 12 (–36.97 kcal/mol vs. 10
and acetylene) rather than barriers to intermediate 11.
2
1
Figure 10. Optimized structures of 11 (left) and 12 (right). Images generated in CYLview.
11 is thermodynamically competitive with intermediate 10, being only 0.83 kcal/mol uphill;
however, the lowꢀbarrier process to eliminate benzene makes this species difficult to rationalize
kinetically, unless alkyne insertion is essentially barrierless. As mentioned before, the ring
puckering of the nickelacycloheptatriene in 10 may hinder the approach of the incoming alkyne,
since the most accessible face of the Ni–C bonds is from below (where we observed unfavorable
thermodynamics in intermediates 9c and 9d). Work is ongoing to map out the transformation
from 10 to 12, and to probe the regioisomers and rotamers accessible in the esterꢀ and etherꢀ
substituted species. Due to steric congestion in the dinickel pocket, it is possible that significant
structural changes will occur in the nickelacycles once substituted alkynes are used, opening the
way for COT formation.

Page 21 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
Figure 11. Reaction profile for the production of benzene and cyclooctatetraenes with HCCH.
Intermediates 9c and 9d, which were determined to be much higher in energy, were excluded for
clarity. Transition states that were not explicitly located are labeled with *.
Figure 11 summarizes the thermodynamics and kinetics for the intermediates and transition
states presented in Figures 6-10. Based on this simplified alkyne model (HCCH), cyclization to

Dalton Transactions
Page 22 of 32
View Article Online
DOI: 10.1039/C6DT04532D
‡
form the nickelacyclopentadiene intermediate 8 is rateꢀdetermining with ꢀG = 27.27 kcal/mol.
The subsequent [2+2], ring expansion, and reductive elimination steps should be facile in
‡
comparison with ꢀG of 22.53, 16.21, and 1.64 kcal/mol, respectively. This may explain our
1
3
inability to isolate intermediate 8 like Uyeda, but it is important to caution that these energetics
may change significantly once the alkyne substituents are included. Computational work is
ongoing to better understand the regioselectivity of this system.
Summary and Conclusions
A new dinickel complex ligated by a semiꢀrigid xantheneꢀbridged bis(iminopyridine)
dinucleating ligand serves as a catalyst for cycloaddition of various terminal alkynes. Relatively
high conversion rates (≥90%) are observed for alkyl propiolates using low catalyst loading (1
mol%) at room temperature. For phenyl acetylene and methyl propargyl ether, heating and/or
higher catalyst loading are required to achieve comparable conversion rates during the same
reaction runtime. With alkyl propiolates and phenyl acetylene, efficient cyclotrimerisation is
obtained, giving 1,2,4ꢀisomers predominantly. While significant amount of cyclotetramerisation
was observed with methyl propargyl ether, it occurs only at low catalyst concernation (1 mol%).
At higher catalyst concentration (5 mol%), this substrate also forms benzenes exclusively. A
direct comparison (using identical reaction conditions) with related dinuclear
(diiminonaphthyridine) or mononuclear (iminopyridine, diazadiene, bipyridine) catalysts
indicated that our system, similarly to the previously reported diiminonaphthyridine also exhibits
bimetallic cooperativity which enables it to preferentially form benzene products.. There are,
however, several notable differences between the present system and the diiminonaphthyridine
(Uyeda’s) system. While both systems appear to retain their dinuclear composition throughout
catalysis, diiminonaphthyridine system demonstrates a significantly faster reaction rates, and

Page 23 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
somewhat higher selectivity for cyclotrimerisation. We hypothesize that the reactivity
differences stem from the different flexibily of the catalysts. Diiminonaphthyridine positions two
nickel centers side by side, in a rigid arragement that subsequently allows little flexibility in the
the nickelꢀnickel distance. In contrast, our xatheneꢀbased bis(iminopyridine) allows relatively
flexible nickelꢀnickel distance. Of course, two mononuclear nickel complexes allow the
“ultimate” flexibility in nickelꢀnickel distance, if bimetallic intermediates/transition states are
2
4
invoked for monometallic precursors. This flexibility can then be correlated with the catalyst
selectivity in the formation of benzenes vs. cyclooctatetraenes, as the intermediates/transition
states for cyclooctatetraenes require wider range of metalꢀmetal distances (2.38 – 2.77 Å) vs.
benzenes (2.51 – 2.65 Å), according to DFT calculations. Due to the constrained nature of the
Uyeda’s system with the “fixed” nickelꢀnickel distance, it shows a profound preference towards
cyclotrimerisation. A bimetallic complex made of two independent nickel centers as in tom
2
4
Dieck’s work can accommodate wider range of nickelꢀnickel distances, thus leading to
cyclooctatetraenes. Our complex, while being relatively rigid, is not as constrained as Uyeda’s,
and therefore leads preferentially to benzenes, but not as efficiently. It should be cautioned that
all calculations were performed on a model system in both the ligand backbone and the alkyne,
and future work will explore the role of each simplification on the proposed mechanistic
pathway. Future experimental studies will focus on the synthesis of bimetallic complexes in
more rigid ligand environments, the exploration of their reactivity, and the attempted isolation of
key metallacycle intermediates predicted by DFT.
Experimental

Dalton Transactions
Page 24 of 32
View Article Online
DOI: 10.1039/C6DT04532D
General Methods and Procedures. All reactions involving airꢀsensitive materials were executed
in a nitrogenꢀfilled glovebox. Diphenylacetylene, ethyl propiolate, methyl propargyl ether,
phenylacetylene, ethynyltrimethylsilane, and hexamethylbenzene were purchased from Aldrich.
tertꢀButyl propiolate was purchased from Alfa Aesar. Bis(cyclooctadiene)nickel(0) was
purchased from Strem. All chemicals were used as received. The synthesis of the ligand and
1
, 2
Ni (DPA)(COD) were previously reported. All nonꢀdeuterated solvents were purchased from
2
2
Aldrich and were of HPLC grade. The nonꢀdeuterated solvents were purified using an MBraun
solvent purification system. Benzeneꢀd was purchased from Cambridge Isotope Laboratories.
6
All solvents were stored over 3 Å molecular sieves. Compounds were generally characterized by
1
13
H and C NMR, Xꢀray crystallography, elemental analysis, gas chromatographyꢀmass
spectrometry, and electrospray ionization mass spectrometry. NMR spectra were recorded at the
Lumigen Instrument Center (Wayne State University) on an Agilent 400 MHz Spectrometer in
C D at room temperature. Chemical shifts and coupling constants (J) were reported in parts per
6
6
million (δ) and Hertz respectively. Elemental analysis was performed under ambient, airꢀfree
conditions by Midwest Microlab LLC.
Preparation of 2 from Ni (DPA)(COD) (1). A 2 mL hexane solution of Ni (DPA)(COD) (102.4
2
2
2
2
mg, 0.200 mmol) and a 2 mL THF solution of the ligand (106.1 mg, 0.200 mmol) were prepared
and cooled to ꢀ35 °C. The ligand was added dropwise to a stirring solution of Ni (DPA)(COD) .
2
2
Over the next 15 minutes, black precipitate formed and the solution color changed to brown. The
reaction was stirred for 2 hours, upon which the volatiles were removed in vacuo. The resulting
residue was dissolved in 5 mL of THF, covered with 10 mL of hexane, and allowed to stand for
24 hours at ꢀ35 °C. The supernatant was removed, and the solid washed with hexane and dried in
vacuo to afford black crystals (102.7 mg, 0.124 mmol, 62%). Xꢀray quality crystals were

Page 25 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
1
obtained from a saturated ether solution of 2 at ꢀ35 °C. H NMR (C D , 400 MHz) δ 11.31 (d, J
6
6
=
8.0 Hz, 2H, βꢀH on pyridine), 9.50 (s, 2H, iminoꢀH), 8.06 (d, J = 12.0 Hz, 2H, orthoꢀH on
phenylacetylene), 7.61 (d, 2H, J = 1.0 Hz, orthoꢀH on xanthene), 7.35 (d, J = 8.0 Hz, 2H, orthoꢀ
H on phenylacetylene), 7.31 (t, J = 4.0 Hz, 2H, γꢀH on pyridine), 7.15ꢀ7.25 (t + t, 2H + 2H, βꢀH
on pyridine + metaꢀH on phenylacetylene), 7.15 (d, 2H, paraꢀH on xanthene – hidden by
benzene), 7.08 (t, J = 8.0 Hz, 1H, paraꢀH on phenylacetylene), 6.89 (t, J = 8.0 Hz, 1H, paraꢀH
on phenylacetylene), 6.79 (t, J = 8.0 Hz, 2H, metaꢀH on phenylacetylene), 6.62 (d, J = 8.0 Hz,
2H, αꢀH on pyridine), 2.02 (s, 3H, methylꢀH), 1.61 (s, 3H, methylꢀH), 1.37 (s, 18H, tertꢀbutylꢀ
1
3
1
H); C{ H} NMR (C D , 400 MHz) δ 155.86, 153.85, 147.20, 146.64, 145.06, 145.00, 142.92,
6
6
1
41.68, 133.02, 130.69, 130.20, 129.16, 128.78, 127.29, 126.40, 124.35, 123.38, 123.07, 121.92,
19.39, 94.70, 89.36, 37.14, 35.94, 34.91, 32.15, 27.90, 23.99, 14.69. Anal. Calcd for
1
C H N Ni O: C, 71.22; H, 5.86; N, 6.78. Found: C, 70.99; H, 6.04; N, 6.50.
49
48
4
2
OneꢀPot Preparation of 2. A 2 mL hexane solution of diphenylacetylene (39.2 mg, 0.220 mmol)
was added dropwise to a 2 mL hexane solution of Ni(COD) (110.0 mg, 0.400 mmol). Over the
2
next 15 minutes, dark red precipitate formed and the solution color changed to red. The reaction
was stirred for 2 hours and then cooled to ꢀ35 °C. A 2 mL THF solution of the ligand (106.1 mg,
0.200 mmol) cooled to ꢀ35 °C was then added dropwise to the stirred solution of
diphenylacetylene and Ni(COD) . Over the next 15 minutes, black precipitate formed and the
2
solution color changed to brown. The reaction was stirred for 2 hours, upon which the volatiles
were removed in vacuo. The resulting residue was dissolved in 5 mL of THF, covered with 10
mL of hexane, and allowed to stand for 24 hours at ꢀ35 °C. The supernatant was removed, and
the solid washed with hexane and dried in vacuo to afford black crystals (149.3 mg, 0.181 mmol,
90%).

Dalton Transactions
Page 26 of 32
View Article Online
DOI: 10.1039/C6DT04532D
Preparation of 3 from 2. A 0.12 mL aliquot of a 0.247 M solution of ethyl propiolate in C D6
6
(
0.030 mmol) was added dropwise to a 2 mL C D solution of 2 (24.8 mg, 0.030 mmol).
6 6
Immediately, the solution color adopted a slight green tinge. The reaction was stirred for 1 hour,
upon which volatiles were removed in vacuo. The residue was washed with 5 mL of ether and
1
dried in vacuo to afford a black solid (19.1 mg, 0.026 mmol, 85%). H NMR (C D , 400 MHz) δ
6
6
11.99 (d, J = 4.0 Hz, 2H, βꢀH on pyridine), 9.24 (s, 2H, iminoꢀH), 7.71 (d, J = 4.0 Hz, 2H, orthoꢀ
H on xanthene), 7.43 (t, J = 6.0 Hz, 2H, γꢀH on pyridine), 7.19 (d, J = 2.0 Hz, 2H, paraꢀH on
xanthene), 7.12 (t, J = 8.0 Hz, 2H, βꢀH on pyridine), 6.59 (d, J = 8.0 Hz, 2H, αꢀH on pyridine),
4.26 (q, J = 8.0 Hz, 2H, methyleneꢀH on ethyl propiolate), 3.83 (s, 1H, acetyleneꢀH), 2.09 (s, 3H,
methylꢀH on xanthene), 1.69 (s, 3H, methylꢀH on xanthene), 1.36 (s, 18H, tertꢀbutylꢀH), 1.03 (t,
1
3
1
J = 8.0 Hz, 3H, methylꢀH on ethyl propiolate); C{ H} NMR (C D , 400 MHz) δ 167.79,
6
6
155.69, 154.68, 149.32, 146.63, 144.71, 133.24, 131.52, 125.72, 124.26, 122.11, 120.09, 75.97,
6
8.73, 59.69, 36.87, 36.24, 34.95, 32.08, 28.80, 15.19. Anal. Calcd for C H N Ni O ×C H O:
4
0
44
4
2
3
4
10
C, 64.42; H, 6.64; N, 6.83. Found: C, 64.97; H, 6.19; N, 6.32.
Treatment of 2 with 2 equiv. of Ethyl propiolate. A 0.24 mL aliquot of a 0.247 M solution of
ethyl propiolate in C D (0.060 mmol) was added dropwise to a 2 mL C D solution of 2 (24.8
6
6
6
6
mg, 0.030 mmol). Immediately, the solution color adopted a slight green tinge. The reaction was
stirred for 1 hour, upon which NMR spectra was obtained for the reaction mixture. Analysis of
the spectra revealed the absence of a diadduct compound and the presence of 2 and
cyclotrimerisation products.
Treatment of 2 with 2 equiv. of Ethyl propiolate in the presence of internal standard
(hexamethylbenzene). A solution of ethyl propiolate (0.075 mmol) hexamethylbenzene (0.006
mmol) in C D with was added dropwise to a 2 mL C D solution of 2 (24.8 mg, 0.030 mmol).
6
6
6
6

Page 27 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
Immediately, the solution color adopted a slight green tinge. The reaction was stirred for 1 hour,
upon which NMR spectra was obtained for the reaction mixture. Analysis of the spectra revealed
the absence of a diadduct compound and the presence of 3 (0.030 mmol) and cyclotrimerisation
products (0.011 mmol of the 1,2,4ꢀisomer and 0.003 mmol of the 1,3,5ꢀisomer). For analyzing
the presence of 3, the peaks at 12.00 ppm and 9.24 ppm (2H and 2H respectively) were used. For
analyzing the presence of the 1,3,5ꢀisomer, the peak at 9.14 ppm (3H) was used. For analyzing
the presence of the 1,2,4ꢀisomer, the peaks at 8.63 ppm and 7.98 ppm (1H and 1H respectively)
were used.
Treatment of 2 with 1 equiv. of Methyl propargyl ether or Phenylacetylene. A 0.10 mL aliquot of
a 0.296 M solution of methyl propargyl ether in C D (0.030 mmol) or a 0.13 mL aliquot of a
6
6
0
.228 M solution of phenylacetylene in C D (0.030 mmol) was added dropwise to a 2 mL C D
6 6 6 6
solution of 2 (24.8 mg, 0.030 mmol). Immediately, the solution color adopted a slight green
tinge. The reaction was stirred for 1 hour, upon which NMR spectra was obtained for the
reaction mixture. Analysis of the spectra revealed the absence of a monoadduct compound and
the presence of 2, diphenylacetylene, cyclotrimerisation products, and unidentifiable
compound(s).
Treatment of 2 with 2 equivalents of Methyl propargyl ether or Phenylacetylene. A 0.20 mL
aliquot of a 0.296 M solution of methyl propargyl ether in C D (0.060 mmol) or a 0.26 mL
6
6
aliquot of a 0.228 M solution of phenylacetylene in C D (0.060 mmol) was added dropwise to a
6
6
2
mL C D solution of 2 (24.8 mg, 0.030 mmol). Immediately, the solution color adopted a slight
6 6
green tinge. The reaction was stirred for 1 hour, upon which NMR spectra was obtained for the
reaction mixture. Analysis of the spectra revealed the absence of a monoadduct compound and
the presence of 2, diphenylacetylene, cyclotrimerisation products, and an unidentified compound.

Dalton Transactions
Page 28 of 32
View Article Online
DOI: 10.1039/C6DT04532D
General Procedure for Catalytic Cyclotrimerisation Reactions (See Table 1). A scintillation vial
containing 2 (8.3 mg, 0.010 mmol) was treated with the appropriate amount of monosubstituted
alkyne in a C D solution or another solvent of choice, with hexamethylbenzene as the internal
6
6
integration standard. Immediately, the solution color changed to green. The reaction mixture was
stirred for 24 hours at the appropriate temperature, upon which NMR spectra was obtained for
the reaction mixture in C D . For reaction mixtures in other solvents, volatiles were removed in
6
6
vacuo and the resulting residue was reꢀdissolved in C D . Analysis of the spectra revealed the
6
6
quantitative presence or absence of cyclotrimerisation products. For selected reactions, an aliquot
of the reaction mixture was analyzed by GCꢀMS for the quantitative presence of
cyclooctatetraenes and benzenes.
Electronic supplementary information (ESI) available: NMR spectra, crystal and refinement
data for compound 2, and Cartesian coordinates and energetics for all calculated species. The
crystallographic data in cif format for compound 2 was deposited at the Cambridge
Crystallographic Data Centre under the following number: CCDC 1519915.
Acknowledgements
SG thanks Wayne State University for the initial sponsorship of this project and the
National Science Foundation (NSF) for current support under grant number CHEꢀ1349048. RLL
thanks the Research Corporation (Cottrell College Science Award) and the NSF for
computational resources (CHEꢀ1039925 to the Midwest Undergraduate Computational
Chemistry Consortium). Compounds characterization was carried out at Lumigen Instrument
Center at Wayne State University.
References

Page 29 of 32
Dalton Transactions
View Article Online
DOI: 10.1039/C6DT04532D
1
2
. (a) S. Saito and Y. Yamamoto, Chem. Rev. 2000, 100, 2901. (b) P. Kumar and J. Louie,
In TransitionꢀMetalꢀMediated Aromatic Ring Construction; K. Tanaka, Ed.; John Wiley
&
Sons, Inc.: Hoboken, NJ, 2013; pp. 37−70. (c) C. Wang and Z. Xi, Chem. Commun.,
2
007, 5119ꢀ5133.
. (a) R. Diercks, B. E. Eaton, S. Gürtzgen, S. Jalisatgi, A. J. Matzger, R. H. Radde and K.
P. C. Vollhardt, J. Am. Chem. Soc., 1998, 120, 8247ꢀ8248; (b) D. R. McAlister, J. E.
Bercaw and R. G. Bergman, J. Am. Chem. Soc., 1977, 99, 1666ꢀ1668; (c) D. P. Smith, J.
R. Strickler, S. D. Gray, M. A. Bruck, R. S. Holmes and D. E. Wigley, Organometallics,
1
992, 11, 1275ꢀ1288; (d) O. V. Ozerov, F. T. Ladipo and B. O. Patrick, J. Am. Chem.
Soc., 1999, 121, 7941ꢀ7942.
3
4
. (a) M. O. Albers, D. J. A. De Waal, D. C. Liles, D. J. Robinson, E. Singleton and M. B.
Wiege, J. Chem. Soc., Chem. Commun., 1986, 1680ꢀ1682; (b) B. Hessen, A. Meetsma, F.
van Bolhuis and J. H. Teuben, Organometallics, 1990, 9, 1925ꢀ1936; (c) F. Calderazzo,
G. Pampaloni, P. Pallavicini, J. Straehle and K. Wurst, Organometallics, 1991, 10, 896ꢀ
9
01.
. For selected examples, see: (a) W. Reppe, O. Schlichting, K. Klager and T. Toepel,
Justus Liebigs Ann. Chem., 1948, 560, 1ꢀ92; (b) P. Alphonse, F. Moyen and P. J.
Mazerolles, J. Organomet. Chem. 1988, 345, 209ꢀ216; (c). P. Bicev, A. Furlani and G.
Sartori, Gazz. Chim. Ital. 1973, 103, 849ꢀ858; (d) S. K. Rodrigo, I. V. Powell, M. G.
Coleman, J. A. Krause and H. Guan, Org. Biomol. Chem., 2013, 11, 7653ꢀ7657; (e) C.
Xi, Z. Sun and Y. Liu, Y. Dalton Trans. 2013, 42, 13327ꢀ13330.
5
6
7
. K. Yamamoto, H. Nagae, H. Tsurugi and K. Mashima, Dalton Trans., 2016, 45, 17072ꢀ
1
7081.
. (a) F. A. Cotton, W. T. Hall, K. J. Cann and F. J. Karol, Macromolecules, 1981, 14, 233ꢀ
36; (b) F. A. Cotton and W. T. Hall, Inorg. Chem., 1981, 20, 1285ꢀ1287.
. (a) M. Kakeya, T. Fujihara, T. Kasaya and A. Nagasawa, Organometallics, 2006, 25,
131ꢀ4137; (b) M. Matsuura, T. Fujihara, M. Kakeya, T. Sugaya and A. Nagasawa, J.
2
4
Organomet. Chem., 2013, 745, 288ꢀ298.
8
. (a) M. H. Chisholm, B. W. Eichhorn, K. Folting and J. C. Huffman, Organometallics,
1
989, 8, 49ꢀ66; (b) M. H. Chisholm, B. W. Eichhorn, K. Folting and J. C. Huffman,