Cobalt-mediated linear 2:1 Co-oligomerization of alkynes with enol ethers to give 1-alkoxy-1,3,5-trienes: A missing mode of reactivity

Article abstract of DOI:10.1002/chem.201000486

A variety of 1,6-heptadiynes and certain borylalkynes co-oligomerize with enol ethers in the presence of [CpCo(C₂H₄)₂] (Cp = cyclopentadienyl) to furnish the hitherto elusive acyclic 2:1 products, 1,3,5-trien-1-ol ethers, in preference to or in competition with the alternative pathway that leads to the standard [2+2 + 2] cycloadducts, 5-alkoxy-1,3- cyclohexadienes. Minor variations, such as lengthening the diyne tether, cause reversion to the standard mechanism. The trienes, including synthetically potent borylated derivatives, are generated with excellent levels of chemo-, regio-, and diastereoselectivity, and are obtained directly by decomplexation of the crude mixtures during chromatography. The cyclohexadienes are isolated as the corresponding dehydroalkoxylated arenes. In one example, even ethene functions as a linear cotrimerization partner. The alkoxytrienes are thermally labile with respect to 6πelectrocyclization-elimination to give the same arenes that are the products of cycloaddition. The latter, regardless of the mechanism of their formation, can be viewed as the result of a formal [2+2+2] cyclization of the starting alkynes with acetylene. One-pot conditions for the exclusive formation of arenes are developed. DFT computations indicate that cyclohexadiene and triene formation share a common intermediate, a cobaltacycloheptadiene, from which reductive elimination and β-hydride elimination compete.

Full text of DOI:10.1002/chem.201000486

DOI: 10.1002/chem.201000486
Cobalt-Mediated Linear 2:1 Co-oligomerization of Alkynes with Enol Ethers
to Give 1-Alkoxy-1,3,5-Trienes: A Missing Mode of Reactivity
David Lebœuf,[a] Laura Iannazzo,[a] Anaꢀs Geny,[a] Max Malacria,[a]
K. Peter C. Vollhardt,[c] Corinne Aubert,*[a] and Vincent Gandon*[b]
Abstract: A variety of 1,6-heptadiynes
and certain borylalkynes co-oligomer-
ize with enol ethers in the presence of
chemo-, regio-, and diastereoselectivity,
and are obtained directly by decom-
plexation of the crude mixtures during
chromatography. The cyclohexadienes
are isolated as the corresponding dehy-
droalkoxylated arenes. In one example,
even ethene functions as a linear cotri-
merization partner. The alkoxytrienes
are thermally labile with respect to 6p-
electrocyclization–elimination to give
the same arenes that are the products
of cycloaddition. The latter, regardless
of the mechanism of their formation,
can be viewed as the result of a formal
[2+2+2] cyclization of the starting al-
kynes with acetylene. One-pot condi-
tions for the exclusive formation of
arenes are developed. DFT computa-
tions indicate that cyclohexadiene and
triene formation share a common inter-
[CpCoACHTUNGTRENNUNG(C2H4)2] (Cp=cyclopentadienyl)
to furnish the hitherto elusive acyclic
2:1 products, 1,3,5-trien-1-ol ethers, in
preference to or in competition with
the alternative pathway that leads to
the standard [2+2+2] cycloadducts, 5-
alkoxy-1,3-cyclohexadienes. Minor var-
iations, such as lengthening the diyne
tether, cause reversion to the standard
mechanism. The trienes, including syn-
thetically potent borylated derivatives,
are generated with excellent levels of
mediate,
a
cobaltacycloheptadiene,
from which reductive elimination and
b-hydride elimination compete.
À
Keywords: boron · C H activation ·
cobalt · coupling reactions · electro-
cyclic reactions
Introduction
metal-catalyzed cross-coupling.[1,2] However, under specific
conditions, it is possible to assemble trienes in a single oper-
ation from two alkynes and one alkene (Scheme 1). To the
best of our knowledge, the first example of such a reaction
was reported by Cairns and co-workers in 1952.[3] They
1,3,5-Trienes are key subunits of a large variety of natural
products, and are also synthetic precursors of cyclohexa-
dienes through 6p-electrocyclization of the cis isomers. They
are usually prepared in several steps, often including a
found that a mixture of [Ni(CO)4] and PACTHNUGTRNEGNU(C6H5)3 catalyzed
the addition of acetylene to acrylic compounds to give hep-
tatrienoic acid derivatives. In 1999, Cheng extended this re-
action to substituted alkynes by employing mixtures of [Ni-
[a] Dr. D. Lebœuf, L. Iannazzo, A. Geny, Prof. Dr. M. Malacria,
Dr. C. Aubert
AHCTUNGTRENNUNG
(PR3)2X2] and Zn.[4] Ruthenium entered the scene in 1998
UPMC Univ Paris 06, IPCM UMR CNRS 7201
4 Place Jussieu, 75005 Paris (France)
Fax : (+33)1-44-27-73-60
when Yi discovered that acetylene and acrylonitrile trans-
E-mail: corinne.aubert@upmc.fr
[b] Prof. Dr. V. Gandon
Univ Paris-Sud, UMR CNRS 8182
Orsay, 91405 (France)
Fax : (+33)1-69-14-47-47
E-mail: vincent.gandon@u-psud.fr
[c] Prof. Dr. K. P. C. Vollhardt
Department of Chemistry, University of California at Berkeley
and the Chemical Sciences Division
Lawrence Berkeley National Laboratory
Berkeley, California 94720-1460 (USA)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000486.
Scheme 1. Generic competitive outcome of the formation of cyclic versus
linear oligomers of two alkynes and one alkene.
8904
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8904 – 8913

FULL PAPER
formed to heptatrienenitrile in the presence of a Cp*ruthe-
nacyclopentadiene catalyst (Cp*=pentamethylcyclopenta-
dienyl).[5] This process was synthetically exploited by Saꢁ
Results and Discussion
Co-oligomerizations of a,w-diynes with enol ethers: This in-
vestigation began in an attempt to extend the scope of the
[2+2+2] cycloaddition of enol ethers to a,w-diynes, initially
by using the previously tested tert-butyl,[12a] and then with
other vinyl ethers such as phenyl ethenyl ether (Table 1).
Operationally, the diyne was added to a solution of [CpCo-
and co-workers in 2006, who utilized [Cp*RuACTHNUGTRNEUNG(CH3CN)3]PF6
to couple a,w-diynes with acyclic alkenes to give trienic
products.[6] With cyclic alkenes, the standard [2+2+2] cy-
cloaddition pathway prevailed, leading to 1,3-cyclohexa-
dienes (Scheme 1).[7] DFT calculations pinpointed
a
common ruthenacycloheptadiene intermediate, from which
the two products arise by either b-hydride or reductive elim-
ination, respectively.[6b] Very recently, a single case of a
G
diyne–acrylACHTUNGTRENNUNGamide coupling to generate a hepta-2,4,6-trien-
ACHTUNGTRENNUNGamide by using [RhACHTUNTGREN(NUGN cod)2]BF4 (cod=cyclooctadiene) ad-
mixed with a chelating phosphine was reported.[8]
In view of the similarity in the chemistry of these systems
to that of [CpCoL2], (Cp=cyclopentadienyl) particularly
with respect to the cocyclization of alkynes with alkenes,[7] it
is surprising that trienes have remained undetected in the
Co series. The only exceptions are occasional dienylhetero-
cyclic products formed in the [2+2+2] cycloaddition of al-
kynes to heteroaromatic double bonds.[9] However, their as-
sembly is mechanistically distinct, and proceeds through
direct electrophilic attack of the cobaltacyclopentadiene in-
termediate on the heterocycle and an ensuing cobalt-assisted
hydrogen transfer (see below).[10] We report examples of the
missing type of Co reactivity, when certain alkynes are treat-
ed with certain enol ethers in the presence of [CpCoL2] me-
diators to produce trienol ethers in a chemo-, regio-, and
diastereoselective fashion. The reaction tolerates boryl sub-
stitution, thus enabling access to borylated alkoxytrienes.
Subtle changes in the nature of the reagents reverts the out-
come of this process to that of the traditional [2+2+2]
pathway, as evidenced by the formation of fused arenes orig-
inating from the demetallation–dehydroalkoxylation of the
corresponding 5-alkoxycyclohexadiene complexes.[11] The
same type of arene is obtained on thermal 6p-electrocycliza-
tion of the trienol ethers.
The discovery of this new type of reactivity exhibited by
[CpCoL2] was made through what appeared to be a routine
extension of our work on the cobalt-mediated [2+2+2] cy-
cloaddition of borylated monoalkynes and diynes with al-
kenes, including enol ethers, which, without exception, to
produce the corresponding [CpCo(cyclohexadiene)] com-
ACTHNUTRGNEUNG
plexes.[12] It is of relevance that [CpCo(C2H4)2],[13] on its
own, furnished products of ethene incorporation, whereas
added alkenes outcompeted ethene in the cyclization. More-
over, variation of the tether lengths in a range of a,w-diynes
from 3 to 5 atoms did not alter the outcome of this process.
However, examples featuring enol ethers were restricted to
cocyclization with the monoalkyne (phenylethynyl)pinacol-
borate. The synthetic value of the products[14] prompted the
current study.
[a] The crude mixture was treated with FeCl3·6H2O (1.5 equiv) in MeCN
at RT before purification. [b] Complex mixture. [c] No reaction.
[d] BPin=boryl pinacolate. [e] Atom X is indicated by a C. [f] This com-
pound is known, see the Supporting Information.
Chem. Eur. J. 2010, 16, 8904 – 8913
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8905

C. Aubert, V. Gandon et al.
as triene 3b on the basis of the NMR spectral data.[15] Thus,
the H NMR spectrum revealed a set of three diagnostic al-
Ring closure is nevertheless attainable by thermal means,
a process that is accompanied by aromatization. Thus, 3b
transformed quantitatively into 2b in boiling toluene (12 h),
or in hot DMF (10 min) under microwave irradiation, by
electrocyclization–dehydroalkoxylation. Similarly, 3d and
3’g were converted to 4 and 5, respectively, for which the
1
kenyl signals at d=6.13 (d), 6.68 (d), and 6.91 ppm (s), the
trans configuration of the enol ether moiety established by
J=12.2 Hz. The stereochemical arrangement of the other
two double bonds was confirmed by an NOE experiment,
which showed a through-space interaction between the hy-
drogen giving rise to a singlet at d=6.91 ppm and that asso-
ciated with the doublet at d=6.13 ppm. Conversely, irradiat-
ing the peak at d=6.68 ppm gave no response (see also the
Supporting Information). The 13C NMR spectrum contained
the appropriate number of CHn signals, notably the charac-
teristically extreme chemical shifts for the sp2-hybridized
carbons of the tert-butoxy-bearing double bond at d=113.2
(Cb) and 147.0 ppm (Ca).[15]
deACTHNUTRGNENGbU oroACHTUNGTNERaNUGN lkoxylation occurred spontaneously (Scheme 2). The
onset of this cascade is sufficiently slow below 808C that
other useful transformations can be executed on these sys-
tems. For example, 3d underwent successful double Suzuki
cross-coupling with p-iodoanisole, giving 6 in 77% yield
(Scheme 3).[17]
The high level of chemo-, regio-, and diastereoselectivity
of the synthesis of 3b encouraged further investigation of its
scope by incremental changes in the respective starting ma-
terials. For example, switching from phenyl to trimethylsilyl
substitution, as in 1c, shut down reactivity (Table 1, entry 3).
On the other hand, the arguably fairly bulky boryl pinaco-
late (BPin) group in 1d is tolerated, providing the novel di-
borylated alkoxytriene 3d in fairly good yield (Table 1,
entry 4).[16] Notably, diyne 1d may be regarded as a tethered
analog of (phenylethynyl)pinacolborate, which engaged the
tert-butyl enol ether only in the standard [2+2+2] cycload-
dition.[12a] Phenyl vinyl ether as a new alkene component re-
sulted in the corresponding 3’d in 81% yield (Table 1,
entry 5). Changing the three-atom tether to incorporate var-
ious functions, as in 1e–j, gave similar results with both enol
ethers, to produce trienes selectively and in modest to good
yields (Table 1, entries 6–14). In no case could we detect
products of cyclization, signaling a change in mechanism
with a,w-diyne substrates.
Scheme 2. Thermal electrocyclization–elimination of trienol ethers.
BPin=boryl pinacolate.
However, it became evident that this conclusion was pre-
mature on inspection of entries 15–21 of Table 1. In these
1
cases, H NMR inspection of the crude mixtures showed the
presence of the aromatized product 2, in addition to very
sensitive CpCo complexes of 5-alkoxycyclohexadienes. To
simplify this outcome operationally, the crude product was
worked up with ironACHTUNGTRNEG(UN III) chloride, followed by chromatogra-
phy. This treatment effected the demetallation and aromati-
zation of the ligands to create 2, again in moderate to good
yields. Specifically, entry 15 reveals that a simple substituent
change on 1,6-heptadiyne from Ph (entry 2) or BPin
(entry 4) to methoxycarbonyl causes mechanistic reversion
of the reaction to the [2+2+2] pathway, with 2k as the sole
product in 73% yield. Similarly, elongating the tether slight-
ly (entry 16) or more substantially (entries 17–21) leads ex-
clusively to arenes 2 l–o. To ascertain that small amounts of
trienes in the crude mixtures had not been decomposed or
converted during work-up into 2, a preparation of 3b was
subjected to FeIII treatment, without any effect. The stability
of an isolated sample of 3b toward FeIII was also checked.
This experiment also rules out the occurrence of catalyzed
hexatriene ring closure.[2]
Scheme 3. Double Suzuki coupling of 3d.
Two further experiments attest to the possibility that the
scope of the 2:1 linear co-oligomerization of diynes extends
beyond the examples in Table 1. The first featured the cyclic
ether 2,3-dihydrofuran, which, on exposure to 1p, generated
7 (Scheme 4). The unconjugated nature of its triene array is
evident from the NMR spectral data and implicates the oc-
currence of a b-hydride elimination from the cobaltacyclo-
heptadiene intermediate A (see also the mechanistic sec-
tion). Notably, there was no evidence for the formation of
the isomeric phenylethanol of type 2, as observed in a relat-
ed reaction that used Ir catalysis.[11d]
8906
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8904 – 8913

Co-oligomerization of Alkynes with Enol Ethers
FULL PAPER
unambiguous NMR data. Upon the same oxidative treat-
ment, the regioisomeric complexes 9 and 10 delivered an
identical product, the triene 12, which could also be readily
characterized. Thus, it appears that the linear co-oligomeri-
zation pathway is energetically competitive with the stan-
dard cyclization without alkoxy substitution of the ene com-
ponent.
Intermolecular co-oligomerizations of borylalkynes with
enol ethers: The results of the previous section raised the
question of whether monoalkynes could be induced to enter
into the linear 2:1 co-oligomerization. Again extending prior
work,[12] the initial focus was on the use of alkynyl borates.
The results re-emphasize the energetic closeness of the
three regiochemical pathways for alkyne oxidative coupling,
and the ensuing competition toward cyclic versus acyclic
products (Table 2). The starting point was a repetition of the
Scheme 4. Co-oligomerization of 1p with 2,3-dihydrofuran (X=C-
ACHTUNGTRENNUNG(CH2OMe)2.
The second case is more far-reaching, and involved simply
Table 2. Co-oligomerization of borylated alkynes with enol ethers. BPin=
ethene as the alkene partner, again a substrate that had pre-
viously only given cyclohexadiene products, including those
derived from cocyclizations with a variety of diborylated
a,w-diynes.[12] Remarkably, the diyne 1q, not tested previ-
boryl pinacolate.
ously, transformed in the presence of [CpCoACHTUNRGTNEUNG(C2H4)2] not
only to the expected 8 (16%), but also to triene complexes
9 (24%) and 10 (24%). These complexes were separable by
chromatography and fully characterizable, an exercise great-
ly aided by their relative stability. Interestingly, the forma-
tion of 10 is of kinetic origin, as it isomerizes to 9 quantita-
tively by a “CpCo-walk”[18] on prolonged standing in solu-
tion. Further structural proof rested on oxidative demetalla-
tion of 8 to furnish the free ligand 11 (Scheme 5), displaying
Entry Alkyne
R
Y
Yield [%] Yield [%] Yield [%] Yield [%]
of 14
tBu 59[a]
of 14’
of 15/15’ of 16/16’
1
2
3
4
5
6
7
8
13a
13a
13b
13b
13c
13c
13d
13d
Ph
Ph
–
–
–
–
33
71
–
–
–
–
–
–
–
–
Ph
–
55[a]
CH2OMe tBu 88[a]
–
CH2OMe Ph 55[a]
31[a]
–
[b]
[b]
Pr
Pr
iBu
iBu
tBu
Ph
tBu
Ph
–
–
–
–
–
–
–
59
72
–
[a] The crude mixture was treated with FeCl3·6H2O (1.5 equiv) in MeCN at
RT before purification. [b] Observed but unstable (see text).
reaction of 13a with tert-butyl vinyl ether, which had pro-
duced selectively the 1,3-diboryl-2,4-diphenyl-5-tert-butoxy-
cyclohexa-1,3-diene CpCo complex in 65% yield.[12a] Under
the conditions of work-up described in Table 1, this experi-
ment rendered the unsymmetrical 14a in 59% yield, as ex-
pected (Table 2, entry 1). However, by using phenyl vinyl
ether instead, a dramatic change in the reaction outcome
ensued: trienol ether 16’a, displaying a 1,4 arrangement of
the boron atoms, was obtained, as evidenced by the pres-
ence of the characteristically shielded vinyl boron hydrogen
(Table 2, entry 2).[19] A similar, if attenuated, switchover oc-
curred in the case of methoxymethylated 13b: the tert-bu-
toxyalkene led to arene 14b only (Table 2, entry 3), whereas
the phenoxy analog gave substantial amounts of triene 16’b,
in addition to 14b (Table 2, entry 4). In contrast, the alkyl
derivatives 13c and 13d assembled alkoxytrienes irrespec-
tive of the vinyl ether employed, but now in the regioiso-
Scheme 5. Co-oligomerization of 1h with ethene.
Chem. Eur. J. 2010, 16, 8904 – 8913
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8907

C. Aubert, V. Gandon et al.
meric form 15, which is associated with protodeborylation at
2b, albeit in poor yield (Table 3, entry 1). On the other
hand, incomplete decomposition of the triene complexes
and/or incomplete ring closure of 3b was signaled when
using II as the cobalt source (Table 3, entry 2), again in poor
yield. Surprisingly, this run revealed, in addition to 2b and
3b, the formation of another isomer of the latter, iso-3b. Its
NMR spectra are very similar to those of 3b (including the
trans coupling of the enol ether group), with one marked ex-
ception: the singlet for the styrenyl hydrogen at the triene
terminus is shifted upfield by almost 1 ppm, a result of
shielding by the other, now proximal, phenyl ring. This prox-
imity was confirmed by NOE experiments, and was further
supported by the recording of an NOE between the hydro-
gen giving rise to the doublet at d=6.22 ppm (the enol
ether hydrogen next to the alkoxy group) and the allylic
methylene hydrogen atoms resonating at d=2.62–2.69 ppm
(see the Supporting Information). The mechanism of the
isomerization of 3b is obscure, although a number of species
in the thermolysis mixture could be invoked as acting as cat-
alysts for such. Fortunately with respect to the goal of this
line of investigation, pure iso-3b, although more resilient
than 3b, also transformed thermally to 2b (2008C, mW,
10 min). Complex III gave the best mass recovery under
these initial conditions, but still substantial quantities of iso-
3b (Table 3, entry 3).
the 4-position.[20] Moreover, the propyl-substituted trienes
1
15c and 15’c, although clearly observable in the H NMR
spectra of the crude reaction mixtures, were prone to re-
markably
rapid
electrocyclization–dehydroalkoxylation
during purification, and could be isolated only as the corre-
sponding arene 14’c (Table 2, entries 5 and 6). This problem
was circumvented with the bulkier isobutyl group (Table 2,
entries 7 and 8). Although more work is required to delin-
eate the scope and utility of this reaction, it is phenomeno-
logically clear that certain constellations of monoalkynes
and enol ethers will produce 2:1 co-oligomerization prod-
ucts.
One-pot preparation of fused arenes by co-oligomerization
of 1,6-diynes with enol ethers: Regardless of their mecha-
nism of formation, arenes 2 (and 14) are the result of a
formal CpCo-catalyzed [2+2+2] cycloaddition of the re-
spective alkynes with acetylene, a strategy that has been
used previously with other metals to circumvent the draw-
backs associated with acetylene itself.[11] Considering the
sensitivity of the Co complexes generated under the low-
temperature conditions employed above and the electrocy-
ACHTUNGTRENNUNGcliACHTUNGTRENNUNGzation–aromatization cascade traversed by the triene
products at elevated temperatures, it seemed possible that
executing the co-oligomerization reaction above 1108C
would channel it toward exclusive arene formation. Initial
tests of the feasibility of this notion were carried out with
1b and tert-butyl vinyl ether, by using the three cobalt com-
Because boiling toluene was not sufficient to convert all
triene products to 2b, the reaction temperature was in-
creased to 2008C (DMF, microwave reactor). This change
had the desired effect, even though yields were still moder-
ate (Table 3, entries 4–6). We next engaged substoichiomet-
ric amounts of metal (Table 3, entries 7–9). This change was
most successful in the case of complex III: with 0.4 equiva-
lents, the isolated yield of 2b
plexes [CpCo(CO)2] (I), [CpCoACTHUNTRGNEUNG(C2H4)2] (II), and the recent-
ly reported air-stable catalyst III[21] (Table 3). Gratifyingly,
already in boiling toluene (12 h), I gave rise exclusively to
was essentially quantitative
(Table 3, entry 9). Unfortunate-
ly, lesser Co loading was detri-
Table 3. Optimization of the reaction conditions for a formal [2+2+2] cycloaddition of diynes to acetylene.
mental
(Table 3,
entry 10).
Therefore, the conditions of
Table 3, entry 9 were chosen in
order to explore further the
scope and limitations of this
one-pot process (Table 4). First,
several enol ethers other than
tert-butyl were tested as surro-
gates for acetylene in the reac-
tion of 1b. However, none of
Entry
[Co] ([mol%])[a]
Conditions[b]
Yield [%] (2b:3b:iso-3b)[c]
them
proved
comparable
1
2
3
4
5
6
7
8
I (100)
II (100)
III (100)
I (100)
II (100)
III (100)
I (40)
II (40)
III (40)
III (20)
toluene, reflux, 12 h
toluene, reflux, 12 h
toluene, reflux, 12 h
DMF, mW, 2008C, 10 min
DMF, mW, 2008C, 10 min
DMF, mW, 2008C, 10 min
DMF, mW, 2008C, 20 min
DMF, mW, 2008C, 20 min
DMF, mW, 2008C, 20 min
DMF, mW, 2008C, 20 min
32 (1:0:0)
37 (0.3:1:0.4)
68 (0.6:0:1)
18 (1:0:0)
51 (1:0:0)
38 (1:0:0)
28 (1:0:0)
28 (1:0:0)
96 (1:0:0)
58 (1:0:0)
(Table 4, entries 1–5). Next,
substituents on the three-
carbon tether of the diyne
(Table 4, entries 6 and 9), the
tether length (Table 4, entries 8
and 10), and the terminal
alkyne groups (Table 4, entry 7)
were altered. In all cases mod-
erate to good yields of the
fused arene 2 were obtained.
9
10
[a] Based on diyne. [b] The diyne (2 equiv), enol ether (5 equiv), and Co species were mixed prior to heating.
[c] Products obtained directly after chromatography on silica gel.
8908
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8904 – 8913

Co-oligomerization of Alkynes with Enol Ethers
FULL PAPER
Table 4. Scope and limitations of the formal [2+2+2] cycloaddition of
diynes to acetylene.
Entry Diyne
R
X
Y
Product Yield [%]
1
2
3
4
5
1b
1b
1b
1b
1b
Ph
Ph
Ph
Ph
Ph
CH2
CH2
CH2
CH2
CH2
Ph
Et
Bu
2b
2b
2b
53
44
64
23
32
n-C12H25 2b
Ac
Scheme 6. Mechanistic pathways of the co-oligomerization of enol ethers
with alkynes subjected to computational scrutiny. RE=reductive elimina-
tion; b-HE=b-hydride elimination.
2b
6
1e
Ph
tBu
2e[b]
41
7
8
9
10
1k
CO2Et CH2
tBu
tBu
tBu
2k
2m
2r
41
67
that although pathway 1 prevails with ethene, pathway 2
(stepwise) becomes competitive with aromatics, such as ben-
zene, furan, thiophene,[10a] and pyridones.[10b]
To deduce the relative likelihood of the options represent-
ed in Scheme 6, DFT computations were carried out. All ge-
ometries of molecules and transition states were optimized
fully without symmetry constraints by using the Gaussian 03
program.[23] Calculations were carried out with the BP86
1m
1r
Ph
Ph
Ph
G
C
N
74
1s[c]
U
2s
53[d]
[a] Atom X is indicated by a C. [b] This compound is known, see Sup-
porting Information. [c] 1,8-Diphenylocta-1,7-diyne-3,6-dione. [d] Yield
of 5,8-diphenylnaphthalene-1,4-dione, the product of air oxidation.
G
ACHTUNGTRENNUNG
Thus, this method constitutes a viable and possibly more
general alternative to those based on Rh and Ir.[11]
functional as implemented and the LACVPACTHNUGRTENUNG(d,p) basis set.
The cobalt atom was described by a double-z basis set with
Mechanistic considerations and DFT calculations: It is
tempting to interpret our experimental results as mimicking
mechanistically the trails computed previously for Co[10a]
and Ru[6b] (Scheme 1), in which alkynes are exposed to
normal (i.e., nonalkoxylated) alkenes. However, the discov-
ery of a different channel, featuring direct electrophilic
attack of the cobaltacyclopentadiene intermediate coupled
with a cobalt-assisted hydrogen transfer in the case of het-
erocycle dienylation (see the Introduction),[10] and consider-
ation of the relatively electron-rich nature of the enol ether
double bond, made the occurrence of such a pathway a rea-
sonable alternative to the b-hydride elimination route.
Three generic routes were addressed in the following inves-
tigation (Scheme 7). The most likely is the initial oxidative
coupling of the two alkyne units, leading to a metallacyclo-
pentadiene intermediate. The standard mechanism would
then proceed by alkene insertion, giving rise to a seven-
membered metallacycle (Scheme 6, pathway 1). Its direct re-
ductive elimination would generate the 1,3-cyclohexadiene
core, whereas b-hydride elimination followed by reductive
elimination would result in the triene framework. The viable
alternative to (Scheme 6, pathway 1) is metathesis of the
the effective core potential of Hay and Wadt
ACTHNUTRGNEUNG
(LANL2DZ),[24] and the 6-31G(d,p) basis set[25] was used for
the other elements. Frequency calculations were performed
to confirm the nature of the stationary points and to obtain
zero-point energies (ZPE). The connectivity between sta-
tionary points was established by intrinsic reaction coordi-
nate calculations (IRC). The energy values (kcalmolÀ1)
refer to enthalpies (DH298 and DH°298). Taking the free ener-
gies into account proved inconsequential with respect to the
conclusions reached. Methyl vinyl ether and acetylene were
chosen as model substrates.
We had established previously that ligand substitution on
CpCo is dissociative, that the ground states of 16-electron
species are triplets, whereas those of their 18-electron coun-
terparts are singlets, and that the oxidative coupling of two
alkynes is energetically more favorable than the coupling of
one alkyne with one alkene.[10a,12a,26] At the level of theory
employed, the removal of ethene from [CpCoACHTUNRGTNEUNG(C2H4)2] leads
3[CpCo
to the triplet species ACTHNUGTRENNUNG AHCTUNGTREN(NUGN C2H4)] at an enthalpic cost of
22.6 kcalmolÀ1 (Scheme 7). Complexation of the enol ether
gives rise to the singlet complex E, lying 2.9 kcalmolÀ1
above the reactants.[27] Activation of any C H bond of the
À
À
C H bond of the alkene moiety (Scheme 6, pathway 2),
enol ether ligand to give a cobalt hydride such as F requires
a prohibitive enthalpy of activation above 30 kcalmolÀ1 (see
for instance TSEF) and is strongly endothermic. On the other
either in a stepwise fashion via a vinyl dienyl metal inter-
mediate (oxidative hydrogen migration),[22] or in a concerted
manner leading directly to a metallacyclopentene, the reduc-
tive elimination of which furnishes the triene system. Un-
likely, but included in our considerations for calibration, is
3[CpCo
hand, complexation of acetylene to ACTHNUGTRENNUGN CAHTUNGTRNEN(UGN C2H4)] to give G
is exothermic. Dissociation of the second ethene ligand re-
quires 2.51 kcalmolÀ1, and the resulting triplet species may
either coordinate the enol ether to furnish H endothermical-
ly, or coordinate acetylene to provide J, which lies
0.9 kcalmolÀ1 below the reactants. Transformation of H into
À
early C H activation at the alkene (Scheme 6, pathway 3),
followed by two carbometallations of the triple bonds and
reductive elimination (RE). Previous calculations suggested
Chem. Eur. J. 2010, 16, 8904 – 8913
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8909

C. Aubert, V. Gandon et al.
À
Scheme 7. Competition between alkyne coupling and vinyl C H activa-
tion pathways in the coordination sphere of CpCo (enthalpies, kcal
molÀ1).
Scheme 8. Computed possible isomerizations of L.
the cobalt hydride I or any other regioisomer again requires
a very high enthalpy of activation (e.g., TSHI) and is strongly
endothermic, whereas the transformation of J into cobalta-
cyclopentadiene K is realized at an enthalpic cost of only
6.9 kcalmolÀ1, and is exothermic by 27.7 kcalmolÀ1. Barrier-
less complexation of the enol ether may give various regio-
and diastereomers L with close-lying energies in the range
of 2.6 kcalmolÀ1,[28] the most stable located at À47.8 kcal
molÀ1. The upshot of this initial set of computations is that
early vinyl-H activations according to pathway 3 depicted in
Scheme 6 are very unlikely. Our focus therefore turned to
species L and its various options of isomerization, summa-
Table 5. Enthalpies of species L–U and associated transition states rela-
tive to the most stable isomer of L [kcalmolÀ1].
Species
R1 =OMe[a]
R2 =OMe[a]
R3 =OMe[a]
R4 =OMe[a]
L
M
N
O
P
Q
R
S
0.1
–
0.0
2.6
À5.3
À5.9
1.7
[b]
[b]
À3.3
–
À9.3
À25.5
À11.8
À10.3
À54.8
À30.8
À37.1
13.4
À9.3
À9.7
[c]
[d]
À35.9
À11.8
À7.2
–
–
À12.6
À13.4
À53.9
À31.2
À36.1
8.5
32.9
20.9
4.9
À12.6
À8.9
À53.9
À21.3
À36.7
10.8
À54.8
[e]
–
–
–
–
[e]
ACHTUNGTRENNUNG
T
U
[e]
[e]
TSLS
TSLU
TSLM
TSMN
TSNO
TSNP
TSNQ
TSPR
TSQS
TSST
24.3
17.5
23.1
17.2
(pathway 2 in Scheme 6) to dienyl vinylcobalt intermediates
U, and metathesis to metallacyclopentenes of type S, could
be modeled, they appear to be noncompetitive. Thus, the
former are substantially endothermic and require apprecia-
ble activation (TSLU ꢀ17–21 kcalmolÀ1). The latter are exo-
thermic but still require more than 23 kcalmolÀ1 of activa-
tion. In contrast, insertion of the enol ether ligand into a
[e]
–
5.3
3.8
À1.1
16.2
3.8
5.6
[f]
[f]
–
À0.6
–
–
[c]
[g]
15.0
4.6
3.0
À3.8
À8.7
À1.8
–
À1.6
8.5
5.0
À5.3
À8.4
1.3
6.4
8.1
À3.8
À5.3
À13.0
À2.6
[e]
–
–
[e]
À
Co C bond to give seven-membered complexes N, either in
stepwise fashion via the cobalt carbene M or concertedly, is
straightforward, with enthalpies of activation between 3.8
and 5.6 kcalmolÀ1. The respective formation of both M and
[a] Position of the OMe group in the respective complexes. The other R
groups in each case represent H. [b] Collapses to N. [c] Not applicable
(R3 =H). [d] Not computed, as the corresponding TS could not be locat-
ed. [e] Not applicable (R2 =H). [f] Not computed, as species M collapses
to N in this case. [g] Not found.
N
is reasonably exothermic by approximately 3–
10 kcalmolÀ1. Thus, insertion remains the preference of the
bound alkene in L, regardless of the presence of the alkoxy
group (Table 5).
consequence of the h2-bonding of the metal to the distal
double bond, revealed that R3 is positioned quite close to
the a’ carbon. This finding, in turn, suggested the possibility
of a hitherto unrecognized 1,5-shift (R3 =H), which would
assemble O, from which triene complexes would emerge by
Having ascertained the intermediacy of N, the next task
was to quantify the bifurcation toward products R and T.
Before considering the traditional intermediates P and Q,
inspection of the rigid and severely bent structure of N, a
8910
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8904 – 8913

Co-oligomerization of Alkynes with Enol Ethers
FULL PAPER
valence tautomerization. Indeed, such a rearrangement
could be computed for two isomers of N, but at a prohibi-
tively high enthalpic cost (TSNO). Returning to the estab-
lished mechanisms,[10a] but now as potentially modified by
methoxy substitution, the energies connected to the isomeri-
zation of N to R via P, and of N to T via Q and S were com-
puted (Table 5). Both processes traverse through relatively
low-lying transition states, and no pronounced energetic
predilection for either pathway can be discerned, as ob-
served experimentally. Interestingly, with the methoxy group
at R1 or R4, the formation of Q appears kinetically favored
over that of P. Moreover, TSNQ (R3 =OMe) is relatively
large, providing a tentative explanation for the emergence
of the alkoxy groups at the termini of the product trienes.
Similarly, the trajectory for R1 =OMe, leading to the trans
stereochemistry observed, appears slightly favored. Evident-
ly, however, structural factors must influence this outcome
strongly, as indicated by the results of Scheme 4. These pro-
cesses do not stop at a cobalt hydride of the type shown in
and b-hydride elimination pathways, and rule out alternative
vinyl hydrogen activation routes toward the triene products.
Experimental Section
General methods: Reactions were carried out under argon. Except for
the [2+2+2] cycloaddition reactions involving III, which were carried
out without pretreatment, solvents were purified as follows: CH2Cl2 was
distilled from calcium hydride, toluene and xylenes from NaK2.8, and
THF and Et2O from sodium benzophenone ketyl. Solvents were degassed
(Ar) prior to the use of [CpCo(CO)2] and [CpCo
ACHTUNGTRENNUNG
was purchased from Aldrich and used as received. [CpCoAHCTUNGTRENNUNG
prepared according to the known procedure.[13] Reaction mixtures em-
ploying [CpCo(CO)2] were irradiated (visible light) by using a 300 W hal-
ogen lamp. Microwave heating was carried out in a Biotage microwave
synthesizer (400 W max). Thin layer chromatography (TLC) was per-
formed on Merck 60 F254 silica gel. Merck Gerudan SI 60 ꢂ silica gel
(35–70 mm) was used for column chromatography. 1H, 19F, 11B, and
13C NMR spectra were recorded at 208C at 400, 377, 128, and 100 MHz,
respectively, on a Bruker AVANCE400 spectrometer. 13C NMR signals of
carbon atoms attached to boron are very broad and were not observed.
Chemical shifts (d) are given in ppm, for 1H NMR referenced to the re-
sidual proton resonance of the solvents (d=7.26 for CDCl3; d=7.16 for
C6D6), and for 13C NMR to the corresponding carbon resonance (d=
77.16 for CDCl3; d=128.06 for C6D6). Coupling constants (J) are given
in Hertz (Hz). The terms m, s, d, t, q, and quin refer to multiplet, singlet,
doublet, triplet, quartet, quintet, respectively; br stands for broad. When
possible, NMR signals were assigned on the basis of NOE, DEPT (Dis-
tortionless Enhancement by Polarization Transfer), and 2D-NMR
(COSY, HMBC) experiments. Infrared spectra (IR) were recorded on a
Bruker Tensor 27 spectrometer. High resolution mass spectral data
(HRMS) were obtained by ESI. Melting points were measured with an
SMP3 Stuart Scientific melting point apparatus and were not corrected.
Starting materials were either commercial or prepared according to the
literature.[29]
b
Scheme 6, but the H atoms are transferred directly to the
a’ carbons to give cobaltacyclopentenes of type S. Reductive
elimination of these species provides the triene complexes
T.
Overall, the calculations suggest that trienes and cyclo-
hexadienes share a common intermediate, and that the for-
mation of trienes follows the b-hydride elimination path-
way 1 depicted in Scheme 6. This result is in consonance
with those obtained for the isoelectronic CpRu systems.
Conclusion
Procedure for preparation of products 3, 3’, 14, and 14’ (Tables 1 and 2):
[CpCoACTHUNRGTNE(NUG C2H4)2] (360 mg, 2 mmol) was dissolved in THF (8 mL). The enol
Long overdue, the competitive occurrence of b-hydride
elimination from a cobaltacycloheptadiene in the co-oligo-
merization of alkynes with alkenes by stoichiometric CpCo
has been established experimentally. It surfaces when the
alkene is an enol ether with certain alkadiyne and -monoyne
substrates. In these cases it is generally completely chemo-,
regio-, and stereospecific, with the resulting free alkoxy-
ACHTUNGTRENNUNGhexaACHTUNGTRENNUNGtrienes emerging exclusively with a terminal ether func-
tion in the trans configuration after column chromatography.
Particularly promising synthetically are cases involving bo-
ACHTUNGTRENNUNGrylACHTUNGTRENNUNGated alkynes, which furnish borylated alkoxytrienes. The
potential further generality of this reaction channel is indi-
cated by the observation that ethene itself can be a co-oligo-
merization partner. The trienes undergo 6p-electrocycliza-
tion followed by dehydroalkoxylation, giving rise to benzene
derivatives formally arising from [2+2+2] cycloaddition of
alkynes to acetylene. The same type of arene is accessed
from the regular cocyclization channel by oxidative demetal-
lation–dehydroalkoxylation of the resulting alkoxycyclohex-
adiene complexes. By using high-temperature conditions for
the co-oligomerization reaction, both pathways converge to
the same arene products, providing a one-step protocol for
using enol ethers as acetylene surrogates. DFT calculations
confirm the energetic closeness of the [2+2+2] reductive
ether (5 equiv) was added neat and the mixture was allowed to stir for
4 h at RT. The alkyne (4 mmol) or the diyne (2 mmol) in THF (2 mL)
was transferred by using a cannula into the mixture at À408C. The cold
bath was removed and the mixture was stirred at room temperature for
4 h. After evaporation of the solvent, the residue was purified with flash
column chromatography by using gradient mixtures of pentane and dieth-
yl ether.
Procedure for preparation of products 2, 14, and 16’ (Tables 1 and 2):
[CpCo
alkyne (4 mmol) or diyne (2 mmol) were reacted as above. The solvent
was removed with rotary evaporator. Acetonitrile (25 mL) and
ACHTNUGTNER(UNGN C2H4)2] (360 mg, 2 mmol), the enol ether (5 equiv), and the mono-
a
FeCl3·6H2O (810 mg, 3 mmol) were added in one portion at RT. The red
color turned violet instantaneously. The mixture was promptly evaporat-
ed and the residue was submitted to flash column chromatography over
silica gel by using gradient mixtures of pentane and diethyl ether.
Procedure for direct arene formation from diynes and tert-butyl vinyl
ether (Table 3, entry 9, and Table 4): In a microwave vial, a solution of
diyne (0.68 mmol), tert-butyl vinyl ether (3.4 mmol), and III (0.06 mmol)
in DMF (4 mL) was heated to 2008C for 20 min. After cooling to RT,
water and Et2O were added, and the aqueous layer was extracted with
Et2O. The combined organic extracts were washed with water, dried over
MgSO4, and concentrated. The residue was purified by flash column
chromatography by using gradient mixtures of pentane and diethyl ether.
2k: Colorless oil; 1H NMR (CDCl3): d=1.40 (t, J=7.2 Hz, 6H), 2.09
(quin, J=7.2 Hz, 2H), 3.29 (t, J=7.6 Hz, 4H), 4.37 (q, J=7.1 Hz, 4H),
7.85 ppm (s, 2H); 13C NMR (CDCl3): d=14.4 (CH3), 24.8 (CH2), 33.7
(CH2), 61.1 (CH2), 128.0 (CH), 130.1 (C), 148.3 (C), 166.8 ppm (C); IR
(neat): n˜max =1025, 1124, 1268, 1713, 2978 cmÀ1
;
HRMS: calcd for
C15H18O4Na: 285.1097; found: 285.1096.
Chem. Eur. J. 2010, 16, 8904 – 8913
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8911

C. Aubert, V. Gandon et al.
2o: Yellow oil; 1H NMR (CDCl3): d=1.33 (s, 24H), 1.68–1.81 (m, 2H),
2.12–2.27 (m, 4H), 3.08–3.19 (m, 4H), 7.47 ppm (s, 2H); 13C NMR
(CDCl3): d=22.3 (CH2), 24.8 (CH3), 27.7 (CH2), 33.1 (CH2), 83.4 (C),
131.8 (CH), 149.2 ppm (C), C-B not observed; IR (neat): n˜max =1108,
(C), 125.7 (CH), 131.1 (CH), 135.9 (CH), 144.2 (C), 148.7 ppm (C), C-B
not observed; 11B NMR (CDCl3): d=32 ppm; IR (neat): n˜max =1209,
1349, 1612, 1679, 3078 cmÀ1; HRMS: calcd for C24H43BO3Na: 413.3197;
found: 413.3194.
1212, 1270, 1315, 1369, 2858, 2930, 2975 cmÀ1
; HRMS: calcd for
15’d: Yellow oil; 1H NMR (CDCl3): d=0.88 (d, J=8.3 Hz, 6H), 0.91 (d,
J=8.3 Hz, 6H), 1.28 (s, 12H), 1.60–1.70 (m, 1H), 1.80–1.90 (m, 1H), 2.09
(d, J=7.5 Hz, 2H), 2.37 (d, J=7.5 Hz, 2H), 5.35 (s, 1H), 5.70 (s, 1H),
6.53 (d, J=12.9 Hz, 1H), 6.80 (d, J=12.9 Hz, 1H), 6.97–7.03 (m, 2H),
7.06–7.12 (m, 1H), 7.33 ppm (t, J=9.2 Hz, 2H); 13C NMR (CDCl3): d=
22.4 (CH3), 24.5 (CH3), 24.8 (CH3), 30.2 (CH), 30.3 (CH), 45.3 (CH2),
45.4 (CH2), 83.0 (C), 115.3 (CH), 120.1 (CH), 126.4 (CH), 127.7 (CH),
129.5 (CH), 130.0 (CH), 141.3 (C), 155.5 ppm (C), C-B not observed; IR
C23H36B2O4Na: 421.2697; found: 421.2700.
1
3b: Yellow oil; H NMR (CDCl3): d=1.18 (s, 9H), 1.66 (quin, J=7.5 Hz,
2H), 2.16 (t, J=7.5 Hz, 2H), 2.73 (t, J=7.5 Hz, 2H), 6.13 (d, J=12.2 Hz,
1H), 6.68 (d, J=12.2 Hz, 1H), 6.91 (s, 1H), 7.21–7.30 (m, 4H), 7.34–
7.42 ppm (m, 6H); 13C NMR (CDCl3): d=23.5 (CH2), 28.1 (CH3), 32.3
(CH2), 32.4 (CH2), 84.4 (C), 113.2 (CH), 126.0 (CH), 126.1 (CH), 126.5
(CH), 128.0 (CH), 128.1 (CH), 128.9 (CH), 129.7 (CH), 133.1 (C), 137.8
(neat): n˜max =1216, 1269, 1368, 1381, 1500, 1607, 1718, 2869, 2928 cmÀ1
.
(C), 139.0 (C), 142.7 (C), 143.6 (C), 147.0 ppm (CH); IR (neat): n˜max
=
1156, 1628, 3054 cmÀ1; HRMS: calcd for C25H28ONa: 367.2024; found:
16’a: Yellow oil; 1H NMR (CDCl3): d=0.70 (s, 12H), 1.21 (s, 12H), 5.71
(s, 1H), 6.42 (d, J=7.7 Hz, 1H), 6.83–6.89 (m, 3H), 6.99–7.04 (m, 1H),
7.23–7.37 ppm (m, 12H); 13C NMR (CDCl3): d=24.0 (CH3), 24.7 (CH3),
82.9 (C), 83.0 (C), 115.4 (CH), 116.9 (CH), 123.1 (CH), 127.3 (CH), 127.6
(CH), 128.0 (CH), 129.3 (CH), 129.5 (CH), 129.6 (CH), 129.9 (CH),
141.8 (C), 141.9 (C), 146.8 (C), 149.6 (CH), 156.9 (C), 159.5 ppm (C), C-
367.2034.
iso-3b: Yellow oil; 1H NMR (CDCl3): d=1.15 (s, 9H), 1.77 (quin, J=
7.3 Hz, 2H), 2.64 (t, J=7.3 Hz, 2H), 2.69 (t, J=7.3 Hz, 2H), 5.48 (s, 1H),
5.96 (d, J=11.8 Hz, 1H), 6.22 (d, J=11.8 Hz, 1H), 6.93–7.06 (m, 3H),
7.15–7.22 (m, 3H), 7.26–7.39 ppm (m, 4H); 13C NMR (CDCl3): d=24.1
(CH2), 28.1 (CH3), 32.1 (CH2), 34.0 (CH2), 84.1 (C), 113.7 (CH), 125.0
(CH), 125.7 (CH), 126.8 (CH), 127.9 (CH), 128.7 (CH), 128.9 (CH),
129.8 (CH), 132.4 (C), 135.2 (C), 139.3 (C), 141.0 (C), 142.7 (C),
B not observed; IR (neat): n˜max =1143, 1264, 1421, 1735, 2927, 3054 cmÀ1
;
HRMS: calcd for C36H42B2O5Na: 599.3116; found: 599.3118.
146.2 ppm (CH); IR (neat): n˜max =1155, 1625, 3022 cmÀ1
.
1
3d: Yellow oil; H NMR (CDCl3): d=1.26 (s, 9H), 1.27 (s, 12H), 1.33 (s,
12H), 1.62–1.67 (m, 2H), 2.50 (t, J=7.1 Hz, 2H), 2.64 (t, J=7.9 Hz, 2H),
5.75 (s, 1H), 6.43 (d, J=12.9 Hz, 1H), 6.86 ppm (d, J=12.9 Hz, 1H);
13C NMR (CDCl3): d=23.6 (CH2), 24.9 (2 C, CH3), 28.1 (CH3), 34.2
(CH2), 34.5 (CH2), 76.5 (C), 82.6 (C), 83.3 (C), 111.3 (CH), 146.1 (CH),
147.2 (C), 163.7 ppm (C), C-B not observed; IR (neat): n˜max =1111, 1164,
1264, 1426, 1732, 2930, 3053 cmÀ1; HRMS: calcd for C25H42B2O5Na:
461.3116; found: 467.3118.
Acknowledgements
We thank UPMC, UPS, the Ministꢃre de la Recherche, CNRS, IUF, and
the NSF (CHE 0907800, K.P.C.V.) for financial support of this work.
Technical assistance (MS) was generously offered by ICSN (Gif sur
Yvette, France). We also thank Pierre Garcia for technical assistance
during the synthesis of [CpCoACHTUNRGTNE(UGN C2H4)2]. Calculations were performed at
CRIHAN, plan interrꢄgional du bassin parisien (project 2006-013).
3’d: Yellow oil; 1H NMR (CDCl3): d=1.28 (s, 12H), 1.33 (s, 12H), 1.65
(quin, J=7.9 Hz, 2H), 2.60 (t, J=7.9 Hz, 2H), 2.67 (dt, J=7.9, 2.1 Hz,
2H), 5.75 (br s, 1H), 6.71 (d, J=12.9 Hz, 1H), 6.98–7.05 (m, 4H), 7.27–
7.34 ppm (m, 2H); 13C NMR (CDCl3): d=21.1 (CH2), 24.5 (CH3), 24.9
(CH3), 32.7 (CH2), 33.9 (CH2), 83.2 (C), 83.4 (C), 115.3 (CH), 120.5
(CH), 125.8 (CH), 129.6 (CH), 133.0 (CH), 143.5 (C), 150.7 (C),
169.5 ppm (C), C-B not observed; IR (neat): n˜max =1120, 1264, 1357,
[1] S. Koo in Science of Synthesis: Houben-Weyl Methods of Molecular
Transformations, Vol. 45b (Eds.: J. Siegel, Y. Tobe), Thieme, Stutt-
gart, 2010, pp. 1349–1420.
[2] For a recent topical paper, see: L. M. Bishop, J. E. Barbarow, R. G.
Bergman, D. Trauner, Angew. Chem. 2008, 120, 8220; Angew. Chem.
Int. Ed. 2008, 47, 8100.
[3] T. L. Cairns, V. A. Engelhardt, H. L. Jackson, G. H. Kalb, J. C.
Sauer, J. Am. Chem. Soc. 1952, 74, 5636. For a related paper, see:
G. N. Schrauzer, Chem. Ber. 1961, 94, 1403.
[4] T. Sambaiah, L.-P. Li, D.-J. Huang, C.-H. Lin, D. K. Rayabarapu, C.-
H. Cheng, J. Org. Chem. 1999, 64, 3663.
1717, 2980 cmÀ1
.
1
3g: Yellow oil; H NMR (CDCl3): d=1.26 (s, 12H), 1.27 (s, 9H), 1.32 (s,
12H), 2.39 (s, 2H), 2.53 (s, 2H), 3.21 (s, 4H), 3.31 (s, 6H), 5.74 (s, 1H),
6.39 (d, J=12.9 Hz, 1H), 6.86 ppm (d, J=12.9 Hz, 1H); 13C NMR
(CDCl3): d=24.9 (CH3), 25.0 (CH3), 28.1 (CH3), 39.7 (CH2), 39.8 (CH2),
44.3 (C), 59.3 (CH3), 75.8 (CH2), 76.6 (C), 82.7 (C), 83.4 (C), 111.1 (CH),
145.8 (C), 146.2 (CH), 161.9 ppm (C), C-B not observed; IR (neat):
n˜max =1107, 1371, 1457, 1716, 2924 cmÀ1; HRMS: calcd for C30H50B2O7Na:
567.3640; found: 567.3642.
[5] C. S. Yi, J. R. Torres-Lubian, N. Liu, A. L. Rheingold, I. A. Guzei,
Organometallics 1998, 17, 1257.
[6] a) J. A. Varela, S. G. Rubꢅn, C. Gonzꢁlez-Rodrꢅguez, L. Castedo, C.
Saꢁ, J. Am. Chem. Soc. 2006, 128, 9262; b) J. A. Varela, S. G. Rubꢅn,
C. Gonzꢁlez-Rodrꢅguez, L. Castedo, C. Saꢁ, J. Org. Chem. 2008, 73,
1320; c) S. Garcꢅa-Rubꢅn, J. A. Varela, L. Castedo, C. Saꢁ, Chem.
Eur. J. 2008, 14, 9772; d) S. Garcꢅa-Rubꢅn, J. A. Varela, L. Castedo,
C. Saꢁ, Org. Lett. 2009, 11, 983.
3’g: Yellow oil; 1H NMR (CDCl3): d=1.27 (s, 12H), 1.32 (s, 12H), 2.50
(s, 2H), 2.57 (s, 2H), 3.23 (s, 4H), 3.28 (s, 6H), 5.74 (s, 1H), 6.67 (d, J=
12.9 Hz, 1H), 6.98–7.03 (m, 3H), 7.27–7.34 ppm (m, 3H); 13C NMR
(CDCl3): d=24.5 (CH3), 24.9 (CH3), 38.6 (CH2), 40.2 (CH2), 47.4 (C),
59.3 (CH3), 76.3 (CH2), 83.2 (C), 83.4 (C), 115.3 (CH), 120.5 (CH), 125.6
(CH), 127.7 (CH), 129.6 (CH), 133.4 (CH), 141.4 (C), 149.2 (C),
155.7 ppm (C), C-B not observed.
[7] For reviews on [2+2+2] cycloaddition reactions, see: a) K. P. C.
Vollhardt, Angew. Chem. 1984, 96, 525; Angew. Chem. Int. Ed. Engl.
1984, 23, 539; b) J. Varela, C. Saꢁ, Chem. Rev. 2003, 103, 3787; c) Y.
Yamamoto, Curr. Org. Chem. 2005, 9, 503; d) V. Gandon, C. Aubert,
M. Malacria, Curr. Org. Chem. 2005, 9, 1699; e) S. Kotha, E. Brah-
machary, K. Lahiri, Eur. J. Org. Chem. 2005, 4741; f) V. Gandon, C.
Aubert, M. Malacria, Chem. Commun. 2006, 2209; g) P. R. Chopade,
J. Louie, Adv. Synth. Catal. 2006, 348, 2307; h) N. Agenet, V.
Gandon, O. Buisine, F. Slowinski, C. Aubert, M. Malacria in Organic
Reactions, Vol. 68 (Ed.: T. V. RajanBabu), Wiley, New York, 2007,
pp. 1–302; i) B. Heller, M. Hapke, Chem. Soc. Rev. 2007, 36, 1085;
j) K. Tanaka, Synlett 2007, 1977; k) B. E. Maryanoff, H.-C. Zhang,
ARKIVOC 2007, 12, 7; l) J. Varela, C. Saꢁ, Synlett 2008, 2571; m) K.
Tanaka, Chem. Asian J. 2008, 3, 508; n) W. Hess, J. Treutwein, G.
Hilt, Synthesis 2008, 3537; o) T. Shibata, K. Tsuchikama, Org.
1
14a: Yellow oil; H NMR (CDCl3): d=0.79 (s, 12H), 1.10 (s, 12H), 7.25–
7.52 (m, 11H), 7.72 ppm (d, J=7.9 Hz, 1H); 13C NMR (CDCl3): d=24.5
(CH3), 24.6 (CH3), 83.3 (C), 83.5 (C), 126.3 (CH), 126.6 (CH), 127.0
(CH), 127.9 (CH), 129.2 (CH), 130.2 (CH), 134.2 (CH), 143.3 (C), 143.8
(C), 147.4 (C) 151.5 ppm (C), 1 CH peak is accidentally isochronous with
another, C-B not observed; IR (neat): n˜max =1140, 1267, 2926, 2978 cmÀ1
;
HRMS: calcd for C30H36B2O4Na: 505.2697; found: 505.2701.
15d: Colorless oil; 1H NMR (CDCl3): d=0.85 (d, J=6.8 Hz, 6H), 0.88
(d, J=6.8 Hz, 6H), 1.26 (s, 9H), 1.29 (s, 12H), 1.68 (non, J=6.8 Hz, 1H),
1.82 (non, J=6.8 Hz, 1H), 2.02 (d, J=7.2 Hz, 2H), 2.33 (d, J=7.2 Hz,
2H), 5.33 (s, 1H), 5.50 (s, 1H), 6.15 (d, J=12.4 Hz, 1H), 6.65 ppm (d, J=
12.4 Hz, 1H); 13C NMR (CDCl3): d=22.5 (CH3), 22.6 (CH3), 24.6 (CH3),
24.9 (CH3), 27.2 (CH), 27.3 (CH), 44.9 (CH2), 45.6 (CH2), 80.2 (C), 83.2
8912
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8904 – 8913

Co-oligomerization of Alkynes with Enol Ethers
FULL PAPER
Biomol. Chem. 2008, 6, 1317; p) C. J. Scheuermann nꢄe Taylor, B. D.
Ward, New J. Chem. 2008, 32, 1850; q) B. R. Galan, T. Rovis,
Angew. Chem. 2009, 121, 2870; Angew. Chem. Int. Ed. 2009, 48,
2830.
[19] We have noted previously that the regioselectivity of the alkyne
coupling (1,3 vs. 1,4) is critically (and so far unpredictably) depen-
dent on the nature of the enol ether, see reference [12].
[20] For recent work, see: E. Khana, B. Wrackmeyer, Z. Naturforsch. B
2009, 64, 1098.
[21] A. Geny, N. Agenet, L. Iannazzo, M. Malacria, C. Aubert, V.
Gandon, Angew. Chem. 2009, 121, 1842; Angew. Chem. Int. Ed.
2009, 48, 1810.
[22] J. Oxgaard, R. P. Muller, W. A. Goddard III, R. A. Periana, J. Am.
Chem. Soc. 2004, 126, 352.
[23] Gaussian 03, Revision B.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford CT, 2004.
[8] See footnote 20 in: Y. Shibata, Y. Otake, M. Hirano, K. Tanaka,
Org. Lett. 2009, 11, 689.
[9] Pyridones: a) C. Aubert, P. Betschmann, M. J. Eichberg, V. Gandon,
T. J. Heckrodt, J. Lehmann, M. Malacria, B. Masjost, E. Paredes,
K. P. C. Vollhardt, G. D. Whitener, Chem. Eur. J. 2007, 13, 7443; in-
doles: b) R. Boese, A. P. Van Sickle, K. P. C. Vollhardt, Synthesis
1994, 1374; c) S. Amslinger, C. Aubert, V. Gandon, M. Malacria, E.
Paredes, K. P. C. Vollhardt, Synlett 2008, 2056; d) pyrimidines: H.
Pelissier, J. Rodriguez, K. P. C. Vollhardt, Chem. Eur. J. 1999, 5,
3549; e) furans and thiophenes: R. Boese, D. F. Harvey, M. J. Malas-
ka, K. P. C. Vollhardt, J. Am. Chem. Soc. 1994, 116, 11153.
[10] a) V. Gandon, N. Agenet, K. P. C. Vollhardt, M. Malacria, C.
Aubert, J. Am. Chem. Soc. 2006, 128, 8509; b) C. Aubert, V.
Gandon, A. Geny, T. J. Heckrodt, M. Malacria, E. Paredes, K. P. C.
Vollhardt, Chem. Eur. J. 2007, 13, 7466.
[11] Because these arenes are formed by the elimination of the alkoxy
group of the enol ether after cyclohexadiene formation, the alkene
functions as a substitute for acetylene. For the use of liquid enol
ethers as acetylene equivalents in [2+2+2] cyclizations, see: rhodi-
um: a) H. Hara, M. Hirano, K. Tanaka, Org. Lett. 2008, 10, 2537;
b) H. Hara, M. Hirano, K. Tanaka, Tetrahedron 2009, 65, 5093; c) H.
Hara, M. Hirano, K. Tanaka, Org. Lett. 2009, 11, 1337; iridium: d) S.
Kezuka, S. Tanaka, T. Ohe, Y. Nakaya, R. Takeuchi, J. Org. Chem.
2006, 71, 543.
[12] a) A. Geny, D. Lebouf, G. Rouquiꢄ, K. P. C. Vollhardt, M. Malacria,
V. Gandon, C. Aubert, Chem. Eur. J. 2007, 13, 5408; b) V. Gandon,
D. Leboeuf, S. Amslinger, K. P. C. Vollhardt, M. Malacria, C.
Aubert, Angew. Chem. 2005, 117, 7276; Angew. Chem. Int. Ed. 2005,
44, 7114.
[13] a) K. Jonas, E. Deffense, D. Habermann, Angew. Chem. 1983, 95,
729; Angew. Chem. Int. Ed. Engl. 1983, 22, 716; For a review, see:
b) V. Gandon, C. Aubert, e-EROS 2008, DOI: 10.1002/
047084289X.rn00943.pub2.
[14] For the use of borylated alkynes in [2+2+2] cycloadditions, see ref-
erence [9c], [12] and, inter alia: a) Y. Yamamoto, J.-i. Ishii, H. Nish-
iyama, K. Itoh, J. Am. Chem. Soc. 2004, 126, 3712; b) A. Goswami,
C.-J. Maier, H. Pritzkow, W. Siebert, Eur. J. Inorg. Chem. 2004,
2635; c) V. Gandon, D. Leca, T. Aechtner, K. P. C. Vollhardt, M.
Malacria, C. Aubert, Org. Lett. 2004, 6, 3405; d) Y. Yamamoto, J.-i.
Ishii, H. Nishiyama, K. Itoh, J. Am. Chem. Soc. 2005, 127, 9625;
e) Y. Yamamoto, K. Hattori, J.-i. Ishii, H. Nishiyama, Tetrahedron
2006, 62, 4294.
[15] For NMR assignments of enol ethers, see: M. P. Strobel, C. G. An-
drieu, D. Paquer, M. Vazeux, C. C. Pham, Nouv. J. Chim. 1980, 4,
101.
[16] For diborylated trienes, see: a) M. Shimizu, T. Kurahashi, T.
Hiyama, Synlett 2001, 1006; b) B. Wrackmeyer, W. Milius, O. L. Tok,
Y. N. Bubnov, Chem. Eur. J. 2002, 8, 1537.
[24] P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299.
[25] a) R. Ditchfield, W. J. Hehre, J. A. Pople, J. Chem. Phys. 1971, 54,
724; b) W. J. Hehre, R. Ditchfield, J. A. Pople, J. Chem. Phys. 1972,
56, 2257; c) P. C. Hariharan, J. A. Pople, Theo. Chim. Acta 1973, 28,
213; d) P. C. Hariharan, J. A. Pople, Mol. Phys. 1974, 27, 209;
e) M. S. Gordon, Chem. Phys. Lett. 1980, 76, 163.
[26] N. Agenet, V. Gandon, K. P. C. Vollhardt, M. Malacria, C. Aubert, J.
Am. Chem. Soc. 2007, 129, 8860.
[27] As the crucial steps of the mechanism involve 18-electron species,
no attempt to locate the crossing points of the singlet and triplet sur-
faces was made. For a more sophisticated treatment, see reference-
s [10a] and [26].
[28] Because the alkene is complexed unsymmetrically to the cobaltacy-
clopentadiene, the metal in these complexes is stereogenic. For a
discussion, see reference [10b].
[29] The borylated diynes were prepared according to reference [12];
Other diynes were prepared as follows: 1b and 1m: a) B. L. Lucht,
S. S. H. Mao, T. D. Tilley, J. Am. Chem. Soc. 1998, 120, 4354; 1c:
b) P. C. Auderset, A. S. Dreiding, E. R. F. Gesing, Synth. Commun.
1983, 13, 881; 1e: c) S. U. Son, Y. K. Chung, J. Org. Chem. 2000, 65,
6142; 1h and 1r: d) J.-J. Lian, P.-C. Chen, Y.-P. Lin, H.-C. Ting, R.-S.
Liu, J. Am. Chem. Soc. 2006, 128, 11372; 1i: e) T. Kudoh, T. Mori,
M. Shirahama, M. Yamada, T. Ishikawa, S. Saito, H. Kobayashi, J.
Am. Chem. Soc. 2007, 129, 4939; 1k: f) Y. Matano, T. Miyajima, T.
Nakabuchi, Y. Matsutani, H. Imahori, J. Org. Chem. 2006, 71, 5792;
1p: g) N. Cadran, K. Cariou, G. Hervꢄ, C. Aubert, L. Fensterbank,
M. Malacria, J. Marco-Contelles, J. Am. Chem. Soc. 2004, 126, 3408;
1 s: P. Wessig, G. Mueller, C. Pick, A. Matthes, Synthesis 2007, 464.
[17] For reviews, see: a) N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95,
2457; b) N. Miyaura, Top. Curr. Chem. 2002, 219, 11; c) N. Miyaura
in Metal-Catalyzed Cross-Coupling Reactions, Vol. 1 (Eds.: A. de
Meijere, F. Diederich), Wiley-VCH, Weinheim, 2004, pp. 41–124.
[18] J. A. King, Jr., K. P. C. Vollhardt, J. Organomet. Chem. 1989, 369,
245.
Received: February 24, 2010
Published online: June 23, 2010
Chem. Eur. J. 2010, 16, 8904 – 8913
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8913