Reaction of [TpRh(C2H4)2] with Dimethyl Acetylenedicarboxylate: Identification of Intermediates of the [2+2+2] Alkyne and Alkyne–Ethylene Cyclo(co)trimerizations

Article abstract of DOI:10.1002/chem.201601927

The reaction between the bis(ethylene) complex [TpRh(C₂H₄)₂], 1, (Tp=hydrotris(pyrazolyl)borate), and dimethyl acetylenedicarboxylate (DMAD) has been studied under different experimental conditions. A mixture of products was formed, in which TpRh^Ispecies were prevalent, whereas the presence of trapping agents, like water or acetonitrile, allowed for the stabilization and isolation of octahedral TpRh^IIIcompounds. An excess of DMAD gave rise to a small amount of the [2+2+2] cyclotrimerization product hexamethyl mellitate (6). Although no catalytic application of 1 was achieved, mechanistic insights shed light on the formation of stable rhodium species representing the resting state of the catalytic cycle of rhodium-mediated [2+2+2] cyclo(co)trimerization reactions. Metallacyclopentene intermediate species, generated from the activation of one alkyne and one ethylene molecule from 1, and metallacyclopentadiene species, formed by oxidative coupling of two alkynes to the rhodium centre, are crucial steps in the pathways leading to the final organometallic and organic products.

Full text of DOI:10.1002/chem.201601927

DOI: 10.1002/chem.201601927
Full Paper
&
Rhodium Complexes
Reaction of [TpRh(C2H4)2] with Dimethyl Acetylenedicarboxylate:
Identification of Intermediates of the [2+2+2] Alkyne and
Alkyne–Ethylene Cyclo(co)trimerizations
Giovanni Bottari,[a] Laura L. Santos,*[a] Cristina M. Posadas,[a] Jesffls Campos,[a] Kurt Mereiter,[b]
and Margarita Paneque*[a]
Abstract: The reaction between the bis(ethylene) complex
[TpRh(C2H4)2], 1, (Tp=hydrotris(pyrazolyl)borate), and di-
methyl acetylenedicarboxylate (DMAD) has been studied
under different experimental conditions. A mixture of prod-
ucts was formed, in which TpRhI species were prevalent,
whereas the presence of trapping agents, like water or ace-
tonitrile, allowed for the stabilization and isolation of octahe-
dral TpRhIII compounds. An excess of DMAD gave rise to
a small amount of the [2+2+2] cyclotrimerization product
hexamethyl mellitate (6). Although no catalytic application
of 1 was achieved, mechanistic insights shed light on the
formation of stable rhodium species representing the resting
state of the catalytic cycle of rhodium-mediated [2+2+2] cy-
clo(co)trimerization reactions. Metallacyclopentene inter-
mediate species, generated from the activation of one
alkyne and one ethylene molecule from 1, and metallacyclo-
pentadiene species, formed by oxidative coupling of two al-
kynes to the rhodium centre, are crucial steps in the path-
ways leading to the final organometallic and organic prod-
ucts.
Introduction
the reaction, cyclohexadienes are the usual products of reduc-
tive elimination, however an additional pathway involving the
formation of linear hexatrienes and their subsequent rear-
rangement to cyclic products has also been demonstrated.[6]
The reductive elimination of the cyclic structure in each case
and the coupling of a new pair of alkynes regenerate the met-
allacyclopentadiene species in the catalytic process. Many com-
plexes based on Groups 8–10 metals[7] have proved to be effi-
cient catalysts for [2+2+2] cyclotrimerization reactions and
among them rhodium complexes show wide applicability.[8] In
fact, Wilkinson’s catalyst[9] and CpRh/IndRh complexes[10] (Cp=
cyclopentadienyl, Ind=indenyl) have been the object of kinet-
ic and theoretical studies aimed to elucidate the reaction
mechanism although direct detection and identification of the
intermediates involved in the catalytic cycle are still limited.[11]
Metal-mediated [2+2+2] alkyne cyclotrimerization[1] and
alkyne-alkene (co)cyclotrimerization[2] constitute extremely
useful classes of reactions in organic synthesis for the prepara-
tion of highly substituted benzenes or cyclohexadienes in a se-
lective, efficient, and atom-economical manner. Formation of
metallacyclopentadiene intermediates has often been ob-
served and invoked as the first step in these oxidative cou-
plings.[3] For the first case, and from a mechanistic point of
view,[4] the coordination of a third alkyne may result in the for-
mation of a metallacycloheptatriene by insertion of the unsatu-
rated molecule into a MÀC bond, or alternatively in metalla-bi-
cyclic species similar to a norbornadiene or bicyclo-hepta-
triene[5] (Scheme 1). If both alkenes and alkynes are involved in
[a] Dr. G. Bottari, Dr. L. L. Santos, Dr. C. M. Posadas, Dr. J. Campos,
Prof. Dr. M. Paneque
Instituto de Investigaciones Químicas
Departamento de Química Inorgµnica
Centro de Innovación en Química Avanzada (ORFEO-CINQA)
Consejo Superior de Investigaciones Científicas (CSIC)
Universidad de Sevilla
Avenida AmØrico Vespucio 49, 41092 Sevilla (Spain)
E-mail: paneque@iiq.csic.es
laura@iiq.csic.es
[b] Prof. Dr. K. Mereiter
Department of Chemistry
Vienna University of Technology
Getreidemarkt 9/164SC, 1060 Vienna (Austria)
Supporting information and the ORCID number(s) for the author(s) of this
article can be found under http://dx.doi.org/10.1002/chem.201601927.
Scheme 1. Simplified reaction pathways for metal-catalyzed acetylene and
ethylene cycloadditions.
Chem. Eur. J. 2016, 22, 13715 –13723
13715
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
In previous studies, we investigated the reactions of two
TpMe Ir (TpMe =hydrotris(3,5-dimethylpyrazolyl)borate) com-
pounds with dimethyl acetylenedicarboxylate (DMAD). One of
the compounds bearing a 2,3-dimethylbutadiene ligand yield-
ed the water adduct of a highly stable iridacycloheptatriene,
with displacement of the diene ligand (Scheme 2(a)).[5,12a] In
I
2
2
turn, the related ethylene complex [TpMe Ir(C2H4)2] eventually
2
Scheme 3. Reductive elimination of a substituted naphthalene by the oxida-
tion of a benzoiridacycloheptatriene.
afforded a TpMe Ir allyl species formed by coupling of two
III
2
DMAD and one ethylene, through the steps depicted in
Scheme 2(b).[13] The metallacycle of A can be trapped as an
aqua-adduct by adding excess water to the reaction mixture,
and its reaction with alkynes other than DMAD results in new
IrIII coupling products with different structural motifs.[14] We did
not observe the actual product of alkyne cyclotrimerization or
on the reactivity of complex [TpRh(C2H4)2], 1 (Tp=hydrotris-
(pyrazolyl)borate), towards DMAD. Complex 1 has already
been applied as an effective catalyst in the regioselective addi-
tion of amines to alkynes[15] and in the substitution of allylic
carbonates by organolithium reagents.[16] Mechanistic studies
on phosphine-ethylene [TpRh(PR3)(C2H4)] complexes and their
reactivity towards hydrogen have also been performed.[17] In
comparison with its TpMe analogue, [TpMe Rh(C2H4)2], which is
only stable below À108C,[18] complex 1 can be handled in air
and at room temperature.
the cyclohexadiene in any of these reactions, probably due to
III
the reluctance of these TpMe Ir products to experience reduc-
2
tive elimination. Only in a related benzannulated system, and
under very forcing oxidative conditions, we observed this pro-
cess with the generation of the substituted naphthalene
shown in Scheme 3.[12]
2
2
Due to the higher tendency of Tp’RhIII (Tp’=any type of hy-
drotris(pyrazolyl)borate ligand) derivatives to undergo reduc-
tion to a RhI species, the catalytic cycle shown in Scheme 1
could be more feasible with this metal. Thus, we now report
Results and Discussion
The synthesis of the known compound [TpRh(C2H4)2][19] (1) was
improved to excellent yields (ꢀ95%) by adapting the proce-
dure reported for the synthesis of the related [TpMe Rh(C2H4)2]
2
(see the Experimental Section).[18] The reaction of 1 with
3 equiv of DMAD (C6H6, 258C, 4 h) gave a mixture of rhodium
complexes 2–5 and C6(CO2Me)6 (6), that is, the already known
product resulting from the [2+2+2] cyclotrimerization of the
alkyne, being 2 and 3 the more abundant organometallic spe-
cies of this complex mixture (Scheme 4). Under these room
temperature conditions, the mixture is under kinetic control
and there was no change in the final composition of the mix-
ture upon increasing the reaction time; this is also the case
when a larger amount (8 equiv) of DMAD was used. By con-
trast, heating the reaction mixture overnight (C6H6, 608C, 3–
8 equiv DMAD) resulted in a prevailing yield of 3, which repre-
sented more than 75% of the crude product, with only
a minor presence of 4, 5, and 6, and 2 being undetectable.
Nevertheless, a small amount of the free diene present in 2 as
a ligand was also formed, indicating the transient formation of
2 and the subsequent evolution under these reaction condi-
tions (see below).
Complexes 2, 3, and 4 are RhI species that all have h4-diene
ligands, either open or cyclic, which differ in the number of
DMAD incorporated into them. Thus, 2 is the result of the
formal alternate coupling of the two ethylene ligands present
in the starting material and one molecule of DMAD. The vinyl
termini and the ethyl groups of the resulting diene are gener-
ated by a formal H-shift from one to the other. In turn, the
cyclic diene of complex 3 forms through the simple coupling
of one ethylene and two DMADs. Finally, in compound 4,
a molecule of C6(CO2Me)6 (6), resulting from the trimerization
of the alkyne, is h4-bonded to the metal center. In contrast, the
remaining Rh complex in this reaction, product 5, is an allyl
RhIII species akin to the IrIII complex depicted in Scheme 2.
4
Scheme 2. Reactions of DMAD with (a) [TpMe Ir(h -CH2=C(Me)C(Me)=CH2)]
2
and (b) [ TpMe Ir(C2H4)2].
2
Chem. Eur. J. 2016, 22, 13715 –13723
www.chemeurj.org
13716
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 4. Reaction of [TpRh(C2H4)2] with 3 equiv of DMAD (C6H6, 258C, 4 h);
isolated yields of the organometallic species are reported in parentheses.
Whereas this type of allyl product was the only species formed
for the Ir case, the allyl Rh product constitutes a very minor
part of the mixture.
All of these organometallic products are air-stable and can
be easily separated by column chromatography or crystalliza-
tion (see the Experimental Section). They are also highly ther-
mally stable, although 2 and 4 react with excess alkyne (see
Figure 1. ORTEP diagram of compound 4; 50% thermal ellipsoids are
shown; hydrogen atoms have been omitted for clarity.
1
below). In the H NMR of compound 2, the -CH=CH2 coordinat-
ed termini of the h4-diene ligand is characterized by signals at
d=5.28 (t), 2.27 (dd), and 2.07 ppm (m), and the chemical
shifts for the diene carbon nuclei range from d=95.4 to
32.5 ppm. Finally, the ethyl substituent of the diene shows the
expected signals for
a diastereotopic CH2 (at 2.07 and
1.09 ppm, 2JHH =13.6 Hz). The NMR data of 3 exhibit the ex-
pected 2:1 equivalence for the three pyrazolyl (pz) rings of the
Tp ligand, according to the symmetry of the molecule. In addi-
tion, the -CH2CH2- unit is responsible for a set of resonances
1
corresponding to an AA’XX’ spin system in the H NMR spec-
trum (with d=2.46 and 0.99 ppm) and only one signal in the
13C{1H} NMR spectrum (23.0 ppm, 2JCRh =4 Hz). In spite of the
symmetry of the cyclic ligand formed in 4, the ligand coordina-
tion through only two double bonds is responsible for the 2:1
1
equivalence of the pz rings of the Tp ligand in both H and
13C{1H} NMR spectra. Accordingly, three types of OMe groups
are observed (see the Experimental Section for details). Com-
pound 5 was easily characterized by comparison with the pre-
viously reported spectroscopic data of the Ir analogue depict-
ed in Scheme 2.[13]
Figure 2. ORTEP diagram of compound 5; 50% thermal ellipsoids are
shown; most hydrogen atoms have been omitted for clarity.
Compounds 3, 4, and 5 gave suitable single crystals for X-
ray analysis (Figure S1 in the Supporting Information, Figures 1,
and 2, respectively), which confirmed their h4-cyclohexadiene-
rhodium (3, 4) and chelating (h3-allyl)(h1-allyl)-rhodium (5)
structures. Compounds 3 and 4 are structurally similar, with
RhÀC(‘diene’) bond lengths of around 2.12 Š in both cases,
and CÀC distances of around 1.42 (internal diene CÀC bond)
and 1.46 Š (terminal diene CÀC bonds). The organic hexacyclic
fragments of both species are equally bent with an angle of
lengths of 1.410(3) and 1.427(4) Š for the allylic carbons,
whereas the corresponding RhÀC bond distances are around
2.09 Š, only slightly shorter than in 3 and 4, and more similar
to the RhÀC(h1-allyl) distance of 2.106(2) Š.
To gain insight into the mechanism of formation of the dif-
ferent rhodium species, we carried out the reaction of 1 with
1
1 equiv of DMAD in C6D6 at 108C and recorded a H NMR spec-
trum at the initial time. Compound 1 reacted instantaneously
with DMAD and, other than small signals from the appearance
about 1388. The single versus double bond character of the CÀ of compound 2, a new major species was observed; its struc-
C fragment, arising from ethylene (3) or the third DMAD mole-
cule (4), resulted in CÀC distances of 1.533(3) and 1.335(2) Š,
respectively. The h3-allyl moiety in 5 demonstrates CÀC bond
ture (compound 7 in Scheme 5) was assigned according to the
characteristic NMR pattern. In the absence of planes of symme-
try, the pyrazolyl protons once again give distinct signals in
Chem. Eur. J. 2016, 22, 13715 –13723
www.chemeurj.org
13717
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
the H NMR spectrum. The ethylene resonance of 7 appears as
more quickly than 7 even at 108C, while the transformation of
7 into 2 is complete in about 2 h. The ratio between 7 and 8
changed slightly depending on the concentration of reactants
and temperature but no conversion of 7 into 8 was observed
by adding another equivalent of DMAD to the mixture of both
compounds, indicating that reversible loss of ethylene is not
occurring between the two RhIII metallacycle species. Thus, the
data discussed so far are in agreement with the high reactivity
and facile activation of the residual ethylene molecule in the
RhIII metallacycle species 7 and 8 and their easy transformation
to the RhI complexes 2 and 3, which must be thermodynami-
cally favored processes. An example of a stable metallacyclo-
pentadiene-ethylene was reported by O’Connor and coworkers
in the reaction of TpIr(C2H4)2 with DMAD:[21] the addition of the
first equivalent of alkyne in THF leads to a solvated iridacyclo-
pentene complex (similar to 7) that converts into an iridacyclo-
pentadiene-ethylene species (similar to 8) by addition of
a second equivalent of DMAD in benzene. Moreover, by heat-
ing the latter compound up to 1008C, no formation of an h4-
cyclohexadiene-iridium species (similar to 3) is observed, nor
are other transformations, according to the normally higher
stability of the TpIrIII when compared to the TpIrI species. In ad-
dition, this behavior also denotes a very high stability of the
TpIr-ethylene species in comparison with the corresponding
an AA ’BB ’spin system at d=3.98 ppm (d=80.4 ppm in the
13C NMR spectrum) due to fast rotation around the coordina-
tion axis.[20] The signals for the two methoxy groups are locat-
ed at d=3.47 and 3.21 ppm (d=50.8 and 50.9 ppm in the
13C NMR spectrum), and the -CH2CH2- protons of the metallacy-
cle generate three multiplets at d=3.13, 2.85, and 2.23 ppm in
a 2:1:1 integration ratio (d=36.5 and 23.1 ppm for the b and
a carbons to the rhodium, respectively). The species 7, result-
ing from the oxidative coupling of one ethylene and one
alkyne molecule, disappeared in about 2 h at room tempera-
ture to give compound 2. The insertion of ethylene may take
place into either the RhÀCsp or the RhÀCsp bond, but the
2
3
former case (Scheme 5), followed by b-H elimination and sub-
sequent H-migration to the RhÀCH2 end of the organic chain,
provides a mechanistic explanation for the formation of com-
pound 2. The full regioselectivity of the insertion process may
be favored by the formation of a less congested metallacycle
with the carbomethoxy groups located far away from the
metal environment.
Rh complex or the TpMe Ir (see Introduction), which rapidly un-
2
dergoes olefin insertion into the metallacycles even at 108C.
With the aim of obtaining more stable metallacycle species,
we carried out the reaction of 1 with 1 equiv of DMAD in THF/
H2O and in CH3CN, obtaining the corresponding aqua- and
acetonitrile-metallacyclopentadiene adducts 9 and 10, respec-
tively (Scheme 6). The results are the same in the presence of
2 equiv of DMAD and NCMe, indicating that trapping of a RhIII
species by NCMe is very effective and the initially formed spe-
cies, the otherwise non-isolable metallacycle of 7, is trapped
before the second molecule of DMAD can enter the com-
pound.
These adducts 9 and 10 were obtained in high yields and
do not react with a second equivalent of alkyne at room tem-
perature, whereas reaction with an excess of DMAD (2–
3 equiv) at 608C in benzene solution slowly converts the ad-
ducts into a mixture of 3 and 5, with 3 being the major prod-
Scheme 5. Formation of 2 and 3 through intermediates 7 and 8.
To verify whether 7 is also a precursor for other rhodium
species containing more than one alkyne or whether other in-
termediates are involved in the mechanism, the addition of
2 equiv of DMAD to 1 in C6D6 was studied and monitored by
1H NMR spectroscopy at 108C using 1,1,1,3,3,3-hexamethyldisi-
lazane as an internal standard. At the initial reaction time,
a mixture of 7 and another new TpRh species were formed,
the latter being identified by NMR as the metallacycle 8
(Scheme 5), with its peculiar symmetry giving rise to character-
1
istic signals in the H NMR spectrum. The ethylene ligand of 8
resonates at d=4.33 ppm as a singlet (d=85.6 ppm in the
13C NMR spectrum), which corresponds to four equivalent pro-
tons from rapid rotation around the coordination axis.[20] The
methoxy groups generate two singlets at d=3.48 and
3.14 ppm (d=51.4 and 51.1 ppm in the 13C NMR spectrum).
The initial ratio between 7 and 8 was 1:4, and they undergo
conversion into 2 and 3, respectively, however 8 converts
Scheme 6. Synthesis of the rhodacyclopentene adducts 9 and 10 and their
reaction with DMAD to give a mixture of 3 and 5; isolated yields in paren-
thesis.
Chem. Eur. J. 2016, 22, 13715 –13723
www.chemeurj.org
13718
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
uct. The water adduct is more reactive and evolves into the
mixture of 3 and 5 faster. Yields of 3 are higher (ratio 3:5=
85:15) than in the case of the NCMe adduct 10, which is more
stable and requires harsher conditions to react with DMAD
(longer reaction times or higher temperatures), and in compar-
ison, yields a mixture with greater amounts of 5 (ratio 3:5=
65:35).
tion pathways (reductive elimination E!3 and b-H elimination
E!5) could operate in competition, and the relative stabilities
and energy barriers for 3 and 5 would determine the ratio of
the two types of compounds in each case. Although both 3
and 5 are very stable, 5 seems to require stronger reaction
conditions to form (it is not observed in the reaction carried
out at 108C, Scheme 5, and is formed in higher yields from 10,
which requires stronger conditions to evolve, than from 9,
which reacts faster with DMAD).
It can be proposed that formation of compounds 3 and 5
from metallacycles 9 or 10 requires the loss of the correspond-
ing Lewis base and the subsequent coordination and insertion
Finally, with the data available we cannot propose a detailed
mechanism for the formation of 3 from either E or 8. The
latter could generate E, as mentioned above, or alternatively
the transformation 8!3 could take place by a direct [4+2] cy-
cloaddition through a metallabicycle of a metallanorbornene
type (Figure 3).[22] [4+2] pathways have been proposed not
only for the cycloaddition of alkynes to alkenes, with formation
of metallanorbornenes, but also for cyclotrimerization of al-
kynes. For instance, an interesting bridging [4+2] tantallanor-
bornadiene formed during the formal cyclotrimerization of 3-
hexyne has been recently described,[3a] and also a recent theo-
retical work proposes the participation of this type of [4+2] in-
of the second alkyne molecule into the metallacycle RhÀCsp
2
bond to give the rhodium-cycloheptadiene intermediate E
(Scheme 7), a type of intermediate commonly proposed for
[2+2+2] cycloadditions of alkynes to alkenes.[2,3] Attempts to
trap this seven-membered metallacycle failed because, as men-
tioned, reaction of 1 with 2 equiv of DMAD in CH3CN (RT) led
to compound 10, with the excess of DMAD remaining unreact-
ed. Heating 10 in benzene with an excess of DMAD (ca.
5 equiv) in the presence of NCMe (ca. 6 equiv), also led directly
to the final compounds 3 and 5. Hence, it is reasonable to
admit that, in benzene at 608C, this purported unsaturated
RhIII metallacycle intermediate E is short-lived and rapidly con-
verts into the corresponding (h4-diene)-RhI and (h3-allyl)(h1-
allyl)-RhIII compounds, 3 and 5, respectively, before being
trapped by the Lewis base.
termediate in TpIr-mediated cyclotrimerization of alkynes.[23]
A
further possibility for the generation of 3 from E would be the
formation of a linear hexatriene with
subsequent thermal electrocyclization,[6]
but this pathway seems unlikely in view
of the reluctance of the organic frag-
ment to be displaced from the rhodium
center (see below). In fact, in situ NMR
monitoring of the evolution of 9 and
10 (Scheme 7) has never shown the for-
mation of linear compounds released
by the metallacycle E.
Due to its open-chain diene-like fea-
tures, compound 2 is expected to still
be reactive towards an excess of
alkyne. Although it did not react with
Figure 3. Possible rho-
danorbornene inter-
mediate formed by
a direct [4+2] cycload-
dition in the formation
of 3 from 8.
DMAD at room temperature, heating a mixture of 2 and
6 equiv of DMAD in benzene at 658C resulted in dissociation
of the organic diene 11, along with formation of a mixture
(Scheme 8). This mixture consisted of compound 4, the cyclo-
trimerization product 6, and new rhodium(III) species 12, aris-
ing from the coupling and activation of the diene ligand of 2
with two additional molecules of alkyne (see Scheme 9 below).
Compound 4, which was a minor species in the reaction at
room temperature (Scheme 4), was the main product in the re-
action of Scheme 8. It is poorly soluble in benzene and was
formed as orange crystals by carrying out the reaction in an
NMR tube without stirring. The soluble products were separat-
ed by column chromatography. Compound 12 gave yellow
cubic crystals by slow evaporation from a hexane/diethyl ether
solution and was fully characterized by NMR and X-ray analy-
ses (Figure 4). 12 can be considered analogous to 5, where the
end of the h3-allyl ligand derives from the diene present in 2
instead of ethylene. In fact, 12 shows a geometrical pattern
around the rhodium center comparable to that one found in
5, with bond lengths and angles contained in the (h3-allyl)(h1-
Scheme 7. Proposed mechanism for the formation of 3 and 5 from 9 and
10.
Reductive elimination of intermediate E would yield com-
pound 3, and this may also be the pathway for its generation
from 8 in Scheme 5 (insertion of ethylene into the RhÀC bonds
of 8 would give rise to E). In addition, and in view of the reac-
tivity shown in Scheme 2 for the related aqua-adduct iridacy-
cloheptadiene (being quantitatively transformed into an allyl
derivative), we propose E as precursor of 5 through a sequence
of b-H elimination and H-migration to the alkenyl end of the
carbon chain. Hence, we postulate that these two decomposi-
Chem. Eur. J. 2016, 22, 13715 –13723
www.chemeurj.org
13719
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 8. Reaction of 2 with excess DMAD; isolated yields are shown for
rhodium species.
Scheme 9. Pathways proposed for the formation of compounds 4, 6, 11,
and 12 from 2 and excess DMAD at 658C.
the rhodium complex, as evidenced by the disappearance of
1
pyrazolyl signals in the H NMR spectrum.
In view of the formation of compound 3 in this system,
which contains the cyclohexadiene derived from the coupling
of two DMAD and one ethylene, we also investigated the cata-
lytic activity of 4 towards DMAD under the same conditions
mentioned but under an atmosphere of ethylene. The reaction
took place with quantitative transformation into 3, without no-
ticeable formation of products deriving from the catalytic tri-
merization or cycloaddition of the unsaturated substrates
(DMAD and ethylene). This is in agreement with the high ther-
mal and kinetic stability observed for 3, which, once formed, is
stable towards displacement of the cyclohexadiene fragment,
even upon heating the reaction mixture in the presence of
a large excess of reactive substrates (DMAD or ethylene). It
also indicates that ethylene competes favorably with DMAD
for the coordination to the metal center of the rhodacyclopen-
tadiene formed by the oxidative coupling of two DMAD mole-
cules. A similar reaction took place by heating compounds
1 or 2 and DMAD (60 equiv) under ethylene atmosphere and
the conditions of the catalytic reaction (858C in toluene, 24 h).
Compound 3 was formed in both cases as the only organome-
tallic species, whereas the reactions carried out under nitrogen
(Scheme 4 and 8, respectively) formed compounds 2, 3, 4, and
5 (for the case of 1), or 4 and 12 (for the case of 2). Again in
these reactions, we never observed the formation of the ex-
pected [2+2+2] cyclization products.[25]
Figure 4. ORTEP diagram of compound 12 at 50% thermal ellipsoids; most
hydrogen atoms have been omitted for clarity.
allyl)-RhIII moiety almost identical to those in 5. Moreover, no
particular geometrical distortion due to the pendant ethyl-
DMAD fragment is observed, which indicates that there is no
significant steric stress with respect to 5 despite the bulkiness
of the ethyl-DMAD termini.
The NMR data for the identical parts of the skeletons of 5
and 12 (see the Experimental Section) are also very similar in
both cases. Thus, by analogy with the formation of 5, we pro-
pose for 12 the generation of a transient unsaturated h2-olefin
species from 2, which would permit the incorporation of two
more molecules of alkyne. For simplicity, we propose the for-
mation (Scheme 9) of a rhodium metallacyclopentadiene inter-
mediate F, which could be a common intermediate for the for-
mation of 4 and 12. F might undergo insertion of the h2-coor-
dinated diene 11 and then convert into compound 12 after
two steps, or release the diene to coordinate more alkyne and
form 4 and 6.
It is noteworthy that compound 4 has a high thermal stabili-
ty and, consequently, is a poorly effective [2+2+2] cyclotrime-
rization catalyst. Reaction of 60 equiv of DMAD at 858C in tolu-
ene with 4 (2 mol%) gave a yield of only 40% (11 TONs) for
the desired hexamethyl mellitate, 6. Higher temperatures (re-
fluxing toluene) led to even lower yields and decomposition of
Conclusions
In sharp contrast to the reactivity of the parent TpMe Ir system,
2
the bis(ethylene) complex TpRh(C2H4)2 reacts with DMAD yield-
Chem. Eur. J. 2016, 22, 13715 –13723
www.chemeurj.org
13720
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ing a manifold of RhI diene complexes and only a small
amount of an allyl RhIII species, which finds precedent in the
d=139.2, 134.5, 104.8 (CHpz), 48.8 ppm (d, J=13 Hz, CH2ethylene); el-
emental analysis calcd (%) for C13H18BN6Rh: C 42.0; H 4.9; N 22.6;
found C 41.8; H 4.5; N 22.6.
related TpMe Ir chemistry. The observed air- and moisture-
2
stable rhodium complexes originate from the facile activation
of at least one ethylene ligand, which competes with DMAD in
the oxidative coupling process to the rhodium center. In addi-
tion, NMR studies performed at low temperature revealed the
formation of transient rhodacycle species (rhodacyclopentene
and rhodacyclopentadiene), and in the presence of appropriate
trapping agents, the corresponding rhodacyclopentene ad-
ducts were isolated and characterized. The allyl-RhIII species 5
and the h4-cyclohexadiene-RhI 3 are thermodynamic sinks, and
display a high stability towards excess of alkyne. In contrast,
the h4-diene-RhI derivative that features an open-chain diene
(compound 2), still reacts with additional alkyne at higher tem-
peratures, forming an unusually stable rhodium complex (4)
with the alkyne-cyclotrimerization product h4-bound to the
metal center. Compound 4 can release the cyclotrimerization
product upon heating, although only moderate catalytic activi-
ty in the [2+2+2] cyclotrimerization reaction was observed. In
contrast, as mentioned, product 3 containing an h4-bonded
metallacyclohexadiene, a model for an intermediate in the
[2+2+2] cocyclotrimerization of alkynes and alkenes, is inert in
the catalytic reaction. We are carrying out additional studies in
our laboratory with different alkynes to address some unre-
solved mechanistic issues and explore other substrates that
can lead to catalytic activity.
Reaction with DMAD: To a solution of 1 (200 mg, 0.54 mmol) in
benzene (10 mL), neat DMAD (0.196 mL, 1.60 mmol) was added
dropwise and the mixture was stirred for 4 h at RT. Then, the sol-
vent was evaporated off and the residual orange solid was purified
by column chromatography on silica gel (5:1!1:2, hexane/Et2O) to
yield 2 (17%), 3 (28%), 4 (4%), and 5 (7%).
Complex 2: Yield: 46 mg (17%). 1H NMR (CDCl3): d=
7.98,7.76,7.72,7.54,7.50,7.22,6.31,6.05,6.03 (9s, 1H each, 9CHpz),
3
5.28 (t, JCA,CB ꢁ7 Hz, Hc), 3.90 (s, 3H, CO2Me), 3.18 (s, 3H, CO2Me),
3
2
2.27 (dd, JBC ꢁ7 Hz, JBA ꢁ4 Hz, Hb), 2.07 (m, 2H, Ha and CHHCH3),
2
3
1.09 (dq, JHH ꢁ13.6 Hz, JHH ꢁ7 Hz, 1H, CHHCH3), 0.99 ppm (t, 3H,
CH2CH3); 13C{1H} NMR (CDCl3): d=175.4, 170.0 (s, CO2Me), 142.6,
142.2, 139.8, 135.2, 134.5, 134.4, 105.3, 105.3, 104.3 (s, CHpz), 95.4
1
1
1
(d, JCRh =6 Hz, C3), 89.3 (d, JCH =167 Hz, JCRh =6 Hz, C2), 52.7, 51.3
(s, 1JCH =147 Hz, CO2Me), 42.5 (d, 1JCRh =16 Hz, C4), 32.5 (d, JCH
=
1
152 Hz, 1JCRh =16 Hz, C1), 21.0 (s, 1JCH =130 Hz, CH2CH3), 13.9 ppm
1
(s, JCH =122 Hz, CH2CH3); IR (nujol): n(C=O)=1739, 1700 cmÀ1; ele-
mental analysis calcd (%) for C19H24BN6O4Rh: C 44.4; H 4.7; N,16.4;
found: C 44.7; H 5.1; N 16.0.
Experimental Section
General methods: Manipulations were performed either in air or
under oxygen-free dinitrogen atmosphere, by means of conven-
tional Schlenk techniques. Solvents were freshly distilled prior to
use over sodium/benzophenone or CaH2 according to standard
procedures. Microanalyses were conducted by the Microanalytical
Service of the Instituto de Investigaciones Químicas (Sevilla). HRMS
data were obtained on a JEOL JMS-SX 102A mass spectrometer by
the Mass Spectrometry Services of the University of Seville (CITIUS).
Infrared spectra were obtained by using PerkinElmer spectrome-
ters, models 577 and 684. NMR characterization was carried out by
Bruker DRX-500, DRX-400, and DPX-300 spectrometers. Spectral as-
signments were performed by means of routine one- and two-di-
mensional NMR experiments where appropriate. The complex
TpRh(C2H4)2, 1, was synthesized according to a slightly modified
procedure developed by our research group[24] for the synthesis of
Complex 3: Yield: 95 mg (28%). 1H NMR (C6D6): d=8.30 (s, 1H,
3
CHpz), 7.96 (s, 1H, CHpz), 7.41 (d, JHH ꢁ1.6 Hz, 1H, CHpz), 7.35 (d,
3JHH ꢁ1.6 Hz, 1H, CHpz), 6.06 (s, 1H, CHpz), 5.86 (s, 1H, CHpz), 3.65 (s,
6H, 4CO2Me), 2.90 (s, 6H, 4CO2Me), 2.46, 0.99 ppm (AA’XX’ spin
2
system, 2H each, JHH ꢁ10.3 Hz, 2CH2); 13C{1H} NMR (C6D6): d=
172.6, 166.2 (s, CO2Me), 144.4,142.1,134.9,134.8,105.7,104.7 (s,
1
2:1:1:2:1:2, 9CHpz), 93.8 (d, JCRh =7.4 Hz, C2), 52.3, 51.3 (s, CO2Me),
1
45.4 (d, 1JCRh =15.6 Hz, C1), 23.2 ppm (d, JCH ꢁ132 Hz, 1JCRh =4 Hz,
C3); IR (nujol): n(CO)=1752, 1715 cmÀ1; elemental analysis calcd
(%) for C23H26BN6O8Rh: C 44.0; H 4.2; N 13.4; found: C 43.9; H 4.1;
N 13.3.
TpMe Rh(C2H4)2, which led to an improved yield and shorter reac-
2
tion time with respect to the reported procedure.[19] DMAD was
purchased from Sigma Aldrich and used without any further purifi-
cation.
Complex 4: Yield: 16 mg (4%). 1H NMR (400 MHz, CDCl3): d=
3
8.61,7.82,7.71,7.52,6.45,6.06 (d, d, d, d, t, t, 1:1:2:2:1:2, JHH ꢁ2.3 Hz
Complex 1: To a suspension of [Rh(C2H4)2Cl]2 (500 mg, 1.28 mmol)
in THF (30 mL) at 08C, KTp salt (680 mg, 2.70 mmol) was added as
a solid. The reaction mixture was warmed to room temperature
and left stirring for 1 h. The solvent was evaporated off under
vacuum and the dark orange solid was dissolved in a 1:1 mixture
of dichloromethane/diethyl ether and filtrated over celite. Volatiles
were evaporated under reduced pressure to give a bright yellow
each, 9CHpz), 4.05 (s, 6H, 6CO2Me), 3.77 (s, 6H, 6CO2Me), 3.27 ppm
(s, 6H, 6CO2Me); 13C NMR (101 MHz, CDCl3): d=170.0,165.4,164.8
(2:2:2, CO2Me), 143.3,140.7,135.2,135.1,106.7,106.2 (2:1:1:2:1:2,
CHpz), 139.5 (d, JCRh =2 Hz, CCO2Me), 97.1 (d, JCRh =5 Hz, CCO2Me),
49.1 (d, JCRh =14 Hz, CCO2Me), 53.7, 52.4, 52.1 ppm (CO2Me); ele-
mental analysis calcd (%) for C27H28BN6O12Rh: C 43.7; H 3.8; N 11.3;
found
C 43.6; H 4.3; N 10.8; HRMS (FAB): m/z calcd for
1
solid in excellent yield (910 mg, 95%). H NMR (400 MHz, C6D6): d=
C27H28BN6O12NaRh 765.0811; found 765.0779 [M+Na]+.
7.60 (d, J=1.9 Hz, 3H, Hpz), 7.41 (d, J=1.9 Hz, 3H, Hpz), 5.91 (t, J=
1.9 Hz, 3Hpz), 2.52 ppm (s, 8H, CHethylene); 13C NMR (101 MHz, C6D6):
Chem. Eur. J. 2016, 22, 13715 –13723
www.chemeurj.org
13721
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
(d, JCRh =3 Hz, CO2Me), 174.4 (d, JCRh =31 Hz, RhCq), 141.7 (d JCRh
=
2.5 Hz, CHpz), 140.4,139.6,135.2,134.6,134.4,105.2,105.1,104.6 (8s,
CHpz), 138.2 (CqCH2), 119.47 (d, JCRh =8.2 Hz, CNMe), 51.1, 50.6
(CO2Me), 36.6 (RhCH2CH2), 20.4 (d, JCRh =21 Hz, RhCH2CH2), 3.9 ppm
(NCMe); HRMS (FAB): m/z calcd for C19H23BN7O4Rh 527.0960; found
527.0947.
Reaction of 2 with excess of DMAD: Compound 2 (60 mg,
0.12 mmol) was dissolved in 4 mL of benzene and DMAD
(0.090 mL, 0.72 mmol) was added. The resulting yellow solution
was heated at 608C overnight, after which time an orange solution
was formed. The solution was concentrated to about 2 mL leading
to the precipitation of an orange solid. The supernatant solution
was separated through a cannula and the solvent was evaporated.
The orange solid firstly separated was washed with hexane and
dried under vacuum and was found to be compound 4 in good
purity (42 mg, 48%). The solution was purified by column chroma-
tography on silica gel (hexane/diethyl ether) to yield 11 (59%) and
12 (25%).
Complex 5: Yield: 23 mg (7%). 1H NMR (400 MHz, CDCl3): d=
7.80,7.71,7.70,7.67,7.59,7.49,6.31,6.23,6.08 (d, d, d, d, d, d, t, t, t,
3
3
1H each, JHH ꢁ2.3 Hz each, 9CHpz), 6.38 (dd, 1H, JHH =12.6, 8.0 Hz,
Hc), 4.19 (d, JRhH =3.4 Hz, 1H, Hd), 3.77 (s, 3H, 4CO2Me), 3.70 (s, 3H,
4CO2Me), 3.49 (s, 3H, 4CO2Me), 3.28 (s, 3H, 4CO2Me), 3.69 (d, 1H,
3JHH =8.0 Hz, Hb), 3.59 ppm (d, 1H, 3JHH =12.5 Hz, Ha); 13C NMR
(101 MHz, CDCl3): d=180.1, 171.5, 165.7, 165.0 (CO2Me), 147.1 (s,
C4), 143.9, 142.3, 141.5, 135.5, 135.5, 135.2, 106.6, 105.6, 104.9
(CHpz), 133.9 (s, C5), 107.4 (d, JCRh =7 Hz, C2), 65.7 (d, JCRh =12 Hz,
C3), 49.7 (d,
33.8 ppm (d,
J
CRh =10 Hz, C1), 52.4, 52.2, 52.0, 50.8 (CO2Me),
CRh =22 Hz, C6); HRMS (FAB): m/z calcd for
J
C23H26BN6O8NaRh 651.0858, found 651.0853 [M+Na]+; elemental
analysis calcd (%) for C23H26BN6O8Rh: C 44.0; H 4.2; N 13.4; found:
C 44.3; H 4.2; N 13.0.
Intermediates 7 and 8: To a solution of 1 (20 mg, 0.054 mmol) in
0.5 mL of C6D6 at 108C, 1 or 2 equiv of DMAD were added to in-
stantaneously give 7 and 8 as major species. No elemental analysis
1
or HRMS data could be obtained. Compound 7: H NMR (400 MHz,
1
Compound 11: Yield: 14 mg (59%). H NMR (300 MHz, CDCl3): d=
C6D6): d=7.81,7.69,7.42,7.33,7.23,6.97,5.98,5.84,5.81 (d, d, d, d, d,
d, t, t, t, 1H each, JHH =2.3 Hz each, 9CHpz), 3.99 (AA’XX’ spin
system, 4H, CH2 =CH2), 3.48 (s, 6H, 2CO2Me), 3.21 (s, 6H, 2CO2Me),
3.11–2.21 ppm (m, 4H, RhCH2CH2); 13C NMR (101 MHz, C6D6): d=
6.65 (dd, 1H, 3JHH =17.5, 10.9 Hz, Ha), 5.54 (d, 1H, 3JHH =10.9 Hz,
Hb), 5.48 (d, 1H, JHH =17.4 Hz, Hc), 3.85 (s, 3H, CO2Me), 3.77 (s, 3H,
3
3
3
CO2Me), 2.48 (q, 2H, JHH =7.5 Hz, CH2CH3), 1.09 ppm (t, 3H, JHH
=
7.5 Hz, CH2CH3); 13C NMR (101 MHz, CDCl3): d=168.8,167.4 (2
CO2Me), 140.2 (CqCH2), 133.6 (CqCHa), 129.4 (CH=CH2), 123.0 (CH=
CH2), 52.3 (2 CO2Me), 21.3 (CH2CH3), 13.5 ppm (CH2CH3); HRMS
(FAB): m/z calcd for C10H14O4Na 221.0790; found 221.0789
[M+Na]+.
173.6,172.2 (2 CO2Me), 170.5 (d,
140.4,140.3,139.7,135.1,134.8,134.7,105.7,105.5,105.1
J
RhC =28 Hz, RhCq),
(9CHpz),
135.3 (CqCH2), 80.4 (d, JRhC =7 Hz, CH2 =CH2), 36.2 (s, RhCH2CH2),
22.8 (d, JRhC =19 Hz, RhCH2CH2).
Compound
8:
1H NMR
(400 MHz,
C6D6):
1
Compound 12: Yield: 24 mg (25%). H NMR (400 MHz, CDCl3): d=
8.01,7.60,7.27,7.06,5.84,5.66 (6s, 1:2:2:1:2:1, 9CHpz), 4.32 (s, 4H,
7.92,7.71,7.66,7.60,7.45,7.37,6.26,6.26,6.08 (d, d, d, d, d, d, t, t, t,
CH2 =CH2), 3.48 (s, 6H, 4 CO2Me), 3.14 ppm (s, 6H, 4 CO2Me).
1H each, JHH ꢁ2.2 Hz each, 9CHpz), 6.89 (d, 1H, 3JHH =11.8 Hz,
3
13C NMR (101 MHz, C6D6): d=163.0, 163.8 (CO2Me), 156.8 (d, JRhC
=
3
CHallyl), 4.59 (d, 1H, JHH =11.8 Hz, CHallyl), 4.34 (d, 1H, JHRh =3.3 Hz,
41 Hz, RhCq), 141.3,140.8,135.6,135.3,105.5,104.8 (1:2:1:2:2:1,
RhCH), 3.78,3.73,3.66,3.49,3.23,2.50 (6s, 3H each, 6CO2Me), 3.19–
2.53 (m, 2H, CH2CH3), 1.08 ppm (t, 3H, 3JHH =7.4 Hz, CH2CH3);
13C NMR (101 MHz, CDCl3): d=179.9, 171.2, 168.0, 167.6, 165.6,
164.6 (CO2Me), 147.8 (Cq), 144.3, 143.1, 142.5, 135.3, 135.1, 135.1,
106.7, 105.7, 104.3 (CHpz), 139.3 (Cq), 137.5 (Cq), 133.9 (Cq), 102.7 (d,
CRh =6, 8 Hz, CHallyl), 60.5 (d, JCRh =6, 8 Hz, CHallyl), 62.6 (d, JCRh
13 Hz, Cq), 52.5, 52.3, 52.1, 52.0, 50.9, 50.8 (CO2Me), 34.3 (d, JCRh
21 Hz, RhCH), 21.4 (CH2CH3), 12.7 ppm (CH2CH3); HRMS (FAB): m/z
calcd for C31H36BN6O12NaRh 821.1437; found 821.1423 [M+Na]+.
CHpz), 122.7 (CqCO2Me), 85.4 (d,
50.8 ppm (CO2Me).
JRhC =6 Hz, CH2 =CH2), 51.2,
Synthesis of metallacycles 9 and 10: To a solution of 1 (80 mg,
0.22 mmol) in THF (4.5 mL+0.5 mL H2O) or CH3CN (5 mL), neat
DMAD (0.027 mL, 0.22 mmol) was added dropwise and the mixture
was stirred for 1 h at RT. The respective solvent was evaporated off
in vacuo and the resulting yellow solid was washed several times
with pentane to give compounds 9 and 10 in high yields. Com-
pound 9: Yield: 82 mg (73%). 1H NMR (400 MHz, CDCl3): d=
7.79,7.72,7.70,7.61,7.59,7.27,6.24,6.08 (d, d, d, d, d, d, t, t,
J
=
=
3
1:1:1:1:1:1:2:1, JHH ꢁ2.3 Hz each, 9CHpz), 3.99 (bs, 2H, H2O), 3.67 (s,
Acknowledgements
3H, CO2Me), 3.36 (s, 3H, CO2Me), 3.03–2.83 ppm (m, 4H,
RhCH2CH2); 13C NMR (101 MHz, CDCl3): d=179.5, 163.0 (CO2Me),
Financial support (FEDER contribution) from the Spanish Minis-
terio de Economía y Competitividad (grants CSD2007-0006 and
CTQ2014-51912-REDC) and the Junta de Andalucía (Grant
FQM-119) is acknowledged.
177.5 (d,
JCRh =32 Hz, RhCq), 142.8,140.2,140.0,135.9,134.8,
105.6,105.5,104.8 (1:1:1:1:2:1:1, CHpz), 139.7 (CqCH2), 51.6, 51.4
(CO2Me), 36.2 (RhCH2CH2), 20.4 ppm (d, JRhC =23 Hz, RhCH2CH2); ele-
mental analysis calcd (%) for C17H22BN6O5Rh: C 40.5; H 4.4; N 16.7;
found: C 40.9; H 4.3; N 16.9.
1
Compound 10: Yield: 88 mg (79%). H NMR (400 MHz, CDCl3): d=
Keywords: alkynes · cycloadditions · cyclotrimerization
rhodium · scorpionate ligands
·
3
7.75 (d, 1H, JHH ꢁ2.2 Hz, CHpz), 7.65–7.63 (m, 2H, 2CHpz),
3
7.56,7.54,7.45,6.19,6.09 (d, d, d, t, t, 1:1:1:2:1, JHH ꢁ2.2 Hz each,
6CHpz), 3.69 (s, 3H, CO2Me), 3.36 (s, 3H, CO2Me), 2.98–2.77 (m, 3H,
RhCH2CH2), 2.64 (dd, 1H, JHH =8.6, 2.8 Hz, RhCH2CH2), 2.26 ppm (s,
3H, CNMe); 13C NMR (101 MHz, CDCl3): d=175.5 (s, CO2Me), 162.4
[1] a) Transition-Metal-Mediated Aromatic Ring Construction (Ed.: K. Tanaka),
Wiley, Hoboken, 2013; b) D. L. J. Broere, E. Ruijter, Synthesis 2012, 44,
2639–2672; c) S. Yamazaki, Inorg. Chim. Acta 2011, 366, 1–18; d) G.
Chem. Eur. J. 2016, 22, 13715 –13723
www.chemeurj.org
13722
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Domínguez, J. PØrez-Castells, Chem. Soc. Rev. 2011, 40, 3430–3444;
e) B. R. Galan, T. Rovis, Angew. Chem. Int. Ed. 2009, 48, 2830–2834;
Angew. Chem. 2009, 121, 2870–2874; f) K. Tanaka, Chem. Asian J. 2009,
4, 508–518; g) P. R. Chopade, J. Louie, Adv. Synth. Catal. 2006, 348,
2307–2327; h) S. Kotha, E. Brahmachary, K. Lahiri, Eur. J. Org. Chem.
2005, 4741–4767; i) S. Saito, Y. Yamamoto, Chem. Rev. 2000, 100, 2901–
2915; j) N. E. Schore, Chem. Rev. 1988, 88, 1081–1119.
[8] a) M. Fujiwara, I. Ojima in Modern Rhodium-Catalyzed Organic Reactions
(Ed.: P. A. Evans), Wiley-VCH, Weinheim, 2005, pp. 129–148; b) E. Ma-
tousovµ, R. Gyepes, I. Císarova, M. Kotora, Adv. Synth. Catal. 2016, 358,
254–267; c) T. Matsuda, K. Suzuki, Eur. J. Org. Chem. 2015, 3032–3035;
d) J. Hara, M. Ishida, M. Kobayashi, K. Noguchi, K. Tanaka, Angew. Chem.
Int. Ed. 2014, 53, 2956–2959; Angew. Chem. 2014, 126, 3000–3003; e) Y.
Shibata, K. Tanaka, Synthesis 2012, 44, 323–350; f) A. Dachs, A. Torrent,
A. Pla-Quintana, A. Roglans, A. Jutand, Organometallics 2009, 28, 6036–
6043; g) K. Tanaka, K. Toyoda, A. Wada, K. Shirasaka, M. Hirano, Chem.
Eur. J. 2005, 11, 1145–1156; h) K. Yoshida, I. Morimoto, K. Mitsudo, H.
Tanaka, Tetrahedron 2008, 64, 5800–5807; i) H. Kinoshita, H. Shinokubo,
K. Oshima, J. Am. Chem. Soc. 2003, 125, 7784–7785; j) M. Herberhold,
H. Yan, W. Milius, B. Wrackmeyer, Organometallics 2000, 19, 4289–4294;
k) A. Borrini, P. Diversi, G. Ingrosso, A. Lucherini, G. Serra, J. Mol. Cat.
1985, 30, 181–195.
[9] a) R. P. Kaiser, F. Hessler, J. Mosinger, I. Cisarova, M. Kotora, Chem. Eur. J.
2015, 21, 13577–13582; b) A. Dachs, S. Osuna, A. Roglans, M. Solà, Or-
ganometallics 2010, 29, 562–569.
[10] L. Orian, M. Swart, F. M. Bickelhaupt, ChemPhysChem 2014, 15, 219–
228.
[11] M. Parera, A. Dachs, M. Solµ, A. Pla-Quintana, A. Roglans, Chem. Eur. J.
2012, 18, 13097–13107.
[12] a) M. Paneque, C. M. Posadas, M. L. Poveda, N. Rendón, L. L. Santos, E.
lvarez, V. Salazar, K. Mereiter, E. OÇate, Organometallics 2007, 26,
3403–3415; b) D. PeÇa, D. PØrez, E. Guitiµn, L. Castedo, J. Am. Chem.
Soc. 1999, 121, 5827–5828.
[2] a) G. Domínguez, J. PØrez-Castells, Chem. Eur. J. 2016, 22, 6720–6739;
b) S. Alvarez, S. Medina, G. Dominguez, J. Org. Chem. 2015, 80, 2436–
2442; c) J.-P. Zhao, S.-C. Chan, C.-Y. Ho, Tetrahedron 2015, 71, 4426–
4431; d) G. Domínguez, J. PØrez-Castells in Comprehensive Organic Syn-
thesis II, Vol. 5, 2nd ed. (Eds.: P. Knockel, G. A. Molander), Elsevier, Am-
sterdam, 2014, pp. 1537–1581; e) N. Weding, M. Hapke, Chem. Soc. Rev.
2011, 40, 4525–4538; f) P. A. Inglesby, P. A. Evans, Chem. Soc. Rev. 2010,
39, 2791–2805; g) J. A. Varela, C. Saµ, J. Organomet. Chem. 2009, 694,
143–149; h) T. Shibata, K. Tsuchikama, Org. Biomol. Chem. 2008, 6,
1317–1323.
[3] For some mechanistic and/or theoretical studies, see: a) K. Yamamoto,
H. Tsurugi, K. Mashima, Chem. Eur. J. 2015, 21, 11369–11377; b) L. Orian,
L. P. Wolters, F. M. Bickelhaupt, Chem. Eur. J. 2013, 19, 13337–13347;
c) A. Dachs, A. Pla-Quintana, T. Parella, M. Solµ, A. Roglans, Chem. Eur. J.
2011, 17, 14493–14507; d) L. Orian, J. N. P. van Stralen, F. M. Bickel-
haupt, Organometallics 2007, 26, 3816–3830; e) V. Gandon, N. Agenet,
K. P. C. Vollhardt, M. Malacria, C. Aubert, J. Am. Chem. Soc. 2006, 128,
8509–8520; f) K. Kirchner, M. J. Calhorda, R. Schmid, L. F. Veiros, J. Am.
Chem. Soc. 2003, 125, 11721–11729.
[13] M. Paneque, C. M. Posadas, M. L. Poveda, N. Rendón, E. lvarez, K. Mer-
eiter, Chem. Eur. J. 2007, 13, 5160–5172.
[4] W. Reppe, W. J. Schweckendiek, Justus Liebigs Ann. Chem. 1948, 560,
104–116.
[14] M. Paneque, M. L. Poveda, N. Rendón, K. Mereiter, Organometallics
2009, 28, 172–180.
[15] Y. Fukumoto, H. Asai, M. Shimizu, N. Chatani, J. Am. Chem. Soc. 2007,
129, 13792–13793.
[16] P. A. Evans, D. Uraguchi, J. Am. Chem. Soc. 2003, 125, 7158–7159.
[17] W. J. Oldham, Jr., D. M. Heinekey, Organometallics 1997, 16, 467–474.
[18] M. Paneque, P. J. PØrez, A. Pizzano, M. L. Poveda, S. Taboada, M. Trujillo,
E. Carmona, Organometallics 1999, 18, 4304–4310.
[19] S. Trofimenko, J. Am. Chem. Soc. 1969, 91, 588–595.
[20] T. A. Albright, R. Hoffmann, J. C. Thibeault, D. L. Thorn, J. Am. Chem. Soc.
1979, 101, 3801–3812.
[5] M. Paneque, M. L. Poveda, N. Rendón, K. Mereiter, J. Am. Chem. Soc.
2004, 126, 1610–1611.
[6] a) J. A. Varela, S. G. Rubín, L. Castedo, C. Saµ, J. Org. Chem. 2008, 73,
1320–1332; b) J. A. Varela, S. G. Rubín, C. Gonzµlez-Rodríguez, L. Caste-
do, C. Saµ, J. Am. Chem. Soc. 2006, 128, 9262–9263.
[7] For examples, see cobalt-catalyzed alkyne cyclotrimerization: a) P.
Jungk, F. Fischer, I. Thiel, M. Hapke, J. Org. Chem. 2015, 80, 9781–9793;
b) I. Thiel, H. Jiao, A. Spannenberg, M. Hapke, Chem. Eur. J. 2013, 19,
2548–2554; c) J. P. Lomont, S. C. Nguyen, M. C. Zoerb, A. D. Hill, J. P.
Schlegel, C. B. Harris, Organometallics 2012, 31, 3582–3587; d) W. Wang,
S. Inoue, S. Enthaler, M. Driess, Angew. Chem. Int. Ed. 2012, 51, 6167–
6171; Angew. Chem. 2012, 124, 6271–6275; e) N. Agenet, V. Gandon,
K. P. C. Vollhardt, M. Malacria, C. Aubert, J. Am. Chem. Soc. 2007, 129,
8860–8871; f) G. Hilt, T. Vogler, W. Hess, F. Galbiati, Chem. Commun.
2005, 1474–1475; g) F. Slowinski, C. Aubert, M. Malacria, J. Org. Chem.
2003, 68, 378–386; nickel-catalyzed alkyne cyclotrimerization: h) S. Pal,
C. Uyeda, J. Am. Chem. Soc. 2015, 137, 8042–8045; i) S. K. Rodrigo, I. V.
Powell, M. G. Coleman, J. A. Krausea, H. Guan, Org. Biomol. Chem. 2013,
11, 7653–7657. Ruthenium-catalyzed alkyne cyclotrimerization: j) K.
Geetharani, S. Tussupbayev, J. Borowka, M. C. Holthausen, S. Ghosh,
Chem. Eur. J. 2012, 18, 8482–8489; k) C.-Y. Wu, Y.-C. Lin, P.-T. Chou, Y.
Wang, Y.-H. Liu, Dalton Trans. 2011, 40, 3748–3753; l) V. Cadierno, S. E.
García-Garrido, J. Gimeno, J. Am. Chem. Soc. 2006, 128, 15094–15095;
m) E. Rüba, R. Schmid, K. Kirchner, M. J. Calhorda, J. Organomet. Chem.
2003, 682, 204–211; n) Y. Yamamoto, T. Arakawa, R. Ogawa, K. Itoh, J.
Am. Chem. Soc. 2003, 125, 12143–12160; iridium-catalyzed alkyne cy-
clotrimerization: o) Y. Masuyama, K. Miyazaki, N. Suzuki, Asian J. Org.
Chem. 2013, 2, 750–754; p) J. M. O’Connor, A. P. Closson, R. L. Holland,
S. K. Cope, C. L. VØlez, C. E. Moore, A. L. Rheingold, Inorg. Chim. Acta
2010, 364, 220–225; q) E. Farnetti, S. Filipuzzi, Inorg. Chim. Acta 2010,
363, 467–473; r) E. Farnetti, N. Marsich, J. Organomet. Chem. 2004, 689,
14–17.
[21] J. M. O’Connor, A. Closson, P. Gantzel, J. Am. Chem. Soc. 2002, 124,
2434–2435.
[22] a) For an example in which a niobanorbornene is proposed to release
cyclohexadiene compounds, see: Y. Obora, K. Takeshita, Y. Ishii, Org.
Biomol. Chem. 2009, 7, 428–431; b) V. M. Visciglio, J. R. Clark, M. T.
Nguyen, D. R. Mulford, P. E. Fanwick, I. P. Rothwell, J. Am. Chem. Soc.
1997, 119, 3490–3499.
[23] A. A. Dahy, N. Koga, Organometallics 2015, 34, 4965–4974.
[24] M. C. Nicasio, M. Paneque, P. J. PØrez, A. Pizzano, M. L. Poveda, L. Rey, S.
Sirol, S. Taboada, M. Trujillo, A. Monge, C. Ruiz, E. Carmona, Inorg. Chem.
2000, 39, 180–188.
[25] During the evaluation of this paper, a reviewer suggested to study the
catalytic activity on the cyclotrimerization of DMAD of compounds 1, 2,
3, and 5, in addition to the already reported activity of 4. Since 3 and 5
are inert to DMAD (see the Results and Discussion), and 1 yields mainly
3 upon heating with DMAD, 2 and 4 are the only species in this system
that could act as catalysts. However, 2 is partially converted in situ to 4,
for which the catalytic activity has been described above.
Received: April 25, 2016
Published online on August 18, 2016
Chem. Eur. J. 2016, 22, 13715 –13723
www.chemeurj.org
13723
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim