An efficient furan synthesis using heterogeneous catalysis

Article abstract of DOI:10.1016/j.tetlet.2007.08.102

A wide variety of 3-alkyne-1,2-diols have been found to undergo exceptionally clean 5-endo-dig cyclisations followed by dehydration at ambient temperature to give the corresponding furans in essentially quantitative yields when exposed to 10 mol % of 10%

Full text of DOI:10.1016/j.tetlet.2007.08.102

Tetrahedron Letters 48 (2007) 7709–7712
An efficient furan synthesis using heterogeneous catalysis
Simon J. Hayes,a David W. Knight,a,* Melanie D. Menzies,a Mark O’Halloranb
and Wen-Fei Tana
aSchool of Chemistry, Cardiff University, Main College, Park Place, Cardiff CF10 3AT, UK
bTechnical and Business Development, GlaxoSmithKline, Shewalton Road, Irvine KA11 5AP, UK
Received 20 July 2007; accepted 6 August 2007
Available online 1 September 2007
Abstract—A wide variety of 3-alkyne-1,2-diols have been found to undergo exceptionally clean 5-endo-dig cyclisations followed by
dehydration at ambient temperature to give the corresponding furans in essentially quantitative yields when exposed to 10 mol % of
10% w/w silver(I) nitrate absorbed on silica gel.
Ó 2007 Elsevier Ltd. All rights reserved.
Despite the best efforts of synthetic chemists during
more than a century of research, there still remains a
need to define new methods for the elaboration of
heteroaromatics which have the degree of flexibility
and generality demanded by modern compound discov-
ery programmes. This is significantly exacerbated by the
ever-increasing demands for environmental protection
and hence the use of a clean technology. The importance
of such compounds to the pharmaceutical sector alone is
self-evident from the statistic that some 70% of drug
therapies in use today contain at least one such residue.
If this were not enough, heteroaromatics play key roles
in many other groups of ‘effect’ chemicals, ranging from
plastics, conducting polymers, agrochemicals, flavours
and fragrances and photochromic and other new mate-
rials to name but a few.1
(Scheme 1).2 Clearly, the residual iodine atom used to
induce the cyclisation can subsequently be deployed to
introduce additional groups, especially by using one of
the plethora of palladium-catalysed coupling methods
which are now available,3 as well as the methods based
around halogen–lithium exchange.4
At the time of our first report of this iodocyclisation, the
5-endo-dig cyclisation mode was something of a ‘Cinder-
ella’ reaction, despite being favoured according to Bald-
win’s rules5 and some earlier indications of its viability
using mercury(II) salts to induce cyclisations related to
those shown in Scheme 1.6 There have subsequently
been a number of reports of alternative approaches to
furans,7 which feature 5-endo cyclisations.8 These
include halide-induced cyclisations of 1,4-diaryl-but-
3-yn-1-ones9 and cyclisations wherein Michael additions
of a nucleophile, usually a simple alcohol, complete
formation of furans from conjugated enones following
exposure to a triggering electrophile,10 including iodo-
nium species and also gold(III)10 and copper(I) salts.11
Conjugated ynones have also been transformed into
furans upon exposure to copper(I) salts in a sequence,
which strongly suggests the intermediacy of the corres-
ponding allenes;12 not surprisingly, Hashmi’s group
There are arguably two basic strategies for synthesis in
this area: prepare a suitable acyclic precursor and cyclise
this to give the heteroaromatic target or elaborate a sim-
pler parent heteroaromatic using appropriate combina-
tions of electrophilic and sometimes nucleophilic
reagents. A potentially attractive version of these meth-
ods, which can be regarded as a combination of both
strategies, is to carry out the cyclisation of an acyclic
precursor using a triggering reagent which remains in
the final product and which can subsequently be used
for further homologations. An example of this is our
use of 5-endo-dig iodocyclisations to convert 3-alkyne-
1,2-diols 1 into the corresponding b-iodofurans 2
R2
R2
R1
I
HO
R1
R3
3.I2
K2CO3
70~95%
R3
O
OH
2
1
*
Scheme 1.
Corresponding author. E-mail: knightdw@cf.ac.uk
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.08.102

7710
S. J. Hayes et al. / Tetrahedron Letters 48 (2007) 7709–7712
have shown that both these types of precursors are use-
ful as furan precursors, especially when using gold(III)
salts.13 Allenic intermediates are also often featured in
the base-catalysed cyclisations of similar precursors
developed by Marshall and DuBay.14 Silver(I) salts have
also long been known to be capable of activating allenes
in cyclisation reactions. For example, both alcohol and
acid groups can be thus induced to undergo 5-endo cyc-
lisations to give 2,5-dihydrofurans15 or butenolides,16
respectively. Conjugated allenic ketones and suitable
alkynyl-pyrimidinones are both converted into the cor-
responding furans using silver(I) nitrate,17 following
the lead provided by the Marshall group, who found
that 2-alkynyl-allylic alcohols could be converted into
the corresponding furans very efficiently using silver(I)
nitrate on silica gel.18
ways, as yields were essentially perfect after simply stir-
ring the catalyst with the precursor for 3 h at ambient
temperature.
Much the same pattern of essentially quantitative con-
versions was observed in syntheses of representative
2,4-disubstituted furans 6 from the corresponding 3-
alkyne-1,2-diols 5 (Scheme 3).
OH
R1
R1
R2
AgNO3-SiO2
CH2Cl2, 20 oC, 3 h
R2
O
OH
5
6
a) [1] R1 = Ph; R2 = Bu [98%]
b) [2] R1 = Ph; R2 = Bu [97%]
c) [2] R1 = Me; R2 = Ph [99%]
d) [2] R1 = Me; R2 = Bu [97%; by NMR/GC]
It was against this background that we began to investi-
gate the prospects for achieving the conversion of 3-
alkyne-1,2-diols 1 (see Scheme 1) into furans using metal
catalysts.6 Our ultimate aim was to define a hetero-
geneous catalyst, which could be recycled and which
would offer all of the advantages usually associated with
such reagents, in contrast to the related homogeneous
catalysts such as gold(III) salts. Although spectacularly
successful in triggering 5-endo-dig and other cyclisa-
tions, their removal might give rise to problems, espe-
cially on scale-up.13 In the event, we were delighted to
find that 10% w/w silver(I) nitrate on silica gel, a mate-
rial more often associated with the chromatographic
separation of alkene stereoisomers, was especially effec-
tive. The initial results leading to 2,5-disubstituted
furans 4 are shown in Scheme 2. The numbers in square
brackets ([1]; [2]) refer to the method of precursor syn-
thesis, as explained below. The yields are the averages
of a number of runs in each case.
OH
OH
AgNO3
O
[97%]
7
8
Scheme 3.
Due to its volatility, furan 6d was assayed by a combina-
tion of NMR and GC analysis. An indication of the
mildness of the procedure is given by the completely
clean conversion of the citronellol-derived diol 7 into
the sesquirosefuran isomer 8 in essentially quantitative
yield. These results are particularly pleasing as the 2,4-
disubstitution pattern is quite difficult to obtain. Our
method featuring 2-lithio-4-methylfuran20 is one of the
few available for the elaboration of terpenoid targets
in this area.21
A final series featured the synthesis of 2,3,5-trisubsti-
tuted furans 10 from the corresponding alkyne-diols 9
(Scheme 4). Again, all yields were essentially quantita-
tive, no matter if the substituents were alkyl or aryl
groups or large t-butyl groups (10g). The only slight
However, as far as it could be judged, these conversions
were quantitative and the slight losses were simply
mechanical or due to product volatility; we could not
detect any by-product formation. Although we carried
out all of these cyclisations in dry dichloromethane, hex-
ane worked equally well and in some cases, cyclisations
were faster in this solvent.18 The rate of cyclisation could
also be enhanced both by heating and by passage
through a column of catalyst contained in a microwave
oven.19 However, while this was not at the expense of
by-product formation, it was judged not worthwhile to
modify an already quantitative conversion in these
OH
R2
R1
R2
R1
R3
AgNO3-SiO2
CH2Cl2, 20 oC, 3 h
R3
O
OH
9
10
a) R1 = R2 = Ph; R3 = Bu [98%]
b) R1 = R2 = Me; R3 = Bu [>95% by NMR; 71%]
c) R1, R2 = (CH2)4; R3 = Bu [99%]
d) R1 = R2 = Ph; R3 = CH2OTBS [94%]*
e) R1 = Me; R2 = CCBu; R3 = Bu [93%]
f) R1 = R2 = Me; R3 = Ph [99%]
HO
R2
AgNO3-SiO2
CH2Cl2, 20 oC, 3 h
R1
R1
R2
OH
O
g) R1 = R2 = Ph; R3 = tBu [98%]
3
4
h) R1, R2 = (CH2)4; R3 = Ph [97%]
i) R1 = R2 = Me; R3 = (CH2)3OBn [99%]
j) R1 = R2 = Ph; R3 = (CH2)3OBn [99%]
k) R1 = R2 = R3 = Ph [98%]
a) [1] R1 = Ph; R2 = Bu [96%]
b) [2] R1 = Me; R2 = Bu [>95%, by NMR/GC]
c) [2] R1 = Me; R2 = Ph [98%]
d) [1] R1 = Ph; R2 = (CH2)2OH [97%]
e) [1] R1 = Me; R2 = (CH2)2OTBDPS [99%]
* isolated as partly desilylated product.
All starting materials prepared by route [2].
Scheme 2.
Scheme 4.

S. J. Hayes et al. / Tetrahedron Letters 48 (2007) 7709–7712
7711
complication was the partial loss of the t-butyldimethyl-
silyl (TBS) protecting group during cyclisation of the
propargyl alcohol derivative 9d. However, the combined
yield, after chromatographic separation of the O-silyl
derivative and the corresponding free alcohol, was still
94%. There was no interference from competing 5-exo-
dig cyclisations involving the distal O-benzyl functions
in reactions of the pentynol derivatives 9i,j (Scheme 4).
sure of triol 14 to AgNO3–SiO2 under the usual condi-
tions (see below) led to a ca. 1:1 mixture of the
expected furan 15 together with ylidene-tetrahydrofuran
16. Fortunately, this event could be blocked by masking
the primary hydroxyl as its TBS ether 17, in which case
only the furan product 18 was obtained,22 in a similar
fashion to the smooth reactions of the O-benzyl deriva-
tives 9i,j (Scheme 4). This bodes well, as silyl groups can
easily be lost in related cyclisations of this type.22
Some limitations:- Inevitably, some limitations to these
otherwise extremely efficient cyclisations came to light.
Starting material synthesis:- There are clearly a variety
of approaches to the precursor 3-alkyne-1,2-diols 3, 5,
7 and 9. During this study, two approaches were used:
route [1] (Scheme 2) comprised a Sonogashira cou-
pling23 between the corresponding 1-haloalkene and
1-alkyne followed by regioselective bis-hydroxylation
of the resulting conjugated enyne,24 as previously
employed by us.2 Route [2] (Schemes 2–6) comprised
condensations between a-silyloxy-aldehydes or ketones
and lithio- or magnesio-acetylides (tetrahydrofuran,
0 °C). To obviate the need for a deprotection step, when
the 1-alkyne was volatile (e.g., 1-hexyne) and hence easily
removed from the products, 2.2 equiv of the lithio-acetyl-
ide was reacted directly with the hydroxy-carbonyl
compound. In all cases, overall yields were in excess of
80% for one or two steps.
The most serious was the complete failure of 1-alkyne-
3,4-diols 11 to undergo cyclisation to give examples of
the 2,3-disubstitution pattern 12 (Scheme 5). The corres-
ponding trimethylsilylated alkynes underwent proto-
desilylation. Efforts to overcome this limitation are
in progress.
The method also appears to be incompatible with the
presence of divalent sulfur: exposure of the 2-thienyl
derivative 13 to silver nitrate on silica gel caused com-
plete decomposition (Scheme 6). Therefore, thioether
functions in general would also probably not survive;
it seems unlikely that anything can be done to address
this limitation.
Finally, 5-exo-dig cyclisations can compete with the
desired 5-endo processes (Scheme 7). For example, expo-
The actual cyclisations were carried out by stirring a
solution of the 3-alkyne-1,2-diol in dichloromethane
with 10 mol % of commercial 10% w/w silver(I) nitrate
on silica gel at ambient temperature for around 3 h
(TLC monitoring), followed by filtration through a
silica gel pad eluted with dichloromethane. The only pre-
caution was to exclude light from the reaction mixture
by wrapping the flask in aluminium foil. Filtration
through silica gel rather than using other methods was
necessary to remove the small amount of silver salt,
which was found to leach from the catalyst. If this slight
weight loss was taken into account, the catalyst isolated
by paper filtration retained essentially the same reactiv-
ity for at least two subsequent runs with the same or
other 3-alkyne-1,2-diols.
OH
R2
R1
R2
R1
H
AgNO3
O
OH
11
12
Scheme 5.
OH
AgNO3
Decomposition
S
This highly efficient furan synthesis would seem to have
the potential to contribute an environmentally friendly
method to the synthetic armoury in this and perhaps
other areas of heterocyclic synthesis. Efforts to illustrate
OH
13
Scheme 6.
OH
HO
O
HO
Ph
AgNO3
HO
Ph
O
+
Ph
ca. 1:1
OH
OH
14
15
16
OTBS
TBSO
HO
Ph
AgNO3
[97%]
Ph
O
OH
17
18
Scheme 7.

7712
S. J. Hayes et al. / Tetrahedron Letters 48 (2007) 7709–7712
syntheses, see: Houpis, I. N.; Choi, W. B.; Reider, P. J.;
Molina, A.; Churchill, H.; Lynch, J.; Volante, R. P.
Tetrahedron Lett. 1994, 35, 9355–9358; Jameba, Z.;
Maklad, N.; Robins, M. J. Nucleosides, Nucleotides,
Nucleic Acids 2005, 24, 1729–1733.
these aspects are underway, as are modifications aimed
at obviating the drawback of the slight silver leaching
during the final filtration.25
12. Kel’in, A. V.; Gevorgyan, V. J. Org. Chem. 2002, 67, 95–
98.
Acknowledgements
13. Hashmi, A. S. K.; Schwarz, L.; Choi, J.-H.; Frost, T. M.
Angew. Chem., Int. Ed. 2000, 39, 2285–2288; Hashmi, A.
S. K.; Sinha, P. Adv. Synth. Catal. 2004, 346, 432–437;
Sromek, A. W.; Rubina, M.; Gevorgyan, V. J. Am. Chem.
Soc. 2005, 127, 10500–10501; Zhou, C.-Y.; Chan, P. W.
H.; Che, C.-M. Org. Lett. 2006, 8, 325–328; For summa-
ries, see: Hashmi, A. S. K. Gold Bull. 2004, 37, 51–65;
Hashmi, A. S. K. Angew. Chem., Int. Ed. 2005, 44, 6990–
6993.
We are very grateful to the EPSRC Mass Spectrometry
Service, University College Swansea, for the provision
of high resolution mass spectrometric data and to
GSK and the EPSRC for financial support.
References and notes
1. For excellent general overviews, see the relevant Chapters:
Comprehensive Heterocyclic Chemistry II; Katritzky, A.
R., Rees, C. W., Scriven, E. F. V., Eds.; Pergamon Press:
Oxford, 1996.
2. Bew, S. P.; Knight, D. W. J. Chem. Soc., Chem. Commun.
1996, 1007–1008; El-Taeb, G. M. M.; Evans, A. B.; Knight,
D. W.; Jones, S. Tetrahedron Lett. 2001, 42, 5945–5949.
3. Li, J. J.; Gribble, G. W. Palladium in Heterocyclic
Chemistry. In Tetrahedron Organic Chemistry Series;
Pergamon Press: Oxford, 2000; Vol. 20.
4. Gschwend, H. W.; Rodriguez, H. R. Org. React. 1979, 26,
1–373; Snieckus, V. Chem. Rev. 1990, 90, 879–933.
5. Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734–
736; Baldwin, J. E.; Cutting, J.; Dupont, W.; Kruse, L.;
Silberman, L.; Thomas, R. C. J. Chem. Soc., Chem.
Commun. 1976, 736–738.
6. Fabrycy, A.; Goshchinskii, S. Zh. Obsch. Khim. 1959, 29,
81–86; Fabrycy, A.; Kubala, J. Zh. Obsch. Khim. 1961, 31,
476–479; Fabrycy, A.; Wichert, Z. Zh. Obsch. Khim. 1979,
49, 2499–2504.
7. For recent reviews of furan synthesis, see: Hou, X. L.;
Yang, Z.; Wong, H. N. C. In Progress in Heterocyclic
Chemistry; Gribble, G. W., Gilchrist, T. L., Eds.; Springer:
Oxford, 2002; Vol. 14, pp 139–179; Gilchrist, T. L. J.
Chem. Soc., Perkin Trans. 1 2001, 2491–2515.
8. Knight, D. W. In Progress in Heterocyclic Chemistry;
Gribble, G. W., Gilchrist, T. L., Eds.; Pergamon Press:
Oxford, 2002; Vol. 14, pp 19–51; Calculations suggest that
5-endo-dig ring closures are rapid and proceed via a late
transition state: Elliott, R. J. Theochem.-J. Mol. Struct.
1985, 22, 79–83.
9. Sniady, A.; Wheeler, K. A.; Dembinski, R. Org. Lett.
2005, 7, 1769–1772; For related iodocyclisations leading to
furanopyrimidines and furopyridines, see: Rao, M. S.;
Esho, N.; Sergeant, C.; Dembinski, R. J. Org. Chem. 2003,
68, 6788–6790; Aillaud, I.; Bossharth, E.; Conreaux, D.;
Desbordes, P.; Monteiro, N.; Balme, G. Org. Lett. 2006, 8,
1113–1116 (see also Refs. 11 and 17).
10. Yao, T.; Zhang, X.; Larock, R. C. J. Am. Chem. Soc.
2004, 126, 11164–11165; Yao, T.; Zhang, X.; Larock, R.
C. J. Org. Chem. 2005, 70, 7679–7685; Lin, Y.; Zhou, S.
Org. Lett. 2005, 7, 4609–4611.
14. Marshall, J. A.; DuBay, W. J. J. Org. Chem. 1993, 58,
3435–3443 (and references cited therein).
15. Olsson, L. I.; Claesson, A. Synthesis 1979, 743–745;
Marshall, J. A.; Wang, X. J. J. Org. Chem. 1990, 55,
2995–2996; Saimoto, H.; Yasui, M.; Ohrai, S.; Oikawa,
H.; Yokoama, K.; Shigemasa, Y. Bull. Chem. Soc. Jpn.
1999, 72, 279–284, and references cited therein.
16. Marshall, J. A.; Pinney, K. G. J. Org. Chem. 1993, 58,
7180–7184; Marshall, J. A.; Bartley, G. S.; Wallace, E. M.
J. Org. Chem. 1996, 61, 5729–5735.
17. Hashmi, A. S. K.; Schwarz, L. Chem. Ber. 1997, 130,
1449–1456; Aucagne, V.; Amblard, F.; Agrofoglio, L. A.
Synlett 2004, 2406–2408; Amblard, F.; Aucagne, V.;
Guenot, P.; Schinazi, R. F.; Agrofoglio, L. A. Biorg.
Med. Chem. 2005, 13, 1239–1248.
18. Marshall, J. A.; Sehon, C. A. J. Org. Chem. 1995, 60,
5966–5968; Marshall, J. A.; Sehon, C. A. Org. Synth.
1998, 76, 263–270.
19. Bagley, M. C.; Jenkins, R. L.; Lubinu, M. C.; Mason, C.;
Wood, R. J. Org. Chem. 2005, 70, 7003–7006. We are
grateful to Mr. Oliver John for obtaining these results.
20. Knight, D. W.; Rustidge, D. C. J. Chem. Soc., Perkin
Trans. 1 1981, 679–683.
21. For example, see: Bernsmann, H.; Hungerhoff, B.; Fech-
ner, R.; Fro¨hlich, R.; Metz, P. Tetrahedron Lett. 2000, 41,
1721–1724; Wenkert, E.; Khatuya, H.; Klein, P. S.
Tetrahedron Lett. 1999, 40, 5171–5174; Brickwood, A.
C.; Drew, M. G. B.; Harwood, L. M.; Ishikawa, T.;
Marais, P.; Morisson, V. J. Chem. Soc., Perkin Trans. 1
1999, 913–921; Liu, H.; Gayo, L. M.; Sullivan, R. W.;
Choi, A. Y. H.; Moore, H. W. J. Org. Chem. 1994, 59,
3284–3288; Hayakawa, K.; Nagatsugi, F.; Kanematsu, K.
J. Org. Chem. 1988, 53, 860–863.
22. See for example Kang, S. H.; Lee, S. B. Tetrahedron Lett.
1993, 34, 7579–7582; Knight, D. W.; Shaw, D. E.; Staples,
E. R. Eur. J. Org. Chem. 2004, 1973–1982.
23. Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron
Lett. 1975, 4467–4470; Sonogashira, K. J. Organomet.
Chem. 2002, 653, 46–49.
24. Jeong, K. S.; Sjo, P.; Sharpless, K. B. Tetrahedron Lett.
1992, 33, 3833–3836.
25. This methodology is the subject of an International Patent
Application: PCT/GB2006/001048.
11. Patil, N. T.; Wu, H.; Yamamoto, Y. J. Org. Chem. 2005,
70, 4531–4534; For related furano-pyridine and uracil