A straightforward alkynylation of Li and Mg metalated heterocycles with sulfonylacetylenes

Article abstract of DOI:10.1039/c4cc07574a

Coupling of alkynyl moieties to heterocyclic rings, without using transition metals, can be easily performed by the reaction of aryl or heteroaryl sulfonylacetylenes with heteroaryl-Li compounds or their corresponding less reactive magnesium derivatives.

Full text of DOI:10.1039/c4cc07574a

ChemComm
View Article Online
View Journal
COMMUNICATION
A straightforward alkynylation of Li and Mg
metalated heterocycles with sulfonylacetylenes†
Cite this:DOI: 10.1039/c4cc07574a
Leyre Marzo, Ignacio Perez, Francisco Yuste, Jose Aleman* and
´
´
´
Received 25th September 2014,
Accepted 5th November 2014
Jose Luis Garcıa Ruano*
´
´
DOI: 10.1039/c4cc07574a
www.rsc.org/chemcomm
Coupling of alkynyl moieties to heterocyclic rings, without using or the dopamine receptor (III),4 which is used against schizo-
transition metals, can be easily performed by the reaction of aryl or phrenia. Particularly interesting are diheteroarylated acetylenic
heteroaryl sulfonylacetylenes with heteroaryl-Li compounds or moieties, like those containing a thienyl residue, which are also
their corresponding less reactive magnesium derivatives.
present in pharmacologically interesting compounds. Thus,
thiophene and furane rings are joined to the triple bond in the
Heteroarylated acetylenic structures are widespread in the field tryptase inhibitor IV,5 used for treating allergic or inflammatory
of materials science, as well as in many biologically interesting disorders, whereas thiophene and thiazole form part of the
compounds1 (Fig. 1). In the latter field, they are present in a structure of the mGlu5 receptors V.6 In addition, the 2,30-
large number of products with biological activity like carlina dithiophene-acetylene moiety is present in the IKK-2 inhibitor
oxide2 (I) with antimicrobial properties, the metabotropic gluta- VI7 and in the anti-tumour agent VII.8
mate receptor II,3 related to the learning and memory processes,
The most employed methodology for the synthesis of hetero-
aryl acetylenes involves the use of the well-known Sonogashira
reaction,9 starting from heterocyclic halides and aryl or hetero-
aryl acetylenes and using palladium as the catalyst. As an
alternative, the direct alkynylation of heteroaromatic com-
pounds with alkynyl halides was developed (inverse Sonoga-
shira),10 but it is restricted to highly activated C–H heterocycles
(e.g. thiazole). Very recently, other strategies, like the use of
hypervalent iodine reagents with AuI as the catalyst11 (eqn (a),
Scheme 1) and the oxidative cross-coupling of terminal alkynes
with heteroarenes,12 have been reported (eqn (b), Scheme 1).
However, the use of the hypervalent iodine is only valid for
preparing SiiPr3 alkynyl derivatives11 and the oxidative cross-
coupling12 is mainly circumscribed to the synthesis of
2-arylalkynyl 5-substituted thiophenes (rather modest yields
were obtained with furane and pyrrol derivatives). Thus, these
Fig. 1 Heteroaryl alkynes of biological interest.
´
´
´
Departamento de Quımica Organica (Modulo 1), Facultad de Ciencias,
´
Universidad Autonoma de Madrid, Cantoblanco, 28049-Madrid, Spain.
E-mail: jose.aleman@uam.es, joseluis.garcia.ruano@uam.es;
Web: www.uam.es/jose.aleman
† Electronic supplementary information (ESI) available: Experimental. See DOI:
10.1039/c4cc07574a
Scheme 1 Recent approaches for the direct alkynylation of heterocycles.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun.

View Article Online
Communication
ChemComm
Table 1 Alkynylation of different heterocycles (2A–M) based on their
C–H activation with t-BuLi or n-BuLia
methods preclude the synthesis of monosubstituted hetero-
cycles, bearing only the alkynyl moiety, which is the case of
many of the compounds shown in Fig. 1. Moreover, these
alternative methods cannot be used for incorporating acetylenic
chains into the less activated positions or the heterocyclic rings
and therefore cannot be considered as general alternatives to the
Sonogashira reaction for alkynylating heterocycles.
All these methodologies require the use of transition metals
for the formation of the Csp–Csp2 bond, which was recognized
as a handicap for pharmaceutical companies,13 which indicated
the convenience of finding alternative methods for C–C coupling,
avoiding the use of transition metals as catalysts. In 2012, our
group reported the unexpected electrophilic behavior of aryl-
sulfonylacetylenes that undergo an unusual a-attack (anti-Michael
addition) in reactions with organolithiums, followed by elimination
of the ArSO2ꢀ moiety to allow the alkynylation of lithiated Csp2 or
Csp3 (eqn (a), Scheme 2).14 This transition-metal free methodo-
logy was applied regardless of the source of the organolithiums
(alkyl-lithium derivatives14b or arenes with activated C–H14a). Our
goal in this work is the use of this methodology to prepare
different alkynyl heterocyles (eqn (b), Scheme 2), paying special
attention to the regioselective synthesis of dialkynyl heterocycles
(useful in materials science) and diheteroaryl acetylenes (present
in pharmacologically interesting compounds, see Fig. 1). As some
of the studied substrates were not stable in the presence of
organolithiums, the behavior of the less reactive Grignard
reagents was studied, which provided surprisingly good results.
We initiated our study by considering the results previously
obtained in the preparation of the 2-phenylethynyl heterocycles
3Aa, 3Ba, and 3Ca (Table 1).14a Typically, the a-lithiation is
performed with n-BuLi or t-BuLi in THF or ether at 0 1C, and the
alkynylation with sulfone 1a at ꢀ78 1C in THF.15 Similar
conditions were successfully used to prepare the alkynyl deriva-
tives 3Da and 3Ea, from the 2-substituted furanes 2D and 2E, in
excellent yields (Table 1). At this point, we studied the behavior
of 2F and observed the exclusive formation of 3Fa in 60% yield
(Table 1), evidencing that the activation provided by the thio-
a
All the reactions were performed with 0.2 mmol of 1a and 0.4 mmol of
b
2A–J. Reaction carried out at the 2.34 mmol scale.
Table 2 Reactions of Li-2A with the p-tolylsulfonylacetylenes 1a–ga
Entry
Sulfone
Product
Yield (%)
1
2
3
4
5
6
7
1a (R = C6H5)
3Aa
3Ab
3Ac
3Ad
3Ae
3Af
89
58
95
76
1b (R = p-MeO-C6H4)
1c (R = p-CF3-C6H4)
1d (R = o-Cl-C6H4)
1e (R = Cy)
b
—
b
1f (R = t-Bu)
—
1g (R = TIPS)
3Ag
54
a
All the reactions were performed with 0.2 mmol of 1a–g and 0.4 mmol
b
of 2A. Reaction mixtures mainly containing Michael addition products.
phene was clearly higher than that of the furane ring. The of the anti-Michael addition. For this study 2-Li furan (Li-2A) was
preparation of other monoalkynylated heterocycles (Table 1), like chosen to react with different alkynylsulfones under standard
those derived from N-methylpyrazol (3Ga), thiazole (3Ha), benzo- conditions shown in Table 1 (THF, ꢀ78 1C) in a 10 min reaction
thiazole (3Ia), and imidazopyridine (3Ja), gave excellent yields (Table 2). Electron donating groups decreased the reactivity
under smooth conditions. In all the cases, the incorporation of (58% yield of 3Ab, compare entries 1 and 2), whereas electron-
the alkynyl residue only took place in the more activated position. withdrawing groups increased it (95% yield of 3Ac, compare entries
Our next step was to study the influence of the substituents at 1 and 3), which is not unexpected taking into account the effect of
the aryl group of p-tolylsulfonylacetylenes 1a–g on the reactivity these groups on the electrophilic character of the triple bond. The
reaction was also compatible with the presence of ortho-substituents
(76% yield of 3Ad, entry 4). The donating character of the alkyl
groups could explain the failure of the reactions with 1e and 1f to
obtain the anti-Michael products 3Ae and 3Af (entries 5 and 6). The
negative influence of the steric effects of these substituents must be
less relevant because the even larger TIPS group, present at 1g, was
not a handicap in its reaction with Li-2A (3Ag is formed in 54%
yield, entry 7), probably due to its favourable electronic effects
(see later). Compound 3Ag is important because it can be depro-
tected and functionalized, thus making possible the preparation
Scheme 2 Previous report and present work.
of alkyl acetylenes that cannot be obtained by direct reaction.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
Scheme 3 Functionalization of monoalkynyl heterocycles.
The absence of 2,5-dialkynylfuranes in the reaction mixtures
obtained from unsubstituted heterocycles 2 (Tables 1 and 2) suggests
the lower reactivity of their monoalkynyl derivatives 3 and allows
their use as starting materials in subsequent functionalization
processes based on the C–H activation of the other a-position
(C-5).16 In this sense, starting from 3Ca, we were able to prepare
2,5-dialkynylthiophenes like 4Cah, bearing two different acetylenic
moieties at the activated positions (eqn (b), Scheme 3), which are not
easily obtained with other methodologies. The introduction of other
electrophiles like CHO by reaction with DMF (5 in Scheme 3)
was also possible. In both cases, the reactions of the monoalkynyl
thiophene 3Ca with the electrophiles were performed at rt (at lower
temperatures the reactions did not work), which revealed its lower
reactivity with respect to the unsubstituted thiophene 2C, which
could be alkynylated at ꢀ78 1C, as was indicated in Table 1.
Scheme 5 Reaction of sulfones 1a and 1i with Grignard compounds.
Once we checked the potential of our method in the synthesis of
heteroaryl aryl acetylenes, our next goal was to obtain diheteroaryl
acetylenes. The synthesis of these compounds required the prepara-
tion of heteroaryl alkynyl sulfones to be used as starting products in
reactions with the heteroaryllithiums. We focus our attention on
sulfones containing the thiophene ring, like 1i,21 which would give
access to the skeleton of the diheteroaryl acetylenes IV–VII shown in
Fig. 1. Reaction of 1i with PhLi (Li–2L) at ꢀ78 1C in THF provided
the expected acetylene 3Ka, but in low yield (25%), along with a large
number of unidentified byproducts (eqn (a), Scheme 5). Reactions
with other heteroaryllithiums derived from 2M–2O were even less
fruitful. These results suggest that 1i is not stable in the presence of
organolithiums. Taking into account that Grignard reagents are less
reactive and, in some cases, more selective too than organolithium
compounds,22 we studied the reaction of 1i with PhMgCl (Mg-2L).
The reaction did not work at ꢀ78 1C and 0 1C but 3Ka was afforded
in an excellent 89% yield when it was conducted at rt (eqn (a),
Scheme 5). Then we explored the reactivity of different heteroaryl
Grignard derivatives (Mg-2M–O) with the phenylethynylsulfone 1a.
These reactions were very clean in all the cases (only one product
was detected by NMR in the reaction crudes) yielding alkynes 3Aa,
3Ca, and 3Ha in excellent yields after 2 h at rt (eqn (b), Scheme 5).
This reveals that the Grignard reagents, working at rt, are as efficient
as the organolithiums at ꢀ78 1C in their anti-Michael reactions with
alkynyl sulfones. Thus, Grignard compounds Mg-2M, Mg-2N, and
Mg-2O provided the corresponding diheteroaryl acetylenes (3Mi and
3Ni at rt and 3Oi at 50 1C) in good yields (decomposition products
were not detected), which confirmed the potential of our methodo-
logy to prepare diheteroaryl acetylenes (eqn (c), Scheme 5).
The mechanism proposed for the anti-Michael reactions of RLi
with substituted sulfonylacetylenes15 (Scheme 6), supported by
theoretical calculations, involves the association of the lithium with
the sulfinyl oxygens as the previous step of the intramolecular
a-attack of the R group on the triple bond and the subsequent
elimination of the metal sulfonate (Scheme 6). The ability of the
R0 to stabilize the carbanionic intermediate would explain the
behaviour of the different substrates shown in Table 2. The lower
reactivity and chelating ability of Grignard derivatives Mg-2 with
respect to those of Li-2 could show that the first ones require higher
temperatures to obtain good conversions. Moreover, these differences
in reactivity also support the nucleophilic character of the a-attack.23
At this point, we investigated the possibilities of our reaction for
introducing the alkynyl moieties into the non-activated positions of
the heterocycles, starting from the appropriate haloderivative and
generating the corresponding organolithium by lithium–halogen
exchange. Thus, starting from 3-bromothiophene (2K), the Li–Br
exchange with n-BuLi took place quickly (less than 15 minutes) at
ꢀ78 1C (Scheme 4) and was followed by the addition of 1a at ꢀ78 1C
that cleanly afforded 3Ka in almost quantitative yield.17
The incorporation of a second alkynyl group in the heteroaryl ring
of 3Ka with two activated C–H heterocycles (at C-2 and C-5) would
yield dialkynyl thiophenes, the formation of two regioisomers being
possible. To our delight, deprotonation of 3Ka with n-BuLi at ꢀ78 1C,
followed by reaction with the sulfone 1h at rt, only yielded the
2,3-dialkynyl derivative 4Kah (57% yield) in a completely regioselec-
tive way. The reactivity of 3Ka and 3Ca, both monoalkynyl derivatives,
was lower (reactions with 1 require rt) than that of the thiophene 2C
(reaction with 1 took place at ꢀ78 1C). However, the formation of the
organolithiums from 3Ka and 3Ca was easier (ꢀ78 1C) than from 3C
(0 1C to rt).18 This suggests a stabilizing influence of the alkynyl
group on the organolithium, which decreases its reactivity.19 The
properties and reactivity associated with the spatial proximity of the
two acetylenic moieties on C-2 and C-3 of the thiophene ring have
been useful in materials science and bionatural products.20
Scheme 4 Double alkynylation of 3-bromo thiophene.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun.

View Article Online
Communication
ChemComm
9, 1763. For a review on the electrophilic alkynylation of the same
group, see: J. P. Branda and J. Waser, Chem. Soc. Rev., 2012, 41, 4165.
12 (a) X. Jie, Y. Shang, P. Hu and W. Su, Angew. Chem., Int. Ed., 2013,
125, 3718; (b) For other related reactions with indoles, see: L. Yang,
L. Zhao and C.-J. Li, Chem. Commun., 2010, 46, 4184.
13 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 and T. Y. Zhang, Green Chem., 2007, 9, 411.
Scheme 6 Mechanistic proposal for the reaction of sulfonylacetylenes 1
with organolithium and organomagnesium reagents.
´
´
14 (a) J. L. Garcıa Ruano, J. Aleman, L. Marzo, C. Alvarado, M. Tortosa,
´
S. Dıaz-Tendero and A. Fraile, Angew. Chem., Int. Ed., 2012, 51, 2712;
´
´
(b) J. L. Garcıa Ruano, J. Aleman, L. Marzo, C. Alvarado, M. Tortosa,
´
S. Dıaz-Tendero and A. Fraile, Chem. – Eur. J., 2012, 18, 8414.
In conclusion we have demonstrated that the anti-Michael addi-
tion of RLi or R-MgX24 to sulfonylacetylenes constitutes an efficient
methodology to obtain different aryl-heteroaryl and diheteroaryl
acetylenes under very mild conditions. The broad scope, excellent
yields, and simplicity of the experimental procedure, which does not
require transition metals, are the main features of this methodology.
The present work can be considered as a general alternative to the
Sonogashira reaction in alkynylation reactions of heterocycles with-
out using transition metals.
15 These yields were obtained starting from 0.2 mmol of 2A–2C but the
reaction can be scaled up (3Ca was obtained in 95% yield on a
2.34 mmol scale).
16 Other C–H functionalizations are not selective strategies and a
mixture of the two activated positions would be found. Therefore,
only substituted thiophenes in the 2 position (with the 20 position
free to be activated) can be used as starting materials (see e.g. ref. 12).
17 The standard reaction time for the Br–Li exchange process was
15 minutes. Longer reaction times for the Br–Li exchange gave equilibria
of 3-Li-2K and 2-Li-2K, and consequenly the corresponding mixture of
monoalkynyl derivatives was formed.
18 This was confirmed by studying deuteration reactions at ꢀ78 1C of the
Li carbaniones generated from 3Ca and 3Ka at the same temperature.
This results in the exclusive formation of the 5-D and 2-D derivatives
respectively. The reaction of thiophene with n-BuLi at ꢀ78 1C only
yielded decomposition products, presumably due to the opening of the
ring with the organolithium (see e.g. K. Chernichenko, N. Emelyanov,
I. Gridnev and V. G. Nenajdenko, Tetrahedron, 2011, 67, 6812), whereas
the reaction at 0 1C produces its 2-Li derivative (unsensitive to the
opening), which is quantitatively deuterated with ND4Cl.
Financial support from the Spanish Government (CTQ2012-
35957) is gratefully acknowledged. J. A. thanks the MICINN for a
´
‘‘Ramon y Cajal’’ contract. I. P. thanks Conacyt-Mexico for a
´
predoctoral fellowship and F. Y. thanks Conacyt-Mexico and
DGAPA-UNAM for a sabbatical fellowship. L. M. thanks the
Spanish government for a FPU fellowship.
19 It can be explained by assuming an –I effect of the alkynyl group (weaker
for longer distance) that stabilizes the lithium carbanion but reduces its
reactivity. The complete regioselectivity observed in the alkynylation reac-
tion of 3Ka indicates that this stabilization is clearly higher for C(2)–Li than
for C(5)–Li, which supports the above statement. Another possible expla-
nation of this regioselectivity would involve the formation of a mixture of
C(2)–Li and C(5)–Li derivatives of 3Ka, with the first one being the most
reactive. Nevertheless this could be discarded by the exclusive deuteration
at C-2 observed by protonation with ND4Cl (see ref. 18).
20 For materials science, see: (a) P.-L. T. Boudreault, J. W. Hennek, S. Loser,
R. Ponce Ortiz, B. J. Eckstein, A. Facchetti and T. J. Marks, Chem. Mater.,
2012, 24, 2929; (b) M. J. O’Connor, R. B. Yelle, L. N. Zakharov and M. M.
Haley, J. Org. Chem., 2008, 73, 4424. For organic synthesis, see: (c) S. Naoe,
Y. Suzuki, K. Hirano, Y. Inaba, S. Oishi, N. Fujii and H. Ohno, J. Org.
Chem., 2012, 77, 4907; (d) M. M. Hansmann, M. Rudolph, F. Rominger,
A. Stephen and K. Hashmi, Angew. Chem., Int. Ed., 2013, 52, 2593.
21 Sulfone 1i was prepared in two steps from the commercially avail-
able 3-ethynyl thiophene by reaction with sodium toluene sulfinate
and NaI in the presence of CAN, followed by reaction with K2CO3 in
refluxing acetone, according to the procedure reported by Nair et al.
(V. Nair, A. Augustine and T. D. Suja, Synthesis, 2002, 2259). Other
heteroarylsulfonyl acetylenes could not be prepared by this proce-
dure, because the starting ethynyl derivatives are not commercially
available. We have tried to prepare them, but in the isolation step we
found serious problems due to their high volatility.
Notes and references
1 Acetylene Chemistry: Chemistry, Biology and Material Science, ed.
F. Diederich, P. J. Stang and R. R. Tykwinski, Wiley-VCH, Weinheim, 2005.
2 F. Herrmann, R. Hamoud, F. Sporer, A. Tahrani and M. Wink, Planta
Med., 2011, 77, 1905.
3 M. Packiarajan, M. Grenon, S. Zorn, A. T. Hopper, A. D. White,
G. Chandrasena, X. Pu, R. M. Brodbeck and A. J. Robichaud, J. Med.
Chem., 2013, 23, 4037.
4 C. Haubmann, H. Hu¨bner and P. Gmeiner, J. Med. Chem. Lett., 1999,
9, 3143.
5 M. J. Costanzo, S. C. Yabut, H.-C. Zhang, K. B. White, L. de Garavilla,
Y. Wang, L. K. Minor, B. A. Tounge, A. N. Barnakov, F. Lewandowski,
C. Milligan, J. C. Spurlino, W. M. Abraham, V. Boswell-Smith,
C. P. Pagec and B. E. Maryanoff, J. Med. Chem. Lett., 2008, 18, 2114.
6 A. Hopper, A. Sams, M. Graven, K. Gitte, M. Packiarajan and
M. Grenon, US Pat. Appl. Publ., US 20110092475 A1 20110421, 2011.
7 D. Bonafoux, S. Bonar, L. Christine, M. Clare, A. Donnelly, J. Guzova,
N. Kishore, P. Lennon, A. Libby, S. Mathialagan, W. McGhee,
S. Rouw, C. Sommers, M. Tollefson, C. Tripp, R. Weier, S. Wolfson
and Y. Min, Bioorg. Med. Chem. Lett., 2005, 15, 2870.
8 (a) R. Romagnoli, P. G. Baraldi, O. Cruz-Lopez, M. Tolomeo, A. Di
Cristina, R. M. Pipitone, S. Grimaudo, J. Balzarini, A. Brancale and
E. Hamel, Bioorg. Med. Chem. Lett., 2011, 21, 2746; (b) G. Liang,
S. Aldous, G. Merriman, J. Levell, J. Pribish, J. Cairns, X. Chen,
S. Maignan, M. Mathieu, J. Tsay, K. Sides, S. Rebello, B. Whitely,
I. Morize and H. W. Pauls, Bioorg. Med. Chem. Lett., 2012, 22, 1049.
9 For recent reviews, see: (a) H. Doucet and J.-C. Hierso, Angew. Chem.,
´
Int. Ed., 2007, 46, 834; (b) R. Chinchilla and C. Najera, Chem. Soc. Rev.,
2011, 40, 5084.
10 (a) S. H. Kim and S. Chang, Org. Lett., 2010, 12, 1868; (b) A. S. Dudnik and
V. Gevorgyan, Angew. Chem., Int. Ed., 2010, 49, 2096; (c) A. S. Dudnik and
V. Gevorgyan, Eur. J. Org. Chem., 2010, 6495; (d) I. V. Seregin, V. Ryabova
and V. Gevorgyan, J. Am. Chem. Soc., 2007, 129, 7742; (e) Y. Gu and X.-M.
22 (a) P. Knochel, W. Dohle, N. Gommermann, F. F. Kneisel, F. Kopp,
T. Korn, I. Sapountzis and V. A. Vu, Angew. Chem., Int. Ed., 2003, 42, 4302;
(b) H. Ila, O. Baron, A. J. Wagner and P. Knochel, Chem. Commun., 2006,
583. For a precedent of the alkylation of Grignards with alkynyl sulfones,
see: R. L. Smorada and W. E. Truce, J. Org. Chem., 1979, 44, 3444.
Wang, Tetrahedron Lett., 2009, 50, 763; ( f ) N. Matsuyama, K. Hirano, 23 The radical character suggested for the a-addition would not explain
`
´
the lower reactivity of the Grignard derivatives (see: J. L. Garcıa Ruano,
´
J. Aleman, A. Parra and L. Marzo, Eur. J. Org. Chem., 2014, 1577).
T. Satoh and M. Miura, Org. Lett., 2009, 11, 4156; (g) F. Besselievre and
S. Piguel, Angew. Chem., Int. Ed., 2009, 48, 9553.
11 (a) J. P. Brand, J. Charpentier and J. Waser, Angew. Chem., Int. Ed., 24 We are currently working on a full article concerning the use of the
2009, 48, 9346; (b) J. P. Branda and J. Waser, Angew. Chem., Int. Ed.,
2010, 49, 7304; (c) Y. Li and J. Waser, Beilstein J. Org. Chem., 2013,
Grignard reagents in alkynylation reactions with sulfonyl acetylenes,
which will be published in the near future.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2014