Interrupted oligomerization revisited: Simple and efficient one-pot multicomponent approach to versatile synthetic intermediates

Article abstract of DOI:10.1021/ol701847b

A novel multicomponent reaction allowing for a one-pot formation of three carbon-carbon bonds has been developed. It is based on in situ generation and anionic dimerization of methylenedithiane and produces a versatile synthetic equivalent of 4-hydroxy-1,3-alkanediones which, among other things, offers expeditious one-pot access to 3(2H)-furanones.

Full text of DOI:10.1021/ol701847b

ORGANIC
LETTERS
2007
Vol. 9, No. 20
4061-4063
Interrupted Oligomerization Revisited:
Simple and Efficient One-Pot
Multicomponent Approach to Versatile
Synthetic Intermediates
Roman A. Valiulin, Logan M. Halliburton, and Andrei G. Kutateladze*
Department of Chemistry and Biochemistry, UniVersity of DenVer,
DenVer, Colorado 80208
akutatel@du.edu
Received August 1, 2007
ABSTRACT
A novel multicomponent reaction allowing for a one-pot formation of three carbon−carbon bonds has been developed. It is based on in situ
generation and anionic dimerization of methylenedithiane and produces a versatile synthetic equivalent of 4-hydroxy-1,3-alkanediones which,
among other things, offers expeditious one-pot access to 3(2H)-furanones.
One-pot multicomponent reactions (MCRs) provide un-
matched opportunities for the expeditious increase of com-
plexity and diversity in synthetic outcomes. Particularly
successful are the isocyanide-based approaches to hetero-
cyclic systems, i.e., the Ugi-type reactions.1 For practical
reasons most of these reactions are limited to three to four
components, as selectivity and reasonable yields are progres-
sively more difficult to ensure in higher dimensional reagent
space.
Interrupted oligomerization is an attractive alternative,
which is normally controlled by a favorable formation of a
5-6-membered cyclic systemsexamples range from the age-
old trimerization of acetone,2 to the trimolecular formation
of thiophene or selenophene ring systems,3 to Posner’s
elegant interrupted anionic polymerization furnishing sub-
stituted cyclohexanes.4 Most of these approaches rely on a
strategically installed latent electrophile. To make this
methodology general, i.e., not dependent on the in situ
presence of the capping electrophile (or an equivalent leaving
group as in the case of vicarious nucleophilic substitution),
one needs to control the degree of polymerization, which
conceivably can be achieved by a rapid buildup of steric
hindrance as oligomerization progresses. The other objective
is to install synthetically useful groups for subsequent
functionalization. With this, we revisited Carlson’s early
studies of nucleophilic additions to 2-methylene-1,3-dithiane
(MDTA).5 Radical polymerization of MDTA is known to
produce a polymer, which is a masked 1,3-polycarbonyl
(3) Nakayama, J.; Yomoda, R.; Hoshino, M. Heterocycles 1987, 26,
2215-2222.
(1) For reviews see: (a) Do¨mling, A.; Ugi, I. Angew. Chem., Int. Ed.
2000, 39, 3168-3210. (b) Domling, A. Chem. ReV. 2006, 106, 17-89.
(2) Ipatieff, V.; Linn, C. B. US Patent 2419142 19470415, 1947.
(4) Posner, G. H.; Shulman-Roskes, E. M. J. Org. Chem. 1989, 54,
3514-3515.
(5) Carlson, R. M.; Helquist, P. M. Tetrahedron Lett. 1969, 173-176.
10.1021/ol701847b CCC: $37.00
© 2007 American Chemical Society
Published on Web 08/30/2007

compound.6 As alkyllithiums are known to add to MDTA,
we reasoned that a controlled anionic dimerization of MDTA
initiated by such addition and terminated with a synthetically
useful electrophile, for example, a carbonyl compound,
would constitute a multicomponent reaction in which four
modules are combined to furnish a linear framework of
masked 4-hydroxy-1,3-dione (Scheme 1). In many cases the
Scheme 2
Scheme 1
products can be simply hydrolyzed and cyclized into dihydro-
3-furanones in a one-pot reaction.
In this Letter we report a successful implementation of
this multicomponent strategy offering one-pot access to
versatile synthons for subsequent transformations.
readily separated by column chromatography or subjected
as a mixture to Hg2+-catalyzed hydrolysis after crude gel-
filtration. The ratio of bis- to monoadduct grows with time,
as more of A-1 is converted into A-2, but the absolute yield
of the bis-adduct declines after 15 min due to the slow
degradation of the anion A-2. We did not see any evidence
for the formation of trimeric dithianyl anion A-3. If desired,
the overall reaction can be accomplished in a stepwise
manner, with MDTA generated with 1 equiv of phenyl- or
butyllithium at the first step, followed by the dimerization
induced with a different alkyl- or aryllithium.
A similar outcome is achieved when excess lithiated
methyldithiane is treated with 1. Thus, several experimental
observations indicate that the dimerization of MDTA via
carbolithiation is fairly efficient, while trimerization is either
unfavorable or slow. Small amounts of bis-dithiane 4 were
isolated. Interestingly, in this case we did not observe
products of (mono) methyldithiane addition to aldehydes.
Our preferred precursor for MDTA is butylthiomethyl-
1,3-dithiane (1), synthesized readily from inexpensive bu-
tanethiol, 2-bromo-1,1-dimethoxyethane, and 1,3-propanedi-
thiol. Alternatively, alkoxymethyl-1,3-dithianes can be used,5,7
although in our hands these precursors proved inferior to 1.
Systematic optimization of reaction conditions with alkyl-
lithiums showed that the yield of a dimeric anion is
maximized at 20 °C in 15 min after the addition of 1.35-
1.45 equiv of RLi. Scheme 2 illustrates the tetracomponent
reaction of 1 initiated by alkyl- or aryllithiums and capped
by benzaldehyde. X-ray crystallography of 2a and 2b
confirmed the anticipated structures (Figure 1).
Generation of MDTA from 1 is very efficient and
completed within minutes at 20 °C. Excess RLi adds at this
temperature to form the initial monomeric dithianyl anion
A-1 within 5 min. The second addition of A-1 to MDTA
produces the bis-dithianyl anion A-2. Formation of this anion
is a slower process that can be controlled by varying reaction
conditions. For example, one can stop this process altogether
at low temperature, i.e., formation of the anion and quenching
of the reaction mixture with benzaldehyde at -30 °C f room
temperature produces only mono-dithiane adduct 3. Appar-
ently the bis-dithianyl anion A-2 is even less reactive toward
MDTA, allowing for its selective accumulation in the
reaction mixture. The bar graph in Figure 2 shows the results
of screening for optimal reaction conditions, with the
maximum yield of the bis-dithianyl adduct (67%) achieved
at 20 °C by quenching the anionic reaction mixture with
benzaldehyde after 15 min. Under these conditions the
monoadduct 3 is formed in 22%. The two adducts are either
(6) Kobayashi, S.; Kadokawa, J.; Shoda, S.; Uyama, H. Macromol. Rep.
1991, A28 (Suppl. 1), 1-5.
(7) Doelling, W. Sci. Synth. 2006, 24, 461-516.
Figure 1. ORTEP drawings of bis-dithianes 2a (left) and 2b.
4062
Org. Lett., Vol. 9, No. 20, 2007

â-hydroxyalkyl bis-dithiane to pyran recently described by
Little.10 In cases when benzaldehyde was used as a capping
electrophile, the Hg2+-catalyzed reaction was accompanied
by dehydrative expansion of one of the dithiane rings to
furnish dithiepin enols 7.
Under the same conditions monomeric adduct 3a produced
the expected product of thioacetal hydrolysis, i.e., 1-phenyl-
1-hydroxy-2-heptanone. It is expected that milder methods
of thioacetal deprotection11 would allow for hydrolysis of
both dithiane moieties in 2 without affecting the sensitive
benzylic hydroxyls.
Figure 2. Optimization of the yield of 2a as a function of
temperature and the time before the quenching of the anion with
benzaldehyde.
Attempts to induce tri- or tetramerization of MDTA by
further decreasing the molar excess of phenyllithium reacting
with 1 resulted only in elimination to furnish MDTA in high
yield. During the column purification, the mixture of MDTA
and benzaldehyde underwent a silica gel-catalyzed condensa-
tion, similar to that previously described.12
The reaction is much cleaner and the yields of adducts 2d,e
are much higher, 82-84%.
Ketones can also be used as capping electrophiles, as
exemplified by acetone. However, esters and acyl chlorides
undergo a more complex electron transfer initiated reaction,
similar to the radical reductive ring opening described earlier
for bulky dithianes.8
The particular synthetic significance of these findings lies
in the fact that this multicomponent reaction allows for a
controlled one-pot formation of three carbon-carbon bonds,
furnishing a useful masked hydroxyalkanedione, which,
among other things, can be utilized in a one-pot hydrolysis
and cyclization into 3(2H)-furanonessa central structural
element of several important natural products.9 For example,
the reaction of 1 initiated by BuLi and capped by propanal
produced 5, which was subjected to HgCl2-catalyzed hy-
drolysis in aqueous acetonitrile to give furanone 6.
In conclusion, we have demonstrated a simple and efficient
one-pot multicomponent approach to versatile synthetic
intermediates based on controlled anionic dimerization of
methylenedithiane.
Acknowledgment. We thank the NSF (CHE-640838) for
financial support of this work.
Supporting Information Available: Synthetic procedures
and spectra. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL701847B
(9) (a) Smith, A. B.; Levenberg, P. A.; Jerris, P. J.; Scarborough, R. M.;
Wovkulich, P. M. J. Am. Chem. Soc. 1981, 103, 1501-1513. (b) Sher, F.;
Isidor, J. L.; Taneja, H. R.; Carlson, R. M. Tetrahedron Lett. 1973, 577-
580. (c) Chimichi, S.; Boccalini, M.; Cosimelli, B.; Acqua, F. D.; Viola,
G. Tetrahedron 2003, 59, 5215-5223. (d) Winkler, J. D.; Oh, K.; Asselin,
S. M. Org. Lett. 2005, 7, 387-389.
(10) Nishiguchi, G. A.; Little, R. D. J. Org. Chem. 2005, 70, 5249-
5256.
(11) For milder methods of thioacetal deprotection see: (a) Cossy, J.
Synthesis 1987, 12, 1113-1115 (MCPBA-CF3CO2H). (b) Lucchetti, J.;
Krief, A. Synth. Commun. 1983, 13, 1153-1161 (CuCl2/CuO). (c) Varma,
R. S.; Saini, R. K. Tetrahedron Lett. 1997, 38, 2623-2624 (clay). (d) Epling,
G. A.; Wang, Q. Tetrahedron Lett. 1992, 33, 5909-5912 (photosensitized
by Methylene Green).
(12) (a) Zhang, Q.; Liu, Y.; Wang, M.; Liu, Q.; Hu, J.; Yin, Y. Synthesis
2006, 3009-3014. (b) Pan, W.; Dong, D.; Sun, S.; Liu, Q. Synlett 2006,
1090-1094. (c) Yin, Y.-B.; Wang, M.; Liu, Q.; Hu, J.-L.; Sun, S.-G.; Kang,
J. Tetrahedron Lett. 2005, 46, 4399-4402.
The efficiency of the second step (55%), i.e., hydrolytic
cyclization, is comparable to that of a similar conversion of
(8) Valiulin, R. A.; Kottani, R.; Kutateladze, A. G. J. Org. Chem. 2006,
71, 5047-5049.
Org. Lett., Vol. 9, No. 20, 2007
4063