Practical alternatives for the synthesis of β-iodofurans by 5-endo-dig cyclisations of 3-alkyne-1,2-diols

Article abstract of DOI:10.1016/S0040-4039(01)01112-1

Iodocyclisations of 3-alkyne-1,2-diols, obtained from acetylides and α-hydroxy-ketones or esters, give generally excellent yields of β-iodofurans by 5-endo-dig cyclisation followed by dehydration.

Full text of DOI:10.1016/S0040-4039(01)01112-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 5945–5948
Practical alternatives for the synthesis of b-iodofurans by
-endo-dig cyclisations of 3-alkyne-1,2-diols
5
Gamila M. M. El-Taeb, Ann B. Evans, Simon Jones and David W. Knight*
Department of Chemistry, Cardiff University, PO Box 912, Cardiff CF10 3TB, UK
Received 1 June 2001; accepted 21 June 2001
Abstract—Iodocyclisations of 3-alkyne-1,2-diols, obtained from acetylides and a-hydroxy-ketones or esters, give generally
excellent yields of b-iodofurans by 5-endo-dig cyclisation followed by dehydration. © 2001 Elsevier Science Ltd. All rights
reserved.
6
Halogenated heteroaromatics have always occupied an
important position as highly valuable intermediates in
general organic synthesis, a status much enhanced by
recent developments in radical chemistry and especially
considerable synthetic potential. More recent examples
generally feature palladium-catalysed cyclisations,
exemplified by a useful approach to indoles from 2-
1,7
alkynylaniline derivatives.
1
in transition metal-catalysed coupling reactions. In
general, the reactivity of five-membered heteroaromat-
ics dictates that a-halo derivatives are usually readily
available whilst the preparation of b-halo isomers has
to rely on suitably powerful directing effects from exist-
ing a-substituents if viable levels of regioselection are to
It was against this background that we developed the
new approach to b-iodofurans summarised in Scheme
8
1
. The key step is an iodine-induced 5-endo-dig cyclisa-
tion of the acetylenic diols 2, readily prepared by highly
regioselective bis-hydroxylations of the corresponding
2
be achieved. Further, in many cases, subsequent
9
enynes 1. Dehydration of the presumed intermediate
removal of such directing or blocking groups is neces-
sary to provide a desired target by, for example, decar-
boxylation of an a-carboxylic acid function. For some
time, we have been investigating the characteristics of
electrophile-driven 5-endo-trig cyclisations which turn
out to be extremely useful for the stereoselective elabo-
ration of highly substituted tetrahydrofurans and
hydroxy-dihydrofurans 3 was evidently very rapid, as
these were not observed; only good to excellent yields
of the iodofurans 4 were isolated. In general, b-iodo-
furans are not easy to prepare selectively, especially
when the two existing a-substituents are electronically
similar. The success of this type of cyclisation, as well
as of a variety of subsequent homologation reactions
involving replacement of the iodine atom by carbon-
3
pyrrolidines. We wondered whether such cyclisation
methodology could be extended to the related 5-endo-
dig mode which, in contrast to 5-endo-trig processes, is
8
based substituents, led us to investigate alternative and
4
potentially more practical approaches to intermediates
2 and the viability or otherwise of iodocyclisations of
these new precursors. Herein, we report that condensa-
tions between acetylides and both a-hydroxy-ketones
and a-hydroxy-esters satisfy these criteria.
favoured according to Baldwin’s rules. Despite this,
such endo-dig cyclisations have been much less
exploited in general, as has been highlighted recently by
5
Carreira, even though there are a number of hints in
the older literature which indicate that these could have
HO
HO
R1
I
I
R2
R2
R1
R1
R2
R1
R2
OH
O
O
1
2
3
4
Scheme 1.
Keywords: iodocyclisation; 5-endo-dig; b-iodofurans; 3-alkyne-1,2-diols; furans.
*
Corresponding author.
040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
0
PII: S0040-4039(01)01112-1

5
946
G. M. M. El-Taeb et al. / Tetrahedron Letters 42 (2001) 5945–5948
Our starting point was a concern that, during our initial
study, no examples involving the presence of a second
the usual iodocyclisation conditions in acetonitrile led
to a 71% isolated yield of the fully substituted furan 18.
8
b-substituent, and thus cyclisations of diols in which
one hydroxyl was tertiary, had been evaluated (see
Scheme 1). We chose to exemplify this using a methyl
group, since the b-methylfuran substructure is a feature
of many furanoterpenes. Condensation between the
Similarly, furoin 19 was converted into the alkyne-diol
20 (Scheme 5). Iodocyclisation failed in the usual sol-
vents but delivered a 25% isolated yield of the trifuryl-
furan 21 if carried out in ethyl acetate, the poor return
reflecting both product sensitivity and iodination of the
10
tetrahydropyranyl ether 5 of hydroxyacetone and 1-
hexynyl lithium followed by deprotection gave an excel-
lent yield of the acetylenic diol 6a. Subsequent exposure
to 3 equivalents each of iodine and sodium hydrogen
carbonate in acetonitrile at 0°C led to a 56% isolated
yield of the hoped for iodofuran 7a (Scheme 2). Simi-
larly, by starting with phenylacetylene, the 2-phenyl
analogue 7b was obtained in 61% isolated yield. These
returns were slightly lower than expected. Combined
GC–MS analysis of the reaction mixtures suggested
these were higher (ca. 80%) and also revealed slower
formation of additional products, the diiodides 8, which
presumably arise by direct iodination of the initial
iodofurans 7 at the vulnerable free a-position. A brief
optimisation study showed that the cyclisations were
faster in dichloromethane and were also viable in tetra-
hydrofuran. For ease of work up, we used the former,
when cyclisation was complete after approximately 5 h
at 0°C. After this, substantial quantities of the diiodides
12
vacant a-positions of the furyl residues.
We then examined the prospects of using a-hydroxy-
esters as starting materials, with a view to carrying a
second alkyne function through the key cyclisation
step. Thus, the bis-acetylenic diols 24 were prepared by
condensations between the protected hydroxy-esters 22
and 2 equivalents of a lithio acetylide or between the
parent hydroxy-esters 23 and 3 equivalents of acetylide
(
Scheme 6).
Overall yields for the former, two-step process were
6
6
2–83% while the latter direct condensation delivered
6–86% isolated yields of the diols 24. As detailed in
Scheme 6, these then underwent smooth cyclisations in
acetonitrile to provide excellent yields of the acetylenic
b-iodofurans 25. The cyclisations were remarkably
clean and could be carried out with equal facility on
3
t
both the O-silyl precursors [24; R =SiBu Me ] or the
8
were formed; the lower isolated yields are also proba-
bly a consequence of product volatility.
2
3
free alcohols [24; R =H] although, in the former case,
it was essential that ca. 5% water was present. Both sets
of substrates gave ca. 90% isolated yields of the iodo-
furans during 1 h or less at ambient temperature. By
Clearly, when the remaining a-position in the final
b-iodofurans is substituted, then further iodination is
not a problem. To probe the viability of such substrates
in this chemistry, we examined similar chemistry using
commercial 3-hydroxy-2-butanone 9 and 2-hydroxycy-
8
further homologations, such compounds may find use
in the elaboration of highly extended p-systems.
1
0
In summary, this modified approach to b-iodofurans
offers flexibility, is usually very efficient and should be
amenable to large scale synthesis. On the negative side,
the availability of a-hydroxy-ketones or -esters could
clohexanone as starting materials.
In the interests of atom efficiency and time, these were
treated directly with solutions of lithio-alkynes (1.1
equiv.) containing an additional equivalent of butyl-
lithium, to obviate the need to protect the free hydroxyl
group. This resulted in isolation of the required diols 11
and 12 in 70–75% yields (Scheme 3). The key cyclisa-
tions were carried out in dichloromethane, again using
R
I
O
HO
R
OH
OH
11
O
3
equivalents each of I and NaHCO , mixed at 0°C
2 3
a) R = Bu
b) R = Ph
followed by stirring without further cooling for 3 h. A
simple work-up then gave the b-iodofurans 13 and 14
in 88–93% isolated yields.
9
13
3
R
I
HO
O
Other possibilities were also examined. Thus, the O-
silyl derivative 15 of benzoin condensed smoothly with
lithiated trimethylsilylacetylene to provide the alkyne-
diol 16 following desilylation using fluoride; subsequent
R
O
OH
OH
10
12
14
11
Sonogashira coupling with 2-iodothiophene then gave
the cyclisation precursor 17 (Scheme 4). Exposure to
Scheme 3.
I
I
O
HO
R
a) R = Bu
b) R = Ph
ThPO
R
I
R
OH
O
O
5
6
7
8
Scheme 2. Reagents and conditions: (i) RCCLi, −78°C, THF; (ii) 3 equiv. I and NaHCO , CH Cl , 0°C.
2
3
2
2

G. M. M. El-Taeb et al. / Tetrahedron Letters 42 (2001) 5945–5948
5947
Ph
I
S
HO
Ph
HO
Ph
Ph
Ph
O
Ph
O
OTBS
Ph
OH
Ph
OH
S
15
16
17
18
Scheme 4.
I
O
O
OH
O
O
O
O
OH
OH
O
O
O
O
19
20
21
Scheme 5.
CO Me
Yields of 25 from 24
2
1
2
3
R1
R = Me; R = Bu; R = TBS [82%]
OTBS
R2
1
2
3
R2
R2
R = Me; R = Bu; R = H [93%]
2
2
OH
1
2
3
R = Me; R = Ph; R = TBS [90%]
I
1
2
3
R = Me; R = Bu; R = H [93%]
1
2
3
CO Me
R1
OR3
R1
R2
R = Ph; R = Bu; R = H [87%]
2
O
1
2
3
R = R = Ph; R = TBS [90%]
R1
OH
1
2
3
R = R = Ph; R = H [94%]
24
25
1
2
3
R = Me; R = CH OTBDPS; R = TBS [93%]
2
3
2
Scheme 6.
present a problem in some cases, although there is a
plethora of methods available for the synthesis of these
starting materials. The alkynylfurans 25 will always
3. (a) Bedford, S. B.; Bell, K. E.; Bennett, F.; Hayes, C. J.;
Knight, D. W.; Shaw, D. E. J. Chem. Soc., Perkin Trans.
1 1999, 2143; (b) Bew, S. P.; Barks, J. M.; Knight, D. W.;
Middleton, R. J. Tetrahedron Lett. 2000, 4447–4451; (c)
Jones, A. D.; Knight, D. W.; Hibbs, D. E. J. Chem. Soc.,
Perkin Trans. 1 2001, 1182.
2
have an element of symmetry (R ) as 2 equivalents of a
single lithio-alkyne are used in their preparation; even if
this were not the case, iodocyclisation of diol deriva-
tives 24 having differing alkyne substituents may well
not be regiospecific. Otherwise, this chemistry should
make many types of highly substituted b-iodofurans
4
5
6
. Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1975, 734,
738.
. Aschwanden, P.; Frantz, D. E.; Carreira, E. M. Org.
Lett. 2000, 2, 331.
. (a) Fabritsy, A.; Goshchinskii, S. Zh. Obshch. Khim.
13
very readily available.
1
955, 29, 81 (Chem. Abstr. 1959, 53, 21868); (b) Fabrycy,
A.; Wichert, Z. Rocz. Chem. 1977, 51, 249 (Chem. Abstr.
977, 87, 68471); (c) See also Fukuda, Y.; Shiragami, H.;
Utimoto, K.; Nozaki, H. J. Org. Chem. 1991, 56, 5816;
Acknowledgements
1
We thank the EPSRC Mass Spectrometry Service, Uni-
versity College, Swansea for the provision of high reso-
lution mass spectral data and the Jowfe Oil Company
Ltd. and the EPSRC for financial support.
(
(
d) Sheng, H.; Lin, S.; Huang, Y. Synthesis 1987, 1022;
e) Marshall, J. A.; Wang, X.-J. J. Org. Chem. 1991, 56,
9
60; (f) Marshall, J. A.; Dubay, W. J. J. Org. Chem.
1993, 58, 3435 and 3602.
7
. See, for example, (a) Taylor, E. C.; Katz, A. H.; Salgado-
Zamora, H.; McKillop, A. Tetrahedron Lett. 1985, 26,
References
5
963; (b) Arcadi, A.; Cacchi, S.; Marinelli, F. Tetra-
1
2
. Li, J. J.; Gribble, G. W. In Tetrahedron Organic Chem-
istry Series. Palladium in Heterocyclic Chemistry;
Elsevier Science Ltd: Oxford, 2000.
hedron Lett. 1989, 30, 2581; For acid-catalysed cyclisa-
tions see, for example, (c) Tischler, A. N.; Lanza, T. J.
Tetrahedron Lett. 1986, 27, 1653; (d) Sakamoto, T.;
Kondo, Y.; Yamanaka, H. Heterocycles 1988, 27, 2225.
. Heaney, H.; Ahn, J. S. Comp. Heterocycl. Chem. 1996, 2,
2
97.

5
948
G. M. M. El-Taeb et al. / Tetrahedron Letters 42 (2001) 5945–5948
8
9
. Bew, S. P.; Knight, D. W. J. Chem. Soc., Chem. Com- 11. Sonogashira, K. Comp. Org. Synth. 1991, 33, 521.
mun. 1996, 1007.
12. Unfortunately, all subsequent attempts to couple this
sensitive compound with various 2-furylmetals, under
palladium catalysis, failed to deliver tetrafurylfuran.
13. Satisfactory spectroscopic and analytical data were
obtained for all compounds reported herein.
. Jeong, K.-S.; Sj o¨ , P.; Sharpless, K. B. Tetrahedron Lett.
1
992, 33, 3833.
1
0. Wakabayshi, Y.; Fakuda, Y.; Shirgami, H.; Utimoto, K.;
Nozaki, H. Tetrahedron 1985, 41, 3659.
.