Expanding the scope of arylsulfonylacetylenes as alkynylating reagents and mechanistic insights in the formation of Csp2-Csp and Csp 3-Csp bonds from organolithiums

Article abstract of DOI:10.1002/chem.201200939

We describe the unexpected behavior of the arylsulfonylacetylenes, which suffer an "anti-Michael" addition of organolithiums producing their alkynylation under very mild conditions. The broad scope, excellent yields, and simplicity of the experimental procedure are the main features of this methodology. A rational explanation of all these results can be achieved by theoretical calculations, which suggest that the association of the organolithiums to the electrophile is a previous step of their intramolecular attack and is responsible for the unexpected "anti-Michael" reactions observed for substituted sulfonylacetylenes. A calculated conclusion: A new transition-metal-free strategy for the synthesis of any kind of alkynyl derivatives in high yields in the reaction of organolithium species with arylsulfonylacetylenes is presented (see scheme). Theoretical calculations provide a rational explanation and suggest that association of the organolithium to the electrophile is a previous step of their intramolecular attack and is responsible for the "anti-Michael" reaction. Copyright

Full text of DOI:10.1002/chem.201200939

DOI: 10.1002/chem.201200939
Expanding the Scope of Arylsulfonylacetylenes as Alkynylating Reagents and
2
3
À
À
Mechanistic Insights in the Formation of Csp Csp and Csp Csp Bonds
from Organolithiums
Josꢀ Luis Garcꢁa Ruano,*[a] Josꢀ Alemꢂn,*[a] Leyre Marzo,[a] Cuauhtꢀmoc Alvarado,[a]
Mariola Tortosa,[a] Sergio Dꢁaz-Tendero,[b] and Alberto Fraile[a]
In memory of Alessandro DeglꢀInnocenti
Abstract: We describe the unexpected behavior of the arylsulfonylacetylenes,
which suffer an “anti-Michael” addition of organolithiums producing their alkyn-
A
Keywords: alkynylation · anti-Mi-
chael reactions · density functional
calculations · organolithiums · sulfo-
nylacetylenes
ty of the experimental procedure are the main features of this methodology. A ra-
tional explanation of all these results can be achieved by theoretical calculations,
which suggest that the association of the organolithiums to the electrophile is a pre-
vious step of their intramolecular attack and is responsible for the unexpected
“anti-Michael” reactions observed for substituted sulfonylacetylenes.
Introduction
report, Truce describes the synthesis, under very mild condi-
tions, of only seven compounds with the alkyne moiety
joined to aryl (phenyl and p-tolyl) and alkyl (nBu and tBu)
residues. Despite the potential interest, to our knowledge,
other contributions based on this work were never reported.
Although this work has been cited only few times in some
reviews,[4] this reaction has been considered rather as a scien-
tific curiosity, scarcely used with synthetic[5] or mechanistic[6]
In recent years, acetylene chemistry has become an increas-
ingly attractive topic for chemists due to its importance in
the synthesis of bioactive natural products and new materi-
als as well as in biochemistry.[1] We have recently reported[2]
one of the most efficient methods for incorporating acety-
lenic fragments to Csp2 consisting in the reaction of com-
À
pounds containing activated aromatic C H with lithium
purACHTNGUTERNNUpG oses in the 1990s and has been inexplicably forgotten.
bases and arysulfonylacetylenes. This work was based on
a reaction found by Smorada and Truce in 1979,[3] which de-
scribed the reactivity of commercial organolithiums or
After considering the scope and limitations of the main
alkynylation methods currently used for the construction of
2
3
À
À
the Csp Csp and Csp Csp bonds (see below), we realized
that the reaction conditions were much milder than those
usually required by the current methods. This, in addition to
the advantages of using lithium instead of transition metals
suggested us that this reaction could open a new entry for
synthesizing alkynyl derivatives whenever its scope was
wide enough. Herein, we decided to explore the possibilities
of this reactivity as a general alkynylating method, in which
are also indicated the efforts for clarifying the reasons of
the unexpected “anti-Michael” addition of the sulfonylacety-
lenes in the presence of organolithiums.
organoACHTUNGTRENNUNGmagnesium (less reactive) with arylsulfonylacety-
lenes, to give alkynes through addition followed by elimina-
tion of the sulfone group (Scheme 1). In this very short
Scheme 1. Work reported by Smorada and Truce.[3]
[a] Prof. Dr. J. L. Garcꢀa Ruano, Dr. J. Alemꢁn, L. Marzo,
Dr. C. Alvarado, Dr. M. Tortosa, Dr. A. Fraile
Departamento de Quꢀmica Orgꢁnica (Mꢂdulo 1)
Universidad Autꢂnoma de Madrid
The main alkynylation methods currently used are collect-
ed in Scheme 2. The reaction of metal-acetylides with alkyl
halides[7] (method a) is typically used for the creation of
alkyl-Csp bonds, but it works satisfactorily only with pri-
mary alkyl halides,[8] mainly due to its competition with
elimination reactions.[9] The addition of acetylides to C=O
and C=N bonds[10] followed by substitution of the C-hetero-
Fax : (+) 3491497466
E-mail: joseluis.garcia.ruano@uam.es
jose.aleman@uam.es
[b] Dr. S. Dꢀaz-Tendero
Departamento de Quꢀmica (Mꢂdulo 13)
Universidad Autꢂnoma de Madrid
Cantoblanco, 28049-Madrid (Spain)
À
À
AHCTUNGTREGaNNUN tom bond by C H (reduction) or C C bonds on the result-
ing heterosubstituted propargylic derivatives, constitutes an
indirect method for preparing sec-alkyl- and tert-alkyl-
AHCTUNGTREGaNNUN lkynes. As they often require long synthetic sequences, the
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201200939.
8414
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8414 – 8422

FULL PAPER
Results and Discussion
We initiated our study with the reaction of 2-para-tolylsul-
fonyl acetylene (1a) with nBuLi (2A). The reaction was car-
ried out at À788C, and “anti-Michael” product 4aA was ob-
tained with complete control of the regioselectivity (entry 1,
Table 1). To determine the influence of the steric bulk of the
alkyllithium, we studied the reactions with smaller (2B,
entry 2) and larger (2C and 2D, entries 3 and 4, Table 1) nu-
cleophiles. In all the cases, only the “anti-Michael” product
was formed in excellent yield, regardless the steric hin-
drance of the reagent (4aB, 4aC, and 4aD). These results
were particularly interesting given that the synthesis of
phenACTHNUGRTENNUGylACHUTGTNRENaNUG cetylenes bearing a tertiary or quaternary carbon is
particularly challenging using existing methods (see above).
3
À
Scheme 2. Main alkynylation methods currently used for preparing Csp
2
À
Csp and Csp Csp bonds.
3
À
Table 1. Formation of Csp Csp bonds by reaction of alkyllithiums with
2-substituted p-tolylsulfonylacetylenes.[a]
creation of these bonds under more satisfactory conditions
remains currently as a challenge.
Method b of Scheme 2 illustrates the three most typical
Sulfone (R1)
Reagent (R2)
Product
[%][b]
2
Entry
À
routes for preparing Csp Csp bonds (aryl-alkynes and con-
jugated enynes). The most general and widely used ap-
proach is the well-known Sonogashira cross-coupling reac-
tion (b-1, Scheme 2).[11] It starts from aryl or alkenyl halides
and terminal alkynes and requires the presence of a palladi-
um(0) catalyst and a copper source (co-catalyst). Although
the scope of this reaction is quite general, special conditions
are often required for couplings involving electron-rich Csp2
components,[12] which interfere with the oxidative addition
step, or electron-poor alkynes,[13] which do not readily form
the copper acetylide intermediates. A more recent strategy
for the formation of aryl-alkynes is the inverse-Sonogashira-
type reaction (b-2, Scheme 2)[14] in which the alkyne source
1
2
3
4
5
6
7
8
1a (Ph)
2 A (nBu)
2B (Me)
2C (sBu)
2D (tBu)
2 A (nBu)
2C (sBu)
2D (tBu)
2D (tBu)
2A (nBu)
2A (nBu)
2C (sBu)
2D (tBu)
2A (nBu)
2C (sBu)
2D (tBu)
4aA (95)
4aB (91)
4aC (85)
4aD (83)
4bA (90)
4bC (91)
4bD (96)
4cD (>99)[c]
4cA (>99)[c]
4dA (90)
4dC (90)
4dD (92)
4eA (60)
4eC (38)
1b (TIPS)
1c (tBu)
1d (Chx)
9
10
11
12
13
14
15
1e (nBu)
À
[d]
is typically a bromo-alkyne and aromatic C H activation is
–
achieved using palladium,[14b] nickel,[14a] or copper[14c] cata-
lysts. This method allows the synthesis of important pharma-
ceutical derivatives but is generally restricted to substrates
[a] All the reactions were performed with 1a–e (0.2 mmol) and 2A–D
(0.4 mmol). [b]Isolated product yield (%) after flash chromatography.
[c] Conversion measured by H NMR spectroscopy. [d] Only the Michael
product could be identified (20% yield) in the complex mixture.
1
À
susceptible to C H activation (azoles or activated aromatic
rings). A third, but even less general, approach is based on
a gold-catalyzed Friedel–Crafts type addition of highly elec-
tron-rich aromatic rings (e.g., 2,4,6-trimethoxyphenyl
groups,[15] thiophenes[16] or indoles[17]) to electron-poor ter-
minal alkynes or alkynyl-iodonium salts derivatives (b-3,
Scheme 2). In addition to the indicated structural limitations
and the price of the catalytic systems, the main handicap of
most of these reactions, which use Pd0 as the catalyst, de-
rives from the waste generated in these reactions, which se-
riously limits their use by the pharmaceutical industry.[18] It
is obvious that this procedure, based on organolithium re-
agents, does not have these drawbacks, which, in addition to
We then extended our study to a more versatile alkyne, 2-
TIPS-1-para-tolylsulfonyl acetylene (TIPS=triisopropylsilyl)
1b, which proved to be an excellent alkynylating agent. Al-
though this reaction may not appear to be atom economi-
cal,[19] the generated lithium-p-toluenesulfinate can be easily
recovered and used again for the synthesis of the ethynylsul-
fone.[20] Reaction with nBuLi (2A), sBuLi (2C) and tBuLi
(2D) (entries 5–7, Table 1) afforded the corresponding 1-
TIPS-hexynes (4bA, 4bC, and 4bD) in good yields. These
compounds can be easily desilylated with tetra-n-butyl-
AHCTUNGTREGaNNUN mmonium fluoride (TBAF), allowing the synthesis of
the milder conditions required and the high ability of the
a wide variety of terminal alkynes.
2
À
lithium for activating many Csp H bonds, confer it a great
To obtain 1,2-dialkyl acetylenes, we performed the reac-
tions of alkylsulfonylacetylenes 1c–e with tertiary, secon-
dary, and primary alkyllithiums 2D, 2C, and 2A (entries 8–
15, Table 1). Starting from 1c (entries 8 and 9, Table 1) di-
interest as a general alkynylating method.
ACHUTNGRENaNUG lkylCAHTUNGTRNEaNUGN lkynes 4cD, and 4cA were obtained in very high con-
Chem. Eur. J. 2012, 18, 8414 – 8422
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8415

J. L. Garcꢀa Ruano, J. Alemꢁn et al.
version, which is specially remarkable in the first case.[21] Di-
alkylalkynes 4dA, 4dC, and 4dD were obtained in excellent
yields. The volatility of some of these products accounts for
the moderate yields observed for 4eA and 4eC (entries 13
and 14, Table 1), since the conversion was almost quantita-
tive by 1H NMR spectroscopy. The only exception to this
general behavior was observed in the reaction of 1e (nBu)
with 2D (tBu), which did not afford the “anti-Michael”
product 4eD but a complex mixture. The only product that
we could isolate and characterize from this mixture was (E)-
1-phenylsulfonyl-2-tert-butyl-1-hexene (20% yield), resulting
from a Michael addition of tBuLi to the alkynylsulfone
(entry 15, Table 1). It should be noted that compound 4eD
is identical to 4cA (entry 9, Table 1), which is easily ob-
tained by reaction of nBuLi with 1c. In this context we must
remark that reactions of sulfonylacetylenes containing pro-
tions of electron-rich aryllithiums were interesting because
their use in the Sonogashira reaction provided poor results
in some cases (mainly with electron-poor alkynes[13,15]). We
were therefore pleased to find that the success of the reac-
tions of p-methoxyphenyllithium (2F) are not affected by
the incorporation to the aryl ring at the acetylenic moiety of
electron-donating nor electron-withdrawing groups (en-
tries 8–10, Table 2). Reaction of electron-poor aryllithium
2G with 1h is similarly successful (entry 11, Table 2), reveal-
ing that the electronic properties of the rings taking part in
the organolithium or in the arylacetylene are not essential
to the reaction. Finally, the alkynylation of pyridines[22,23]
can be also performed from 2-halo pyridines by lithium-
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
4aH and 4bH from 2-bromopyridine (entries 12 and 13,
ACHTUNGTRENNUNGpargylic hydrogen atoms, like compounds 1d and 1e, are
Table 2).[24]
successful except for entry 15 of Table 1, which contrasts
Our method could be also applied to the synthesis of
enynes, which are commonly found in naturally occurring
compounds.[25] Alkenyllithium 2I, easily obtained from (E)-
1-iodo-1-octene, underwent reactions with 1a and 1b to pro-
vide 4aI and 4bI, respectively, in quantitative yields and
with complete retention of the E configuration of the
double bond (Scheme 3, a).[26] Analogously, enyne 4aK, with
with the predictions from Truce[3] based on the acidity crite-
À
ria. We next investigated the formation of Aryl Csp bonds
(Table 2).
2
À
Table 2. Formation of Csp Csp bonds by reaction of aryl or alkenyllithi-
ums with 2-substituted p-tolylsulfonylacetylenes.[a]
Entry
Sulfone (R1)
Reagent (R2)
Product
[%][b]
1
2
1b (TIPS)
1c (tBu)
2E (Ph)
2E (Ph)
4bE (80)
4cE (81)
3
4
5
6
7
8
9
10
11
12
13
1e (nBu)
1a (Ph)
1 f (p-Tol)
1g (p-CF3-C6H4)
1h (3,5-di
1 f (p-Tol)
1a (Ph)
1h (3,5-di
1h (3,5-di
1a (Ph)
2E (Ph)
2E (Ph)
2E (Ph)
2E (Ph)
4eE (–[c]
4aE (98)
4 fE (99)
4gE (99)
4hE (89)
4 fF (60)
4aF (78)
4hF (88)
4hG (81)
4aH (95)
4bH (90)
)
(p-CF3)-C6H4)
2E (Ph)
2F (p-MeO-C6H4)
2F (p-MeO-C6H4)
2F (p-MeO-C6H4)
2G (p-CF3-C6H4)
2H (2-Py)
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
1b (TIPS)
2H (2-Py)
Scheme 3. Synthesis of enynes from vinyllithiums and sulfonylacetylenes.
[a] All the reactions were performed with 1a–h (0.2 mmol) and 2E–H
(0.4 mmol). [b] Isolated product yield (%) after flash chromatography.
[c] Complex mixture.
Z configuration at the double bond (Scheme 3, b), was the
only product observed in the reaction of the Li derivative of
(Z)-1-bromo-1-propene (2K) with 1a.[27] Finally, the 1,3-di-
First, we performed the reactions of PhLi (2E) with para-
toluenesulfonylacetylene 1b, containing a TIPS group which
is easily converted into the terminal acetylene (entry 1,
Table 2). Compound 4bE was obtained in 80% yield. Anal-
ogously, reaction of 2E with the tert-butylacetylene sulfone
1c afforded the alkylarylacetylene 4cE (81%, entry 2). The
behavior of PhLi with 1e (entry 3, Table 2) was similar to
that observed for tBuLi (entry 15, Table 1), giving a complex
mixture, in which the anti-Michael product 4eE could not
be detected. We then studied the reactions of 2E with 1a
and with sulfonyl arylacetylenes containing electron-donat-
ing (1 f) and electron-withdrawing groups (1g and 1h), with
excellent results in all the cases (entries 4–7, Table 2). Reac-
AHCTUNGTREGsNNUN ubstituted enyne 4aJ was obtained in 80% yield by reac-
tion of a-methylvinyl-lithium 2 J with 1a (Scheme 3, b).
Reactions of the organolithiums 2 resulting in ortho-lith-
AHCTUNGTRENNUNG
iation processes[28] from aromatic compounds bearing acti-
À
vated C H (5) were particularly interesting because they do
not require to start from the potentially costly or difficult to
access halo derivatives, nor from using Pd for the coupling
of the alkyne moieties with the aromatic residues[29]
(Scheme 4). Although the results obtained in this study
were reported before,[2] we have included here a summary
of them for having a complete idea about the scope of this
reaction.
8416
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8414 – 8422

Arylsulfonylacetylenes as Alkynylating Reagents
FULL PAPER
Treatment of 5O, bearing two OMe groups, with 1a exclu-
sively provided 4aO in 85% yield (Scheme 6). This yield
was increased up to 92%, working on a 4.0 mmol scale. The
2
À
Scheme 4. Direct alkynylation of activated Csp H.
We first studied the alkynylation of benzenes containing
ortho-directing groups (Table 3). Under the optimized con-
ditions, the lithium derivative 2L was obtained by ortho-lith-
Table 3. Alkynylations of anisole with different sulfones 1.[a]
Entry
Sulfone (R1)
Product
Yield
[%][b]
Scheme 6. Alkynylation of 5O and synthesis of 3-alkynylbenzofuranes.
1
2
3
4
5
1a (Ph)
1b (TIPS)
1 f (p-Tol)
1h [3,5-di
1i (o-ClC6H4)
4aL
4bL
4 fL
4hL
4iL
81
80
53
73
55
absence of dialkynylated products indicates that 4aO is less
reactive than 5O, as would be expected from the deactivat-
ing influence of the alkynyl groups on the formation of the
corresponding aryl-lithium. Starting from 4aO, a new exam-
ple of the synthesis of enynes is shown in Scheme 6. It has
been reported that ortho-alkynyl anisoles react with I2 to
afford 3-iodobenzofurans.[30] Under these conditions, com-
pound 4aO was transformed into 6, which reacted in situ
with nBuLi and then with the sulfonylacetylene 1b, yielding
3-alkynylbenzofurane 7 in 65% yield (two steps from 4aO).
The three-step sequence connecting 5O with 7 (Scheme 6)
illustrates an efficient method for the synthesis of 3-alkynyl-
benzofuranes, which involves two consecutive alkynylation
reactions.
The reaction of 5P was interesting because of its regio-
AHCTUNGTREGsNNNU electivity (Scheme 7, a). Working under the usual condi-
tions (ortho-lithiation at RT), the reaction affords an 80:20
mixture of 4aP and its regioisomer 4-alkynyl-1,3-dimethoxy-
benzene, which were easily separated by chromatography
(54% yield of 4aP). The lower susceptibility of 5Q to
ACHTUGNERTN(NUNG p-CF3)-C6H3]
[a] All the reactions were performed with 1a–b (0.2 mmol) and
(0.4 mmol). [b] Isolated product yield (%) after flash chromatography.
2
ACHTUNGTRENNUNGiation of anisole (5L) with nBuLi or tBuLi at RT. Reaction
of 2L with the sulfonyl acetylene 1a at À788C, afforded
4aL in 5 min (81% yield, Table 3, entry 1). Under similar
conditions, compound 5l reacted with 2b to afford alkynyl
derivative 4bL in 80% yield (Table 3, entry 2), which can be
used as the precursor for terminal alkynes and as an inter-
mediate in the preparation of many other alkynyl deriva-
tives. Moreover, the good results are independent on the
presence of electron-donating (1 f) and electron-withdrawing
groups (1h and 1i) at the aryl ring of the arylacetylenesul-
fones, thus enabling the direct preparation of 4 fL, 4hL and
4iL (Table 3, entries 3–5).
The use of other ortho-directing groups was also success-
ful. Thus, the N,N-diethylcarbamate 5M reacted with 1a
through 2M affording 4aM (99% yield) and the N,N-diiso-
propylcarboxamide 5N gave 4aN (86%) and 4bN (90%) in
their reactions with 1a and 1b, respectively (Scheme 5).
ortho-lithACTHNUTRGNEUGiN ation (likely a result of the high electronic density
Scheme 5. Alkynylation of carbamate 5M and diisopropylcarboxamide
5N.
Scheme 7. Regioselective alkynylation.
Chem. Eur. J. 2012, 18, 8414 – 8422
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8417

J. L. Garcꢀa Ruano, J. Alemꢁn et al.
of the ring) results in the need to perform the reaction with
1a at from À788C to room temperature (the other reactions
were nearly instantaneous at À788C) by using a large excess
of the nucleophile (see the Supporting Information) to
obtain 4aQ[12] in 80% yield.
Concerning regioselectivity,
a particularly interesting
result was obtained from 2-methoxynaphthalene 5R, which
upon lithiation by nBuLi and addition of 1a afforded a 95:5
mixture of the two possible regioisomers, with 4aR being
the major one, which was obtained in 70% after its chroma-
tographic purification (Scheme 7, b). The observed regio-
ACHTUNGTRENNUNGselectivity (attack at C3 is favored) can be explained by as-
suming that electronic and steric factors make both the lith-
iation and subsequent alkynylation at C1 difficult. More re-
gioselective was the evolution of the estradiol derivative 5S,
which reacts with nBuLi and 1a to afford exclusively 4aS in
89% yield (Scheme 7, c).
Other non-benzenoid aromatic systems are also sensitive
to the Truceꢄs reaction. This is the case of the direct alkyn-
ACHTUNGTRENNUNGylation of ferrocene 5T with tBuLi and 1a (Scheme 8) yield-
Scheme 9. Alkynylation of aromatic heterocycles.
performed quantum chemistry calculations using the Gaussi-
an 09 code.[35] Our simulations were carried out with density
functional theory (DFT) using the B3LYP functional.[36,37]
First, we optimized the geometry of compound 1a at the
Scheme 8. Ferrocene alkynylation.
B3LYP/6-31+GACTHUNTGRNEUNG(d,p) level. From the obtained geometry, we
calculated the charge on each atom of the molecule using
Natural Bonding Orbital theory (NBO)[38] (see Figure 1a).
We also evaluated the frontier molecular orbitals (HOMO
and LUMO) of compound 1a; following the dual descrip-
ing 4aT in 67% yield. This is remarkable because, to our
knowledge, all the methods available for introducing alkynyl
groups at ferrocene cores are indirect or give poor yields.[31]
2
À
Given that many Csp H bonds of aromatic heterocycles
are sufficiently activated to undergo lithiation with RLi, we
tried to directly introduce an alkyne into such positions. The
resulting alkynylated heteroaromatic rings are important in-
termediates in the preparation of bioactive molecules.[32]
The a positions of furan and thiophene were alkynylated to
afford 4aU and 4aV in high yields (Scheme 9, a) without
any double alkynylation, which had been observed using
other methods. N-methylpyrrole also gave the expected
product 4aW, although special conditions were required
(see the Supporting Information). Interestingly, N-methyl-
ACHTUNGTRENNUNGindole 5X gave 4aX, the alkynylation product at C2
(Scheme 9, b), whereas the regioselectivity can be changed
starting from the TIPS derivative 5Y, which afforded the C3
alkynylated product 4aY in low yield after silyl-deprotection
(Scheme 9, c).[33] These compounds had been previously pre-
pared using both catalytic[34a–c] and non-catalytic methods[34d]
generally by using long synthetic sequences. Analogously,
the mild conditions used for obtaining 4aZ in 64% yield
(À788C, 5 min, Scheme 9, d) by metalation of benzothiazole
5Z and further reaction with 1a, are in contrast with the
harsh conditions (toluene, 1008C, 24 h) previously used for
the alkynylation of benzothiazoles.[14d]
Figure 1. a) Atomic charges from a Natural Bonding Orbital analysis and
b) electronic density difference between the LUMO and HOMO orbitals
[1LUMOÀ1HOMO] for compound 1a.
Mechanistic considerations: To understand the unexpected
electrophilic behavior of substituted sulfonylacetylenes, we
8418
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8414 – 8422

Arylsulfonylacetylenes as Alkynylating Reagents
FULL PAPER
tor,[39] the most favored site for nucleophilic attack is the
one presenting the highest value of electronic density differ-
ence between the LUMO and the HOMO [1LUMOÀ1HOMO
]
as shown in Figure 1b. Both the NBO study and the molecu-
lar orbital analysis reveal that C-b is the most electrophilic
position of compound 1a, suggesting that nucleophilic
attack should take place through a more typical Michael-
type approach.
An inspection of the obtained geometry shows that the
À
Csp S bond (1.8 ꢅ) is longer than the Csp-Ph (1.5 ꢅ). This
suggests a larger importance of the steric effects on the de-
stabilization of the b-approach exerted by the substituent at
C-b, which in the case of the phenyl group (1a) could par-
tially compensate the electronic tendency. However, the ob-
served regioselectivity is complete, resulting in the exclusive
formation of the “anti-Michael” products in most of the
studied cases, even for substrates bearing primary alkyl
groups at C-b, which are clearly smaller than the sulfonyl
groups. In these cases, the steric effects cannot be invoked
for justifying the regioselectivity, which could be better ex-
plained by assuming the association of the lithium to the sul-
fonyl oxygen as the previous step of an intramolecular
attack of the nucleophile, which takes place to the a carbon
more easily than the b carbon (Figure 2).[40]
Figure 3. Mechanism of a- and b attack of methyl-lithium (2B) on sulfone
1a. Minimum of the sulfone–MeLi coordination and transition states of
both attacks are shown together with the energies.
a smaller energy barrier than attack at the b position
(4.40 kcalmolÀ1 smaller for PhLi and 3.56 kcalmolÀ1 for
MeLi). This result explains the complete selectivity observed
in these reactions. The difference between the used organo-
lithium is in accordance with the size of these two groups,
being notable larger when Ph was employed. It is remark-
able that when the solvent is not considered, the calculations
predict that attack at the b position will be favored, whereas,
when THF is included in the simulation, the trend is invert-
ed, and the theory predicts nearly exclusive formation of the
alkyne resulting from an a attack. Attenuation of the elec-
trostatic interactions by the solvent (mainly favoring the
b attack) and a more efficient solvation of the tetrahedral Li
at TSa than the almost diagonal Li at TSb, could account
for the significant influence of the solvent on the reaction
course.
To gain a deeper insight in the experimental results, in
particular the anomalous cases in which the synthesis of the
“anti-Michael” product was not successful (entries 15 at
Table 1 and 3 at Table 2), and provide an explanation on the
factors governing the observed behavior, we carried out
a systematic theoretical study of the reactions of PhLi with
various sulfonylacetylenes: 1a–c, 1e, 1j–k, and that of
nBuLi with 1a (Table 4). The relative energies of the reac-
tions taking place at C-a (Tsa) in different alkynes
1 depend on the ability of R1 to stabilize the incipient nega-
tive charge developed at C-b (electronic factors) and on its
size (steric effects). Our theoretical results show that elec-
tronic factor dominate against steric effects in Tsa. Thus,
the values are almost constant for 1e, 1j, and 1c (alkyl
groups do not stabilize the negative charge) but significantly
decrease for 1b and 1a (entries 4 and 5, Table 4), which
have Si and the Ph groups, respectively, to stabilize the neg-
ative charge. As further evidence for this explanation, we
studied theoretically the reaction of PhLi with 1k, which has
a strong electron-donating group (NMe2) on the aromatic
ring of the sulfonylacetylene. A substantial increase in the
energy of the transition state was observed (entry 6,
Table 4), as would be expected for an electron-rich aromatic
ring destabilizing a negative charge at the C-b. The influ-
Figure 2. Mechanism of a- and b attack of PhLi (2E) on sulfone 1a. The
minimum of the sulfone-PhLi coordination and transition states of both
attacks are shown together with the energies.
To clarify these alternatives, a more realistic picture of
the reaction mechanism has been obtained by using further
DFT calculations. Initially, we studied the reaction of 1a
with PhLi (2E), which gave exclusively 4aE in almost quan-
titative yield (entry 4, Table 2) and also with MeLi (entry 2,
Table 1). Beginning from non-complexed starting materials,
the calculations indicate the formation of a complex with
the nucleophile coordinated to the electrophile (structure in
the center of Figures 2 and 3) by association of the lithium
to one of the sulfonyl oxygen atoms.[41] It corresponds to
a minimum in the potential energy surface. From this point,
we then obtained the transition states corresponding to
attack at the a- and b positions (right and left structures in
Figures 2 and 3[42]). The attack at the a position shows
Chem. Eur. J. 2012, 18, 8414 – 8422
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8419

J. L. Garcꢀa Ruano, J. Alemꢁn et al.
Table 4. Transition states energies associated with attack at C-a and C-b,
obtained from DFT calculations, for various arylsulfonylacetylenes and
organolithiums.
ꢀ
ilar reasoning can be used for the result that tBu-C C-nBu
can be obtained by the reaction of 1c and 2A (entry 9 in
Table 1), whereas the reaction of 1e and 2D (entry 15,
Table 1) was unsuccessful.[43]
Conclusion
Entry
RLi
Substrate (R1)
TSa
TSb
We have explored the scope and limitations of Truceꢄs reac-
tion, in which the alkynylation with b-substituted 2-para-tol-
ylsulfonylacetylenes of a wide variety of organolithium spe-
cies obtained from commercial sources or prepared in situ
by metal-halogen exchange reaction or direct deprotonation
of aromatic, heteroaromatic and olefinic compounds. The
[kcalmolÀ1
]
[kcalmolÀ1
]
1
2
3
4
5
6
7
Ph
Ph
Ph
Ph
Ph
Ph
nBu
1e (nBu)
1j (sBu)
1c (tBu)
1b (TMS)
1a (Ph)
1k (p-NMe2-C6H4)
1a (Ph)
12.24
12.13
11.76
9.78
8.85
13.29
6.51
10.46
11.35
15.60
13.35
13.26
15.59
8.89
broad scope of the reaction and its simple experimental pro-
3
À
cedure make possible the formation of almost any Csp Csp
2
À
and Csp Csp bonds, with only the limitations inherent to
organolithium chemistry at low temperatures.[44] The meth-
odology can be also applied to complex molecules, allowing
for the introduction of alkyne moieties into naphthol deriva-
tives, steroids and functionalized benzofuranes. A rational
explanation of all these results can be achieved by theoreti-
cal calculations, which suggest that the association of the or-
ganolithiums to the electrophile is a previous step of their
intramolecular attack, as responsible for the unexpected
“anti-Michael” reactions observed for substituted sulfonyl-
ence of the size of the attacking R group on this energy is
also significant, given that the R/SO2Ph interaction could in-
terfere with the a attack. These steric factors could explain
the higher stability of TSa when nBuLi instead of PhLi is
used as the nucleophile (compare the R group size in en-
tries 5 and 7, Table 4).
The factors controlling the stability of the TSb, however,
are different. Their energies depend significantly on the rel-
ative size of R and R1 (steric interactions now being the pri-
mary destabilizing force). As such, there is a distinct in-
crease within the series 1e<1j !1c (entries 1–3, Table 4).
The similar size of TMS (1b) and Ph (1a), both smaller
than tBu (1c), explain the values obtained for TSb in en-
tries 4 and 5 (lower than entry 3, Table 4). However, the
electronic effects resulting from modifications of the elec-
tronic deficiency are also significant. This can be deduced
from the comparison of the entries 1 and 7 (Table 4), in
which the steric interactions are similar (Ph/nBu), but with
the Ph group (ÀI) providing a much greater electronic
effect than the nBu (+I). Similarly, results obtained in en-
tries 5 and 6 (similar sizes of 1a and 1k, but different elec-
tronic effect) also support the important role of electronic
factors. However, these effects are less significant for TSb
than for TSa (DEb=2.33 kcalmolÀ1 and DEa=4.43 kcal
molÀ1 for entries 5 and 6, Table 4).
AHCTUNGTRENNGaUN cetylenes.
Acknowledgements
Financial support from Spanish Government (CTQ-2009-12168,
CTQ2010-17006) and CAM (“programa AVANCAT CS2009/PPQ-1634”)
is gratefully acknowledged. J.A., M.T. and S.D.T thank the MICINN for
“Ramon y Cajal” contracts, C.A. thanks the “Consejo Nacional de Cien-
cia y Tecnologꢀa de Mꢆxico” for a postdoctoral fellowship and L.M.
thanks the Ministerio de Educaciꢂn y Ciencia for a predoctoral fellow-
ship. We gratefully acknowledge computational time provided by the
Centro de Computaciꢂn Cientꢀfica at the Universidad Autꢂnoma de
Madrid CCC-UAM. We also thank Juan Jose Mayor for permission to
use the background picture, and Alfonso Moyano for assistance in the
design.
[1] Modern Acetylene Chemistry (Eds.: P. J. Stang, F. Diederich), Wiley-
VCH, Weinheim, 1995.
To explain the observed regioselectivity, we had to consid-
er the relative stability of the transition states corresponding
to attacks at C-a and C-b. In entries 3–5 (Table 4), the ob-
served preference for C-a is predicted by the calculations
(DE>3.5 kcalmolÀ1). The inversion of the regioselectivity
predicted in entries 1 and 2 (Table 4) can be explained by
the large relative change in the energy of TSb, while that of
TSa remains nearly constant. This prediction was experi-
mentally confirmed by the reaction of 1e with PhLi (2E),
which did not give the anti-Michael product 4eE (entry 3,
Table 2). By contrast, reaction of nBuLi (2 A) with 1a,
yielding 4aA (identical to 4eE), was predicted to be suc-
cessful based on the calculations (entry 7, Table 3), and con-
firmed by the experimental results (entry 1, Table 1). A sim-
[2] J. L. Garcꢀa Ruano, J. Alemꢁn, L. Marzo, C. Alvarado, M. Tortosa,
S. Dꢀaz-Tendero, S. Fraile, Angew. Chem. Int. Ed. 2012, 51, 2712–
2716.
[3] R. L. Smorada, W. E. Truce, J. Org. Chem. 1979, 44, 3444–3445.
[4] For a review, see: a) T. G. Back, Tetrahedron 2001, 57, 5263–5301.
[5] J. J. Eisch, M. Behrooz, S. K. Dua, J. Organomet. Chem. 1985, 285,
121–136.
[6] J. J. Eisch, M. Beharroz, J. E. Galle, Tetrahedron Lett. 1984, 25,
4851–4854.
[7] For examples, see: a) E. V. Dehmlow, U. Fastabend, Gazzetta Chimi-
ca Italiana 1996, 126, 53–55; b) K. H. Dotz, A. J. Gerhardt, Organo-
met. Chem. 1999, 578, 223–228; c) N. Hayashi, H. Noguchi, S.
Tsuboi, Tetrahedron 2000, 56, 7123–7137.
[8] In this report, alkynyl lithium derivatives reacted quite well with the
primary alkyl halides, whereas secondary gave no reaction, see:
a) M. Buck, J. M. Chong, Tetrahedron Lett. 2001, 42, 5825–5827. A
8420
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8414 – 8422

Arylsulfonylacetylenes as Alkynylating Reagents
FULL PAPER
Protocol for diaryl acetylenes by a one-shot double elimination pro-
cess has been also described, see: b) A. Orita, K. Miyamoto, M. Na-
kashima, F. Ye, Ju. Otera, Adv. Synth. Catal. 2004, 346, 767–776;
c) A. Orita, H. Taniguchi, J. Otera, Chem. Asian J. 2006, 1, 430–437.
[9] There is a method which describes a cross-coupling on alkynyl-
indium derivatives with secondary benzylbromides, see: J. Caiero, J.
Pꢆrez-Sestelo, L. A. Sarandes, Chem. Eur. J. 2008, 14, 741–746.
[10] a) B. M. Trost, A. H. Weiss, Adv. Synth. Catal. 2009, 351, 963; b) L.
Lin, R. Wang, Current Org. Chem. 2009, 13, 1565–1576; c) E. Tyr-
rell, Current. Org. Chem. 2009, 13, 1540–1552; d) G. Blay, A. Mon-
lꢆon, J. R. Pedro, Current Org. Chem. 2009, 13, 1498–1539.
[11] For recent reviews, see: a) R. Chichilla, C. Nꢁjera, Chem. Rev. 2007,
107, 874–922; b) M. M. Heravi, S. Sadjavi, Tetrahedron 2009, 65,
7761–7775; c) H. Doucet, J. C. Hierso, Angew. Chem. 2007, 119,
850–888; Angew. Chem. Int. Ed. 2007, 46, 834–871.
[25] For a review in this field, see: a) A. L. K. Shi Shun, R. R. Tykwinski,
Angew. Chem. 2006, 118, 1050–1073; Angew. Chem. Int. Ed. 2006,
45, 1034–1057. For different examples, see: b) M. N. Hopkinson,
J. E. Ross, G. T. Giuffredi, A. D. Gee, V. Gouverneur, Org. Lett.
2010, 12, 4904–4907; c) K. Gueogjian, F. V. Singh, J. M. Pena,
M. F. Z. J. Amaral, H. A. Stefani, Synlett 2010, 427–432; d) N. Sakai,
R. Komatsu, N. Uchida, R. Ikeda, T. A. Konokahara, Org. Lett.
2010, 12, 1300.
[26] These features are more advantageous than those previously report-
ed (longer reaction times, lower yields and more experimentally dif-
ficult), see ref. [25].
[27] The halogen–lithium exchange of the starting bromide proved prob-
lematic, thus required the use of a large excess of 2K for getting
a complete conversion of 1a and only gave a moderate yield (2K
was not consumed and the unreacted material was easily recov-
ered).
[28] a) M. C. Whisler, S. Macneil, V. Snieckus, P. Beak, Angew. Chem.
2004, 116, 2256–2276; Angew. Chem. Int. Ed. 2004, 43, 2206–2225;
b) V. Snieckus, Chem. Rev. 1990, 90, 879–933; c) C. G. Hartung, V.
Snieckus, The Directed ortho Metalation Reaction: A Point of Depar-
ture for New Synthetic Aromatic Chemistry, in Modern Arene
Chemistry (Ed.: D. Astruc), Wiley-VCH, Weinheim, 2004.
[29] The search of coupling reactions fulfilling these two features was
identified as one the most important goal of the pharmaceutical in-
dustries; see ref. [18].
[30] D. Yue, T. Yao, R. C. Larock, J. Org. Chem. 2005, 70, 10292–10296.
[31] For a review, see: a) N. J. Long, C. K. Williams, Angew. Chem. 2003,
115, 2690–2722; Angew. Chem. Int. Ed. 2003, 42, 2586–2617;
b) D. P. Dyadchenko, M. A. Dyadchenko, V. N. Okulov, D. A. Leme-
novskii, J. Organomet. Chem. 2011, 696, 468–472.
[32] See, for example: a) P. A. Donets, G. V. Latyshev, N. V. Lukashev,
I. P. Beletskaya, Russian Chemical Bulletin Int. Ed. 2007, 56, 504–
508; b) M. Huxley, C. Sanchez-Cano, M. J. Browning, C. Navarro-
Ranninger, A. G. Quiroga, A. Rodger, M. J. Hannon, Dalton Trans.
2010, 39, 11353–11364.
[33] a) M. Matsuzono, T. Fukuda, M. Iwao, Tetrahedron Lett. 2001, 42,
7621–7623. For the importance of indole as privileged structure in
biochemistry, medicinal chemistry and science materials see: b) C.
Sandro, F. Giancarlo, Chem. Rev. 2011, 111, PR215–PR283, and ref-
erences therein.
[12] L. S. Hegedus, in Organometallics in Synthesis (Ed.: M. Schlosser),
Wiley-VCH, Weinheim, 2002, p. 1150.
[13] a) N. Yoneda, S. Matsuoka, N. Miyaura, T. Fukuhara, A. Suzuki,
Bull. Chem. Soc. Jpn. 1990, 63, 2124–2126; b) T. Sakamoto, F.
Shiga, A. Yasuhara, D. Uchiyama, Y. Kondo, H. Yamanaka, Synthe-
sis 1992, 746–752; c) N. G. Kundu, S. K. Dasgupta, J. Chem. Soc.
Perkin Trans. 1 1993, 2657–2668.
[14] For a highlight article, see: a) A. S. Dudnik, V. Gevorgyan, Angew.
Chem. 2010, 122, 2140–2142; Angew. Chem. Int. Ed. 2010, 49, 2096–
2098; b) For a pioneering work, see: K. Kobayashi, M. Arisawa, M.
Yamaguchi, J. Am. Chem. Soc. 2002, 124, 8528–8529. For different
examples, see: c) R. Amemiya, A. Fujii, M. Yamaguchi, Tetrahedron
Lett. 2004, 45, 4333–4335; d) N. Matsuyama, K. Hirano, T. Satoh,
M. Miura, Org. Lett. 2009, 11, 4156–4159; e) A. Rodriguez, R. V.
Fennessy, W. J. Moran, Tetrahedron Lett. 2009, 50, 3942–3944; f) T.
Kawano, N. Matsuyama, K. Hirano, T. Satoh, M. Miura, J. Org.
Chem. 2010, 75, 1764–1766; g) M. Tobisu, Y. Ano, N. Chatani, Org.
Lett. 2009, 11, 3250–3252; h) Y. Gu, X.-M. Wang, Tetrahedron Lett.
2009, 50, 763–766; i) B. A. Trofimov, L. N. Sobenina, Z. V. Stepano-
va, T. L. Vakulskaya, O. N. Kazheva, G. D. Aleksandrov, O. A. Dya-
chenko, A. Mikhaleva, Tetrahedron 2008, 64, 5541–5544.
[15] T. de Haro, C. Nevado, J. Am. Chem. Soc. 2010, 132, 1512–1513.
[16] J. P. Brand, J. Waser, Angew. Chem. 2010, 122, 7462–7465; Angew.
Chem. Int. Ed. 2010, 49, 7304–7307.
[17] J. P. Brand, J. Charpentir, J. Waser, Angew. Chem. 2009, 121, 9510–
9513; Angew. Chem. Int. Ed. 2009, 48, 9346–9349.
[18] D. J. C. Constable, P. J. Dunn, J. D. Hayler, G. R. Humphrey, J. L.
Leazer, Jr., R. J. Linderman, K. Lorenz, J. Manley, B. A. Pearlman,
A. Wells, A. Zaks, T. Y. Zhang, Green Chem. 2007, 9, 411.
[19] B. M. Trost, Angew. Chem. 1995, 107, 285–307; Angew. Chem. Int.
Ed. Engl. 1995, 34, 259–281.
[20] Sulfone 1c was reacted with 2A on a 4.0 mmol scale, recovering
3.8 mmol of lithium-p-toluensulfinate, which can be easily trans-
formed into 1c by using ceric ammonium nitrate (CAN) and tert-bu-
tylacetylene, see: V. Nair, A. Augustine, T. D. Suja, Synthesis 2002,
2259–2265.
[21] The simplicity of the procedure (5 min at À788C), which could also
be used for preparing non symmetrical di-t-alkylacetylenes, contrasts
with the complexity of other methods previously reported for pre-
paring 4cD. See, for example: a) J. Wang, Y. Gurevich, M. Botosh-
ansky, M. S. Eisen, J. Am. Chem. Soc. 2006, 128, 9350–9351; b) J.
Terao, N. Kambe, N. Sonoda, Tetrahedron Lett. 1998, 39, 5511–5512;
c) G. Maas, K. Schneider, W. Ando, J. Chem. Soc. Chem. Commun.
1988, 72–74.
[22] See, for example: a) J. L. Bolliger, C. M. Frech, Adv. Synth. Catal.
2009, 351, 891–902; b) S. Saleh, M. Picquet, P. Meunier, J.-C.
Hierso, Tetrahedron 2009, 65, 7146–7150.
[23] See, for example, a) M. Islam, P. Mondal, A. Singha, R. Kazi, Syn-
thesis 2010, 2399–2406; b) N. Ngassa, E. A. Lindsey, B. E. Haines,
Tetrahedron 2009, 65, 4085–4091.
[34] a) A. Minato, K. Tamao, T. Hayashi, K. Suzuki, M. Kumada, Tetra-
hedron Lett. 1981, 22, 5319–5322; b) M. J. Mphahlele, Tetrahedron
2010, 66, 8261–8266; c) V. Fiandanese, D. Bottalico, G. Marchese,
A. Punzi, Tetrahedron 2008, 64, 53–60; d) R. Soley, F. Albericio, M.
ꢇlvarez, Synthesis 2009, 1559–1564.
[35] Gaussian 09, Revision B.01, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V.
Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X.
Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J.
Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
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, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A.
Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, ꢈ.
Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian,
Inc., Wallingford CT, 2010.
[36] A. D. Becke, J. Chem. Phys. 1993, 98, 1372–1377.
[37] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1998, 37, 785–789.
[38] A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev. 1998, 88, 899–
926.
[39] C. Morell, A. Grand, A. Toro-Labbꢆ, J. Phys. Chem. A 2005, 109,
205–212.
[24] a-Alkynylation of pyridines has been achieved under Sonogashira
conditions at temperatures higher than 1208C (see ref. [22]) or
longer reaction times (12–20 h) (see ref. [23]).
[40] We have considered other mechanistic alternatives. The ipso-substi-
tution was excluded because the energy of the corresponding TS, de-
Chem. Eur. J. 2012, 18, 8414 – 8422
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8421

J. L. Garcꢀa Ruano, J. Alemꢁn et al.
termined by DFT calculations, was much higher than that of the
anti-Michael process. Another mechanism to be considered is based
on the evolution of the anion resulting in the Michael addition into
the carbene by elimination of the sulfonyl group and further rear-
rangement of the Ar groups. It was also rejected because the TS for
this evolution is less stable than that observed for anti-Michael addi-
tion (see Figure 2). Additionally, we could not observe any differ-
ence in the reaction rate by changing the nature of the migrating
group (aryl, heteroaryl, or vinyl) which should be the case if the re-
arrangement (carbene mechanism) was the limiting step of the reac-
tion.
[42] The geometry of these species has been optimized at the B3LYP/6-
31+GACHTNUTRGNEUNG(d,p) level of theory. Harmonic vibrational frequencies were
obtained for each structure to characterize them as minima or tran-
sition states. These frequencies were also used to compute the zero-
point energies (ZPE). Evaluation of the final relative energies was
carried out using single-point energy calculations with the same
functional and a larger basis set: B3LYP/6-311+GACHTNUTRGNEUNG(3df,2p), including
the ZPE correction and solvent effects (THF) by using the Polariza-
ble Continuum Model (PCM), see: J. Tomasi, B. Mennucci, R.
Cammi, Chem. Rev. 2005, 105, 2999–3093.
[43] According to Smorada and Truce (see ref. [3]), the poor results ob-
tained in the reactions of entry 15 in Table 1 and entry 3 in Table 2
are due to the acidity of the propargylic protons, which immediately
protonate the organolithium, precluding its attack to C-a. However,
the formation of the Michael products in some extension and
mainly other successful results at Table 1 on substrates having prop-
argylic protons demonstrate that the explanation given by Truce is
wrong and suggest other ones related to the relative facility of the
a- and b attacks as responsible of the lack of reactivity observed in
some cases.
[41] This suggested that the presence of lithium is essential for the anti-
Michael reaction of sulfonylacetylenes with organollithiums. To pro-
vide some experimental support to this claim, we repeated the reac-
tion of entry 2 in Table 1 in the presence of 12-crown-4-ether hoping
to capture the metal. In this case, we did not observe any formation
of 4aB, and similar results were observed in other cases that had
provided the desired product without the crown ether. These unsuc-
cessful results support the idea that the association of the reagent to
the sulfonyl oxygen atoms prior to intramolecular nucleophilic
attack plays an important mechanistic role. However, the best ob-
tained results in the preparation of the indole derivative 4aY
(Scheme 9, b) was achieved in the presence of tetramethylethylene-
diamine (TMEDA), which suggests that a more complex situation
must be considered in those substrates possessing ortho-directing
groups.
[44] For the industrial scale application of this method, temperatures
higher than À788C would be desirable. In this sense, we have pre-
pared 4aA at À208C, without significant erosion of the isolated
yield (95%).
Received: March 20, 2012
Published online: June 4, 2012
8422
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 8414 – 8422