Regioselective Preparation of 2,4-, 3,4-, and 2,3,4-Substituted Furan Rings. 1. [1,4] O → C and [1,4] C → O Silyl Migrations of Silyl Ethers and Esters Attached to Furan and Thiophene Rings

Article abstract of DOI:10.1021/jo971097j

[1,4] O → C and [1,4] C → O rearrangements are described for a variety of furans and thiophenes. Treatment of 3-((silyloxy)methyl)furans and -thiophenes with n-BuLi in HMPA provided 2-silylated-3-(hydroxymethyl)furans and -thiophenes in good to excellent yields. The reaction was shown by crossover studies to proceed via an intramolecular [1,4] O → C silyl migration. Silyl esters of 3-furoic acids also underwent an intramolecular [1,4] O → C silyl migration to provide 2-silylated-3-furoic acids in moderate to good yield when treated with a mixture of LDA and HMPA. Finally, the above silyl migrations were shown to be reversible. Treatment of 2-silylated-3-(hydroxymethyl)-furans and -thiophenes with NaH in DMF provided 3-((silyloxy)methyl)furans and -thiophenes in excellent yields via a [1,4] C → O silyl migration. The [1,4] C → O silyl migration was also shown to be an intramolecular process by a crossover study.

Full text of DOI:10.1021/jo971097j

J . Org. Chem. 1997, 62, 8741-8749
8741
Regioselective P r ep a r a tion of 2,4-, 3,4-, a n d 2,3,4-Su bstitu ted
F u r a n Rin gs. 1. [1,4] O f C a n d [1,4] C f O Silyl Migr a tion s of
Silyl Eth er s a n d Ester s Atta ch ed to F u r a n a n d Th iop h en e Rin gs
Edward Bures, Patrick G. Spinazze´, Giovanna Beese, Ian R. Hunt, Christine Rogers, and
Brian A. Keay*
Department of Chemistry, University of Calgary, Calgary, Alberta, Canada T2N 1N4
Received J une 17, 1997X
[1,4] O f C and [1,4] C f O rearrangements are described for a variety of furans and thiophenes.
Treatment of 3-((silyloxy)methyl)furans and -thiophenes with n-BuLi in HMPA provided 2-silylated-
3-(hydroxymethyl)furans and -thiophenes in good to excellent yields. The reaction was shown by
crossover studies to proceed via an intramolecular [1,4] O f C silyl migration. Silyl esters of 3-furoic
acids also underwent an intramolecular [1,4] O f C silyl migration to provide 2-silylated-3-furoic
acids in moderate to good yield when treated with a mixture of LDA and HMPA. Finally, the
above silyl migrations were shown to be reversible. Treatment of 2-silylated-3-(hydroxymethyl)-
furans and -thiophenes with NaH in DMF provided 3-((silyloxy)methyl)furans and -thiophenes in
excellent yields via a [1,4] C f O silyl migration. The [1,4] C f O silyl migration was also shown
to be an intramolecular process by a crossover study.
Of the many disubstituted furan patterns possible, the
3,4-disubstituted system is probably most difficult to
prepare since furan rings preferentially lithiate1 and add
electrophiles2 in the C-2 or C-5 positions. The second
most difficult is the 2,4-disubstituted pattern, since
3-substituted furans preferentially lithiate at the C-2
position, thereby providing 2,3-disubstituted furan rings.1
Since many natural products incorporate furan rings with
3,4-, 2,4-, or 2,3,4-substitution patterns,3 a need has
emerged for developing synthetic routes toward the
preparation of furan rings containing these substitution
patterns. Some elaborate methods for preparing 3,4-
disubstituted furans have been reported. These include
Diels-Alder/retro-Diels-Alder chemistry,4 chemical modi-
fications of 3,4-furandicarboxylic acid,5 and the synthetic
modification of acyclic precursors.6
butyldimethylsilyl)-3-(hydroxymethyl)furan (3), prepared
by a [1,4] O f C silyl migration of 1, when treated with
2.2 equiv of n-BuLi in DME or THF/HMPA at 0 °C for 1
h, and treated with an electrophile, provided 2-(tert-
butyldimethylsilyl)-3-(hydroxymethyl)-4-substituted furans
(5) in good to excellent yield (Scheme 1). Subsequent
removal of the C-2 silyl group with tetra-n-butylammo-
nium fluoride or via a [1,4] C f O silyl rearrangement
(NaH, DMF) provided the corresponding 3,4-disubstitut-
ed furans (7 or 8) respectively. Thus a direct, high-
yielding synthesis of 3,4-disubstituted furans via lithia-
tion has been achieved. A few years later, we showed
that silyl esters of 3-furoic acid (2) undergo a similar
sequence providing 4-substituted-3-furoic acids in good
to excellent yields (Scheme 1, see 2 f 4 f 6 f 9).8
In this and the following paper, we provide a full
account of our work on the silyl migrations and regiose-
lective lithiations. This paper will provide a full account
of (a) the [1,4] O f C silyl rearrangement of 3-((silyl-
oxy)methyl)furans and -thiophenes,9 (b) the [1,4] O f C
silyl migration of silyl esters of 3-furoic acid and
3-thiophenecarboxylic acids;10 and (c) the [1,4] C f O silyl
rearrangement of 2-silylated-3-(hydroxymethyl)-4-sub-
stituted furans and thiophenes.11 The next paper (i.e.,
part 2) will discuss in detail regioselective lithiations of
2-silylated-3-substituted furan systems.
In 1988 we reported the first successful regiospecific
C-4 lithiation of a 2,3-disubstituted furan.7 Thus 2-(tert-
X Abstract published in Advance ACS Abstracts, November 15, 1997.
(1) Gschwend, H. W.; Rodriguez, H. R. Org. React. 1979, 26, 1.
(2) (a) Bosshard, P.; Eugster, C. H. Adv. Heterocycl. Chem. 1966, 6,
378. (b) Sargent, M. V.; Cresp, T. M. Compr. Org. Chem. 1979, 4, 693.
(c) Dean, F. M. Adv. Heterocycl. Chem. 1982, 30, 167. (d) Dean, F. M.
Adv. Heterocycl. Chem. 1982, 31, 238. (e) Donnelly, D. M. X.; Meegan,
M. J . Compr. Heterocycl. Chem. 1984, 4, 657. (f) Sargent, M. V.; Dean,
F. M. Compr. Heterocycl. Chem. 1984, 4, 599. (g) Heany, H.; Ahn, J . S.
Compr. Heterocycl. Chem. II 1996, 2, 297. (h) Freidrichsen, W. Compr.
Heterocycl. Chem. II 1996, 2, 351.
(3) For some recent examples, see: (a) Keay, B. A.; Dibble, P. W.
Compr. Heterocycl. Chem. II 1996, 2, 395 and references therein. For
the use of 3,4-disubstituted furans as intermediates in synthesis, see:
(b) Kanematsu, K.; Soejima, S.; Wang, G. Tetrahedron Lett. 1991, 32,
4761. (c) Eberbach, W.; Laber, N.; Bussenius, J .; Fritz, H.; Rihs, G.
Chem. Ber. 1993, 126, 975.
(4) (a) Weis, C. D. J . Org. Chem. 1962, 27, 3514 and 3520. (b)
Hutton, J .; Potts, B.; Southern, P. F. Synth. Commun. 1979, 9, 789.
(c) Ko¨nig, H.; Graf, F.; Weberndo¨rfer, V. Liebigs Ann. Chem. 1981, 688.
(d) Ansell, M. F.; Caton, M. P. L.; North, P. C. Tetrahedron Lett. 1981,
22, 1727. (e) Yang, Y.; Wong, H. N. C. Tetrahedron 1994, 50, 9583
and references therein. (f) Baba, Y.; Sakamoto, Y.; Soejima, S.;
Kanematsu, K. Tetrahedron 1994, 50, 5645 and references therein.
(g) Naganawa, A.; Ichikawa, Y.; Isobe, M. Tetrahedron 1994, 50, 8969.
(5) (a) Corey, E. J .; Crouse, D. N.; Anderson, J . E. J . Org. Chem.
1975, 14, 2140. (b) Schultz, A. G.; Motyka, L. A. J . Am. Chem. Soc.
1982, 104, 5800. (c) Schultz, A. G.; Motyka, L. A.; Plummer, M. J . Am.
Chem. Soc. 1986, 108, 1056.
It was reasoned that a new synthetic pathway to 3,4-
disubstituted furans could be possible by employing a
carefully modified 2,3-disubstituted furan ring such that
lithiation would occur at the C-4 position and not the C-5
(7) (a) Bures, E. J .; Keay, B. A. Tetrahedron Lett. 1988, 29, 1247.
(b) Keay, B. A.; Bontront, J .-L. J . Can. J . Chem. 1991, 69, 1326. (c)
Cristofoli, W. A.; Keay, B. A. Tetrahedron Lett. 1991, 32, 5877. (d)
Cristofoli, W. A.; Keay, B. A. Synlett 1994, 625. (e) Maddaford, S. P.;
Keay, B. A. J . Org. Chem. 1994, 59, 6501. (f) We have recently used
this method to prepare a 2,3,4-trisubstituted furan intermediate in
the synthesis of (+)-xestoquinone, see: Maddaford, S. P.; Andersen,
N. G.; Cristofoli, W. A.; Keay, B. A. J . Am. Chem. Soc. 1996, 118, 10766.
(g) Scott has used our methodology for the preparation of 2,3,4-
trisubstituted furans, see: Danso-Danquah, R. E.; Scott, A. I. Tetra-
hedron 1993, 49, 8195.
(8) Yu, S.; Keay, B. A. J . Chem. Soc., Perkin Trans. 1 1991, 2600.
(9) Bures, E. J .; Keay, B. A. Tetrahedron Lett. 1987, 28, 5965.
(10) Beese, G.; Keay, B. A. Synlett 1991, 33.
(6) Friedrichsen, W. Compr. Heterocycl. Chem. II 1996, 2, 351 and
references therein.
(11) Spinazze, P. G.; Keay, B. A. Tetrahedron Lett. 1989, 30, 1765.
S0022-3263(97)01097-9 CCC: $14.00 © 1997 American Chemical Society

8742 J . Org. Chem., Vol. 62, No. 25, 1997
Bures et al.
Sch em e 1
Sch em e 2
12 (R ) H, E ) Si(t-Bu)Me2), which could not be easily
separated by chromatography or distillation.17 While
investigating alternative methods for preparing 3, we
discovered that 3 could be prepared in good yield via a
[1,4] O f C silyl migration.
Goldsmith et al. have reported18 that 3-((tert-butyldi-
methylsilyl)oxy)methyl)furan (1) can be regiospecifically
lithiated at C-2 (1 equiv of n-BuLi, ether, rt, 6 h) to
provide 2-(trimethylsilyl)-3-(((tert-butyldimethylsilyl)oxy)-
methyl)furan (13) in good yield (90%) after quenching the
anion with trimethylsilyl chloride (Scheme 2). In an
unrelated project, we found that treating the C-2 anion
of furan 1 with 1-bromo-3-chloropropane afforded the
alkylated product 14 in poor yield (28%). However,
changing the lithiation conditions (n-BuLi, THF, 0 °C, 6
h) and then adding HMPA simultaneously with the
electrophile resulted in the isolation of starting material
1 (36%) and a slower running spot by TLC, which was
identified as 2-(tert-butyldimethylsilyl)-3-(hydroxymeth-
yl)furan (3) (55%).19 Furan 3 was unexpected from this
reaction and appeared to be formed by a [1,4] O f C silyl
rearrangement.20 Since a search of the literature (in
1987-88) indicated there were only 5 examples of [1,4]
C f O silyl rearrangements,21 and none involved an
initially formed carbanion at an sp2 center, we investi-
gated this reaction in more detail.22
position. Two criteria were necessary for the C-4 lithia-
tion to occur. First, an ortho-lithiation director must be
present at the C-3 position of the furan ring.12 In
addition, this functional group must be easy to modify
for use in the synthesis of natural products. Second, the
C-2 position should have a bulky substituent, which is
easily attached and removed from the furan ring. The
size of this substituent was thought to be essential for a
successful C-4 lithiation, since the steric bulk at C-2
would presumably impede lithium coordination with the
furan ring oxygen atom, thereby reducing the likelihood
of a C-5 lithiation with n-BuLi. The system chosen
initially to meet the above criteria was 2-(tert-butyldi-
methylsilyl)-3-(hydroxymethyl)furan (3) (Scheme 1). The
hydroxymethyl group would act as the ortho-lithiation
director13 while the bulky silane at C-2 would provide
the necessary steric hindrance during the lithiation.
Later, we found that the carboxylic acid group was also
a good ortho-lithiation director in these systems.8 The
silyl group was also chosen since the C-Si bond in
arylsilanes and vinylsilanes can be cleaved by protode-
silylation or treatment with tetra-n-butylammonium
fluoride.14 Thus, removal of the silyl moiety on the furan
ring should be a facile process.
(17) Compounds 3 and 12 (R ) H, E ) Si(t-Bu)Me2) required four
silica gel columns (EtOAc:petroleum ether, 1:9) to afford compound 3
in 56% isolated yield.
(18) Goldsmith, D.; Liotta, D.; Saindane, M.; Waykole, L.; Bowen,
P. Tetrahedron Lett. 1983, 24, 5835.
(19) The spectral data of furan 3, isolated from the reaction involving
furan 1 with HMPA, were identical to the major product isolated from
reaction of the dianion of compound 10 with tert-butyldimethylsilyl
chloride.
[1,4] O f C of Silyl Migr a tion s of Silyl Eth er s
(20) Oxygen-to-carbon silyl migrations are sometimes referred to as
retro[1,n]Brook rearrangements, see: (a) Brook, A. G. J . Am. Chem.
Soc. 1958, 80, 1886. (b) Brook, A. G. ; Warner, C. M.; McGriskin, M.
E. J . Am. Chem. Soc. 1959, 81, 981. (c) Brook, A. G. Pure Appl. Chem.
1966, 13, 215. (d) Brook, A. G.; Pascoe, J . D. J . Am. Chem. Soc. 1971,
93, 6224. (e) Brook, A. G. Acc. Chem. Res. 1974, 7, 77. (f) Brook, A. G.;
Bassendale, A. R. Rearrangements in Ground and Excited States; De
Mayo, P., Ed.; Academic Press: New York, 1980; Vol. 2, pp 149-227.
(g) Brook, A. G.; Chrusciel, J . J . Organometallics 1984, 3, 1317 and
references therein. For recent references to many [1,n] X f Y silyl
migrations, see: He, H.-M.; Fanwick, P. E.; Wood, K.; Cushman, M.
J . Org. Chem. 1995, 60, 5905.
(21) (a) Speier, J . L. J . Am. Chem. Soc. 1952, 74, 1003. (b) Ben,keser,
R. A.; Cunico, R. F. J . Org. Chem. 1967, 32, 395. (c) Evans, D. A.;
Takacs, J . M. J . Am. Chem. Soc. 1979, 101, 371. (d) Hunig, J .; Oller,
M. Chem. Ber. 1980, 113, 3803. (e) Mora, J .; Costa, A. Tetrahedron
Lett. 1984, 25, 3493. (f) Rucker, C. Tetrahedron Lett. 1984, 25, 4349.
For an example in which the proposed mechanism involves a [1,4] O
f C silyl migration, see: Karlsson, J . O.; Nguyen, N. V.; Foland, L.
D.; Moore, H. W. J . Am. Chem. Soc. 1985, 107, 3392.
(22) Since 1988, more [1,4] O f C silyl rearrangements have been
reported, see: (a) Braun, M. Agnew. Chem., Int. Ed. Engl. 1989, 28,
896. (b) Satoh, T.; Oohara, T; Ueda, Y.; Yamakawa, K. J . Org. Chem.
1989, 54, 3130. (c) Kim, K. D.; Magriotis, P. A. Tetrahedron Lett. 1990,
43, 6137. (d) Wang, K. K.; Liu, C.; Gu, Y. G.; Burnett, F. N. J . Org.
Chem. 1991, 56, 1914. (e) Lautens, M.; Delanghe, P. H. M.; Goh, J . B.;
Zhang, C. H. J . Org. Chem. 1992, 57, 3270. (f) Hoffmann, R. W.
Bewersdorf, M. Tetrahedron Lett. 1989, 54, 3130. (g) Hoffmann, R.;
Bruckner, R. Chem. Ber. 1992, 125, 1471 and 2731. (h) Lautens, M.;
Delanghe, P. H. M.; Goh, J . B.; Zhang, C. H. J . Org. Chem. 1995, 60,
4213. (i) Ru¨cker, C. Chem. Rev. 1995, 95, 1009.
Tanis et al.15 and Katsumura et al.16 have reported that
treatment of 3-(hydroxymethyl)furan (10) with 2.2 equiv
of n-BuLi in TMEDA (0 °C) or THF (-78 °C, 2 h, 0 °C, 1
h) affords a 4:1 and 9:1 ratio of the 2,3-disubstituted- and
2,4-disubstituted-furans (11 and 12), respectively, after
quenching the dianions with an electrophile (Scheme 2).
In our hands, treatment of the dianion of 10, formed
using Katsumura’s conditions, with tert-butyldimethyl-
silyl chloride afforded a 9:1 mixture of compounds 3 and
(12) For excellent reviews on directed ortho-lithiation, see: (a) ref
1. (b) Snieckus, V. Chem. Rev. 1990, 90, 879.
(13) For use of a benzylic alcohol moiety as an ortho-lithiation
director, see: (a) Uemura, M.; Tokuyana, S.; Sakan, T. Chem. Lett.
1975, 1195. (b) Meyer, N.; Seebach, D. Chem. Ber. 1980, 118, 1304.
(b) Panetta, C. A.; Dixit, A. S. Synthesis 1981, 59. (d) Uemura, M.;
Take, K.; Tobe, K.; Minani, T.; Hayashi, Y. Tetrahedron 1985, 41, 5771.
(e) Katsuura, K.; Snieckus, V. Can. J . Chem. 1987, 65, 124. For a
synthetic sequence which results in products from the R-lithiation of
the benzylic methylene group, see: Katritzky, A. R.; Fan, W.-Q.;
Kutagawa, K. Synthesis 1987, 415.
(14) (a) Colvin, E. Silicon in Organic Synthesis; Butterworth and
Co. Ltd.: London, 1981. (b) Colvin, E. Silicon Reagents in Organic
Synthesis; Academic Press: London, 1988.
(15) Tanis, S. P.; Head, D. B. Tetrahedron Lett. 1984, 25, 4451.
(16) Katsumura, S.; Hori, K.; Fujiwara, S.; Isoe, S. Tetrahedron Lett.
1985, 26, 4625.

Regioselective Preparation of Substituted Furans
J . Org. Chem., Vol. 62, No. 25, 1997 8743
Ta ble 1. Resu lts of th e [1,4] O f C Silyl Migr a tion of Silyl Eth er s
entry
starting material
product(s) (% yield)a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1
15
16
17
18
19
20
21
22
23
24
25
26
27
X ) O; R1 ) R2 ) Me; R3 ) t-Bu
X ) S; R1 ) R2 ) Me; R3 ) t-Bu
X ) O; R1 ) R2 ) R3 ) i-Pr
X ) S; R1 ) R2 ) R3 ) i-Pr
X ) O; R1 ) R2 ) Ph; R3 ) t-Bu
X ) S; R1 ) R2 ) Ph; R3 ) t-Bu
X ) O; R1 ) R2 ) R3 ) n-Bu
X ) O; R1 ) R2 ) R3 ) Et
3 (87)
28 (87)
29 (86)
30 (89)
31 (87)
32 (63)
33 (85)
34 (79)
35 (75)
X ) S; R1 ) R2 ) Ph; R3 ) Me
X ) S; R1 ) R2 ) Me; R3 ) i-Pr
X ) O; R1 ) R2 ) Me; R3 ) i-Pr
X ) S; R1 ) R2 ) Me; R3 ) Ph
X ) O; R1 ) R2 ) R3 ) Me
36 (29)a + 23 (11)b + 42 (40)b + 39 (15)a
37 (24)a + 24 (15)b + 10 (43)b + 40 (12)a
38 (38)a + 25 (15)b + 42 (25)b + 41 (10)b
X ) S; R1 ) R2 ) R3 ) Me
a
b
Isolated yields. GC yields.
Sch em e 3
The silyl rearrangement was optimized by treating
compound 1 with 1.1 equiv of n-BuLi in THF containing
HMPA (1.1 equiv) at -20 °C for 1 h, followed by stirring
the reaction mixture at room temperature overnight.
Standard workup afforded furan 3 in 87% yield. The
reaction is quite general and is not limited to either the
tert-butyldimethylsilyl group or the furan ring (Table 1).
A variety of silyl derivatives (15-22) undergo the silyl
rearrangement in good to excellent yield to provide
compounds 28-35, respectively. In addition, the silyl
rearrangements occur when a thiophene ring is present
(Table 1, entries 2, 4, 6, and 9). Interestingly, the yield
of the expected 2,3-disubstituted heterocycles decreased
as the steric bulk of the substituents on the silane de-
creased (Table 1, entries 10-12). Thus the isopropyldi-
methylsilyl and dimethylphenylsilyl derivatives (23-25)
afforded the expected 2,3-disubstituted heterocycles 36-
38, respectively, in yields ranging from 24 to 38%. In
addition, 2,5-disilylated-3-(hydroxymethyl)-heterocycles
39-41 were formed (10-15%), plus some unreacted
starting material (23-25, 11-15%) and desilylated start-
ing material 10 and 42 (25-43%). The 3-(((trimethylsi-
lyl)oxy)methyl)furan (26) and -thiophene (27) provided
only desilylated starting material under the rearrange-
ment conditions. This was not unexpected since it is well-
known that trimethylsilyl ethers are readily cleaved by
alkyllithiums.23
The appearance of the 2,5-disilylated-3-(hydroxy-
methyl)-heterocycles (39-41), when the silyl substituents
were either isopropyldimethyl or dimethylphenyl,
prompted us to investigate the mechanism of this reac-
tion in more detail. A crossover experiment confirmed
that the silyl rearrangement was intramolecular in
nature (Scheme 3). Thus an equimolar mixture of
3-(((tert-butyldimethylsilyl)oxy)methyl)furan (1) and
3-(((triisopropylsilyl)oxy)methyl)thiophene (17), when
treated under the rearrangement conditions, provided an
equimolar mixture of furan 3 and thiophene 30.24 Analo-
gous intramolecular rearrangements have been reported
for [1,4] O f C,21,22 [1,4] O f O,25a [1,5] O f O,25b and
[1,4] C f C25c silyl rearrangements.
from the oxygen atom to the C-2 position of the furan or
thiophene ring. This was expected on the basis of the
report of Goldsmith et al.18 that the lithiation of furan 1
in ether was C-2 regiospecific.26 Since the above silyl
rearrangements were observed in THF, we found that
the poor yield of compounds 36-38 and the formation of
the 2,5-disilylated products 39-41 could be explained by
studying the lithiation of furan 1 in THF in the absence
of HMPA.
Treatment of furan 1 with n-BuLi in THF at -20 °C
for 1 h followed by a quench with trimethylsilyl chloride
1
provided a 2:1 ratio (by H NMR)27 of furans 13 and 43
(Scheme 4). Thus, a mixture of C-2 and C-5 monoanions
are being formed initially in THF. When furan 1 was
treated under the same conditions and HMPA added
after 1 h instead of an electrophile, only furan 3 was
obtained in 87% yield. The size of the groups on the
silicon atom in combination with the initial mixture of
C-2 and C-5 anions in THF therefore must be controlling
whether the silyl rearrangement occurs cleanly (to give
1 and 15-22) or provides a mixture of products (23-25).
(23) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, 2nd Ed.; J ohn Wiley & Sons Inc.: New York, 1991.
(24) Compounds 1 and 17 were chosen for the crossover study since
they were found to rearrange at approximately the same reaction rate.
(25) (a) van Boeckel, C. A. A.; van Aelst, S. F.; Beetz, T. Recl. Trav.
Chim. Pays-Bas. 1983, 102, 415. (b) Torisawa, Y.; Shibasaki, M.;
Ikegami, S. Chem. Pharm. Bull. 1983, 31, 2607. (c) Daney, M.;
Lapouyade, R.; Bouas-Laurent, H. J . Org. Chem. 1983, 48, 5055.
(26) In our hands the lithiation of furan 1 with n-BuLi in ether (rt,
6 h) provided a 9:1 mixture of 11:12 (Scheme 2: R ) tert-butyldim-
ethylsilyl; E ) Li). The 9:1 ratio was determined by quenching the
mixture of anions with MeOD and integrating the furan signals in
the 1H NMR spectrum.
The above experiment indicated that the products from
the rearrangements of compounds 1 and 15-23 were
formed by an intramolecular transfer of the silyl moiety
(27) Compounds 13 and 43 could not be separated on a silica gel
column (petroleum ether:EtOAc (9:1)) or by fractional distillation.

8744 J . Org. Chem., Vol. 62, No. 25, 1997
Bures et al.
Ta ble 2. Resu lts of th e [1,4] O f C Silyl Migr a tion of
Sch em e 4
Silyl Ester s
starting
entry material
product(s)
(% yield)a
1
2
3
4
5
6
7
8
2
54
55
56
57
58
59
60
X ) O; R1 ) R2 ) Me; R3 ) t-Bu
X ) S; R1 ) R2 ) Me; R3 ) t-Bu
X ) O; R1 ) R2 ) R3 ) i-Pr
X ) S; R1 ) R2 ) R3 ) i-Pr
X ) O; R1 ) R2 ) Ph; R3 ) t-Bu
X ) S; R1 ) R2 ) Ph; R3 ) t-Bu
X ) O; R1 ) R2 ) R3 ) n-Bu
X ) O; R1 ) R2 ) R3 ) Et
4 (72)b
61 (47)c
62 (56)b
63 (57)c
64 (43)b
65 (64)c
66 (51)b
67 (57)b
Sch em e 5
Isolated yields. 3-Furoic acid was also isolated. c 3-Thio-
a
b
phene carboxylic acid was also isolated.
lithiate the starting material 44) then lithiates compound
49 at the C-2 position to form 50, which rearranges to
form the 2,5-disilylated heterocycles 51. Support for the
proposed mechanism came from a crossover study in
which an equimolar mixture of furan 24 and thiophene
25, when treated with n-BuLi, provided a complex
mixture of 10 products. GC/MS analysis of the mixture
revealed the presence of compounds 10, 24, 25, 37, 38,
40, 41, 42, 2-(dimethylisopropylsilyl)-5-(dimethylphenyl-
silyl)-3-(hydroxymethyl)furan (52, M•+ ) 332) and 2-(di-
methylphenylsilyl)-5-(dimethylisopropylsilyl)-3-(hy-
droxymethyl)thiophene (53, M•+ ) 348). The presence
of compounds 52 and 53 indicates that an intermolecular
transfer of a silyl group is occurring.
In summary, good to excellent yields of 2-silylated-3-
(hydroxymethyl)furans and -thiophenes are obtained via
[1,4] O f C silyl migrations when the groups on the
silicon atom are bulky.
[1,4] O f C Silyl Migr a tion of Silyl Ester s
Since Knight et al.29 have reported that 3-furoic acids
can be lithiated with 2.2 equiv of LDA at the C-2 position,
we became interested to determine if silyl esters would
lithiate at the C-2 position and undergo a [1,4] O f C
silyl migration. At the time this investigation was
started we could not find any examples of silyl esters
undergoing a silyl migration. Most of the work reported
in the literature dealing with silyl esters has involved
studying reduction reactions of silyl esters30 and using
silyl esters as protecting group for carboxylic acids.31
Silyl esters 2 and 54-60 (Table 2) were prepared by
treatment of either 3-furoic or 3-thiophenecarboxylic acid
with a silyl chloride (1.2 equiv) and imidazole (1.2 equiv)
in DMF at 60 °C for 48 h.32 Yields of the silyl esters
ranged from 60 to 89%. During workup, the contact time
with water was minimized by performing fast extractions
since most of the silyl esters hydrolyzed to the acid upon
prolonged exposure to water.
When the groups attached to the silicon atom are large
(entries 1-9, Table 1), the initially formed C-2 anion (46,
Scheme 5) undergoes an intramolecular silyl rearrange-
ment to form the 2,3-disubstituted heterocycle 47; how-
ever, in the presence of HMPA, the C-5 anion 45
undergoes a “equilibration-disproportionation”28 to form
the C-2 anion 46 faster than the C-5 anion can intermo-
lecularly attack another silyl moiety to form 49. This
process leads to the formation of 2,3-disubstituted furans
and thiophenes in excellent yield when the groups
attached to silicon are bulky.
When smaller groups are attached to the silicon atom
(entries 10-12, Table 1) a different reaction path is
taken. The initially formed C-2 anion 46 results in the
intramolecular silyl migration to form 47. The C-5 anion,
however, intermolecularly attacks some unreacted start-
ing material 44 to form 49 and desilylated starting
material 48. The remaining n-BuLi (which did not
(29) (a) Knight, D. W.; Nott, A. P. J . Chem. Soc., Perkin Trans. 1
1981, 1125. (b) Knight, D. W. Tetrahedron 1979, 20, 469.
(30) (a) Corriu, R. J . P.; Lanneau, G. F.; Perrot, M. Tetrahedron Lett.
1987, 28, 3841. (b) Larson, G. L.; Ortiz, M.; Rodriguez de Roca, M.
Synth. Commun. 1981, 11, 583.
(31) (a) Borgulya, J .; Bernauer, K. Synthesis 1980, 545. (b) Hart, T.
W.; Metcale, D. A.; Scheinmann, F. J . Chem. Soc., Chem. Commun.
1979, 156. (c) Corey, E. J .; Kim, C. U. J . Org. Chem. 1973, 38, 1233.
(32) For the preparation of silyl esters, see: (a) Corey, E. J .;
Venkateswarlu, A. J . Am. Chem.Soc. 1972, 94, 6190. (b) Tacke, R.;
Lange, H. Chem. Ber. 1983, 116, 3685.
(28) (a) Zeigler, F. E.; Fowler, K. W. J . Org. Chem. 1976, 41 1564.
(b) Leonard, N. J .; Bryant, J . D. J . Org. Chem. 1979, 44, 4612. (c)
Ronald, R. C.; Winkle, M. W. Tetrahedron 1983, 39, 2031. (d) Marsais,
F.; Quegwiner, G. Tetrahedron 1983, 39, 2009.

Regioselective Preparation of Substituted Furans
J . Org. Chem., Vol. 62, No. 25, 1997 8745
Silyl esters 2 and 54-60 underwent a [1,4] O f C silyl
migration in the presence of LDA to provide 2-silyl-3-
hetereocyclic carboxylic acids 4 and 61-67, respectively
(Table 2). Thus, the addition of silyl esters 2 (or 54-60)
to a mixture of LDA and HMPA (1.2 equiv each) in THF
at -78 °C resulted in a migration of the silyl group into
the C-2 position of the furan or thiophene ring after
approximately 15 min. An acidic workup followed by
purification through a silica gel column provided 2-silyl-
3-furoic acids (4, 62, 64, 66, and 67) and 3-thiophenecar-
boxylic acids (61, 63, and 65) in moderate to good yields
(Table 2). 3-Furoic acid and 3-thiophenecarboxylic acid
were also isolated from the reaction mixture. The
formation of the acids was not unexpected, since it has
been reported that nitrogen nucleophiles can attack the
silicon atom of silyl esters.33 Changing the type of base
(LHMDS), solvent (ether, DME), additive (TMEDA), and
temperature (-78 °C f rt) employed in the reaction did
not noticeably improve the yields.
The silyl migration was shown to be an intramolecular
process by treating an equimolar mixture of compounds
54 and 55 with an LDA/HMPA mixture in THF; acids
61 and 62 were the only compounds detected (by 1H
NMR) and isolated. This result is in accord with the
crossover study performed on the silyl ethers (Scheme
3, vide supra) and others reported in the literature.21,22,25
The above results indicate that 3-substituted silyl
esters of furans and thiophenes undergo a [1,4] O f C
silyl migration when treated with a mixture of LDA/
HMPA in THF at -78 °C to provide 2,3-disubstituted
heterocycles in moderate to good yield.
rings or the tert-butyldimethylsilyl group; both thiophenes
and other silyl groups underwent the migration (entries
1-15, Table 3). Groups present in the C-4 position of
the furan ring36 did not impede the migration and 3,4-
disubstituted furans were obtained in excellent yields
(entries 16-20, Table 3). A few of the reactions were
tried in THF as a solvent and found to take approxi-
mately 16 h for completion. Generally the yields are very
high for furans and thiophenes in both DMF and THF.
The migration was also not limited to primary alcohols
since the secondary alcohols 72 and 73 provided migrated
products 77 and 78 in 95% and 88% yields, respectively.
Some of the silyl ethers were found to be labile under
the conditions of NaH/DMF (entries 12, 14, and 15, Table
3). In these cases, 1 h in DMF provided only 3-(hydroxy-
methyl)furan (10). It was determined that a reductive
cleavage of the expected silyl ether had occurred. Thus,
compounds 22, 24, and 26 were prepared by silylation of
10 and found to be completely desilylated after 1 h in a
NaH (5 equiv)/DMF mixture. In an attempt to isolate
the corresponding silyl ethers from the migrations, the
reactions with compounds 35 and 37 were stopped after
5 min. Silyl ethers 22 and 24 were isolated in good
yields, respectively. Therefore the desilylation occurred
after the migration was complete. These results indicate
that NaH/DMF may be a suitable method for the desi-
lylation of diphenylmethylsilyl- and dimethylisopropyl-
silyl groups.37
Since the above migrations were completed within 5
min when performed in DMF, we reinvestigated the
reaction with compounds 3 and 28 and found that
migrated products 1 and 15 were obtained in 93% and
85% yields, respectively, after only 5 min. In fact, TLC
indicated that the reaction of compound 3 was complete
within a few seconds.38 Since the time of the reaction
varied greatly between THF and DMF, we performed the
migration with furan 3 in diethyl ether and DME (Table
4). The reaction was complete after 30 min in DME39
(entry 2, Table 4) and did not proceed in diethyl ether
after stirring at rt for 7 days (entry 4, Table 4).40 The
addition of excess DMF to the ether solution (after 7
days), however, resulted in a rapid silyl migration to form
furan 1 in 92% yield (entry 4, Table 4). Reducing the
amount of NaH to catalytic quantities in DMF slowed
down the reaction, but it was still complete after 15 min
(entry 5, Table 4).
[1,4] C f O Silyl Migr a tion s of
2-Silyla ted -3-su bstitu ted F u r a n s a n d Th iop h en es
With a variety of 2-silylated-3-(hydroxymethyl)furans
and -thiophenes in hand, we investigated whether a [1,4]
C f O silyl migration would be possible with these
systems. The idea of using a [1,4] O f C migration,
followed by a lithiation at C-4, and then reprotection of
the hydroxymethyl group using the same silyl group that
underwent the initial [1,4] O f C migration seemed very
attractive as a synthetic method (see Scheme 1). Al-
though some [1,4] C f O silyl migrations had been re-
ported by 1989, none had involved the use of furan and
thiophene ring systems.34 Since 1989, two other exam-
ples of [1,4] C f O silyl migrations have been reported.22h,35
Treatment of compound 3 with excess NaH (5 equiv)
in DMF (1 h, rt) provided an excellent yield of furan 1,
in which the silyl group had apparently migrated from
the carbon atom to the oxygen atom (entry 1, Table 3).
Various counterions were employed in the migration
of furan 3 (Table 4). Potassium ions41 gave similar
results as sodium ions (compare entries 3 and 6), and
when NaOH was used instead of NaH, the migration
occurred in DMF after 1 h, producing a mixture of silyl
ether 1 and desilylated material 10 (entry 7, Table 4).
The use of magnesium or lithium cations in THF resulted
in no migration; however, if excess DMF was added after
1
The H and 13C NMR spectra of the product matched the
spectra obtained from the silylation of 3-(hydroxymeth-
yl)furan (10). The migration was not limited to furan
(33) (a) Hudrlik, P. F.; Feasley, R. Tetrahedron Lett. 1972, 1781.
(b) Gornowicz, G. A.; West, R. J . Am. Chem. Soc. 1968, 90, 4478. (c)
Sommer, L. H.; Parker, G. A.; Frye, C. L. J . Am. Chem. Soc. 1964, 86,
3280. (d) Anderson, H. H.; Fischer, H. J . Org. Chem. 1954, 19, 1296.
(34) (a) Woodbury, R. P.; Rathke, M. W. J . Org. Chem. 1979, 43,
1947. (b) Isobe, M.; Kitamura, M.; Goto, T. Tetrahedron Lett. 1979,
20, 3465. (c) Matsuda, I.; Murata, S.; Izumi, Y. Bull. Chem. Soc. J pn.
1979, 52, 2389. (d) Matsuda, I.; Murata, S.; Ishii, Y. J . Chem. Soc.,
Perkin Trans. 1 1979, 26. (e) Isobe, M.; Kitamura, M.; Goto, T.
Tetrahedron Lett. 1980, 21, 4727. (f) Fleming, I.; Floyd, C. D. J . Chem.
Soc., Perkin Trans. 1 1981, 969. (g) Takeda, T.; Naito, S.; Ando, K.;
Fujiwara, T. Bull. Chem. Soc. J pn. 1983, 56, 967. (h) Brook, A. G.;
Chrusciel, J . J . Organometallics 1984, 3, 1317. (i) Isobe, M.; Ichikawa,
Y.; Funabashi, Y.; Mio, S.; Goto, T. Tetrahedron 1986, 42, 2863.
(35) Tietze, L. F.; Feissler, H.; Gewert, J . A.; J akobi, U. Synlett 1994,
2863.
(36) The preparation of compounds 69-73 is discussed in the
following paper in this issue; see ref 55.
(37) For an example of a reductive desilylation using NaH/HMPA,
see: Shekani, M.S.; Kahn, K. M.; Mahmood, K. Tetrahedron Lett. 1988,
29, 6161.
(38) Although the silyl migration is finished within a few seconds,
we did not go back and repeat all the examples in Table 3. All
migrations performed since our initial report with excess NaH in DMF
have been finished with a few seconds, so we have no reason to doubt
that all the examples in Table 3 would be complete within a few
seconds.
(39) The reaction time of 56 h in DME for this reaction in our initial
communication was an error and all reactions done to date in DME
have been complete within 0.5 h.
(40) Other [1,4] C f O migrations using sodium ions in diethyl ether
have been reported not to proceed. See 34f.

8746 J . Org. Chem., Vol. 62, No. 25, 1997
Ta ble 3. Resu lts of th e [1,4] C f O Silyl Migr a tion of Silyl Eth er s
Bures et al.
1 d at rt, migrated product 1 was isolated in excellent
yields (entries 8-10, Table 4). The results with (a)
various counterions (b) the rate enhancement observed
with sodium ions on changing the solvent from ether to
THF to DME to DMF, and (c) the addition of DMF to
either the ether solution containing sodium ions or the
THF solution containing Mg or Li ions can be rationalized
on the degree of ion pair dissociation. A similar argu-
ment has been proposed to explain the effect of counter-
ions on the rate of oxy-Cope reactions42 and other
reactions.43-45 The enhanced ion pair dissociation has
been related to the polarity of the solvent;46 the more
polar the solvent, the more ion pair dissociation occurs
and thus the reaction rate increases.47
An alternative explanation for the rate increase ob-
served in more polar solvents is that the solvent is
activating the silicon atom to attack by the oxygen atom
by forming a pentavalent intermediate with the solvent.
Rate enhancements in the nucleophilic displacement
reactions at silicon have been reported in the presence
(43) Reichardt, C. Solvent Effects in Organic Chemistry; Verlag
Chemie: New York, 1988; pp 232-239.
(44) (a) Kolodziejski, W. J . Mol. Struct. 1987, 159, 335. (b) Stevenson,
G. R.; Hajhim, R. T. J . Phys. Chem. 1986, 90, 3217. (c) Szwarc, M.
Ions and Ion Pairs in Organic Reactions; Wiley and Sons: New York,
1974; Vol. 2, p 386.
(45) For theoretical calculations on Li+, Na+, and K+ interactions
with solvents, see: Portmann, P.; Marvizumi, T.; Welti, M.; Bad-
ertscher, M.; Neszmelyi, A.; Simon, W.; Pretsch, E. J . Chem. Phys.
1987, 87, 493. For theoretical calculations on BF3 affinities for some
solvents, see: Rauk, A.; Hunt, I. R.; Keay, B. A. J . Org. Chem. 1994,
59, 6808. For theoretical calculations on proton affinities with solvents,
see: Yamabe, S.; Hiraa, K.; Wasada, H. J . Chem. Phys. 1992, 96,
10261.
(41) KH was not used in DMF since it is known to reduce DMF,
see: Brown, C. A. J . Org. Chem. 1974, 39, 3913.
(42) (a) Evans, D. A.; Golob, A. M. J . Am. Chem. Soc. 1975, 97, 5765.
(b) Magnera, T. F.; Caldwell, G.; Sunner, J .; Ikuta, S.; Kebarle, P. J .
Am. Chem. Soc. 1984, 106, 6140.
(46) (a) See ref 43, pp 17-25. (b) Maria, P.-C.; Gal, J .-F. J . Phys.
Chem. 1985, 89, 1296 and references therein.

Regioselective Preparation of Substituted Furans
J . Org. Chem., Vol. 62, No. 25, 1997 8747
Ta ble 4. Effect of Va r iou s Solven ts a n d Ba ses on th e
being transferred to the C-2 position of the furan ring.
To prove that the proton source was indeed the hydroxyl
hydrogen atom of unrearranged starting material, the
following experiment was conducted. Furan 80 was
prepared by treating alcohol 3 with excess D2O in DME.
The solution of D2O, furan 80, and DME was passed
through a plug of Na2SO4 directly into a solution of excess
NaH in DMF. Workup after 5 min gave the C-2 deuter-
ated silyl ether 81 exclusively (Scheme 6). Since deute-
rium was incorporated at C-2 of the furan ring, the rate
of the silyl migration must be faster than the rate of
proton abstraction from the hydroxyl group by NaH.
Similar results have been reported with other C f O silyl
migrations.20f,22h This deuteration study suggests why
aldehyde 79 was not formed when the migration was
done in DMF. The reaction of any incipient species at
C-2 of the furan ring is expected to be protonated by the
unreacted alcohol faster than undergoing a reaction with
either MeI or DMF.
[1,4] C f O Silyl Migr a tion of F u r a n 3
product
entry
base
solvent
time
(% yield)
1
2
3
4
5
6
7
8
5 equiv of NaH
5 equiv of NaH
5 equiv of NaH
5 equiv of NaH
1 mol % of NaH
5 equiv of KH
DMF
DME
THF
Et2O
DMF
THF
seconds
0.5 h
1 (88)
1 (96)
1 (80)
16 h
7 d [DMF]a S.M. [1 (88)]b
15 min
2 d
1 (92)
1 (61)
1 (64) + 10 (23)
5 equiv of NaOH DMF
1 h
1 equiv of
vinyl-MgBr
1 equiv of MeLi
THF
1 d [DMF]a S.M. [1 (85)]b
9
10
THF
1 d [DMF]a S.M. [1 (89)]b
1 d [DMF]a S.M. [1 (88)]b
1 equiv of n-BuLi THF
a
After stirring at rt for the time shown excess DMF was added.
Product and yield after excess DMF was added to the solution.
b
S.M. ) starting material.
Sch em e 6
A cyclic intermediate having a pentavalent silicon atom
may also be involved in the reaction, and its formation
may be enhanced by the presence of good electron pair
donor solvents. Intermediates involving pentavalent
silicon atoms have been postulated in other intramolecu-
lar [1,n] silyl migrations20f,34f,49 and have been shown to
be viable intermediates through ab initio level calcula-
tions.50 Such an intermediate could also account for the
nonreactivity with electrophiles (DMF and MeI) other
than protons in this reaction.51
of DMF, DMSO, or HMPA.48 An activation of the silicon
atom by the formation of a pentavalent intermediate with
the solvent has been proposed. A similar effect may be
occurring with the more polar solvents DMF and DME
versus THF and diethyl ether.
We were initially surprised to find that furan 79 was
not formed when the [1,4] C f O silyl migration was
performed in anhydrous DMF (Scheme 6), since it is well-
known that aryl carbanions can be quenched with dry
DMF to provide aldehydes.1 We also tried to trap any
incipient C-2 carbanion with iodomethane. Thus, a
solution of furan 3 in THF containing 2 equiv of MeI was
treated with excess NaH and stirred at rt. After 8 h furan
1 was isolated in 87% yield; 1H NMR indicated that
neither 2-methyl-3-(((tert-butyldimethylsilyl)oxy)meth-
yl)furan nor 2-(tert-butyldimethylsilyl)-3-(methoxymeth-
yl)furan had formed.
To understand further what was happening during the
migration, the mechanism of this reaction was investi-
gated. A crossover study with an equimolar mixture of
furan 3 and thiophene 30 (NaH, DMF) only provided
furan 1 and thiophene 17, indicating the migration is an
intramolecular process. Since aldehyde 79 was not
formed during the migrations performed in DMF, the
question arose as to the source of the proton which is
Con clu sion s
We have shown that [1,4] O f C and [1,4] C f O silyl
migrations are facile reactions with 3-substituted furans
and thiophenes and proceed in good to excellent yields
with silanes containing bulky substituents. Both silyl
ethers and silyl esters undergo these rearrangements.
The next paper describes the C-4 and C-5 regiospecific
lithiations of various 2-silylated-3-substituted furans and
illustrates how 3,4-disubstiuted-, 2,4-disubstituted, and
2,3,4-trisubstituted furan rings can be prepared.
Exp er im en ta l Section
1H and 13C NMR spectra were run at 300 or 200 MHz and
75 or 50 MHz, respectively, and in CDCl3 as a solvent unless
otherwise noted. Elemental analyses were performed by either
Guelph Chemical Laboratories, Guelph, Ontario, Canada, or
by Ms. Dorothy Fox at the University of Calgary. tert-
Butyldimethylsilyl chloride was supplied from the Lithium
Corporation of America (now FMC), Gastonia, NC. All sol-
vents were dried and distilled prior to use. The oil from the
NaH was removed by three successive washes with anhydrous
diethyl ether. The remaining ether was removed under a high
vacuum (4 h, rt). Flash column chromatography was per-
formed using E. Merck silica gel (230-400 mesh A.S.T.M.) by
the method developed by Still et al.52
(47) Although refs 43-46 indicate that solvent polarity can be
related to sodium ion solvation, Abraham and Liszi have reported the
ionic solvation free energies for Na and K ions in DMF, THF, and DME.
Although the free energy change for Na and K ions was more negative
for DMF (-97.3 kcal/mol for Na ions and -79.5 kcal/mol for K ions),
when compared to the values for THF and DME (see below), the free
energy change for sodium ions in THF (-94.3 kcal/mol) was more
negative than that with DME (-92.5 kcal/mol), while with K ions, the
values were about the same (-73.3 kcal/mol in THF vs -73.1 kcal/
mol in DME). These results indicate that the observed rate enhance-
ments may be due to a combination of factors, see: Abraham, M. H.;
Liszi, J . J . Chem. Soc., Faraday Trans. 1 1978, 1604.
(49) Corriu, R. J . P.; Young, J . C. Hypervalent Silicon Compounds.
In The Chemistry of Organic Silicon Compounds; Patai, S., Rappoport,
Z., Eds.; J ohn Wiley and Sons: New York, 1989; Part 2, Chapter 20,
pp 1241-1288.
(50) (a) Antoniotti, P.; Canepa, C.; Tonachini, G. J . Org. Chem. 1994,
59, 3952. (b) Antoniotti, P.; Tonachini, G. J . Org. Chem. 1993, 58,
3622.
(51) 29Si NMR indicated that a pentavalent silicon atom does not
form in d10-diethyl ether when furan 3 was treated with excess NaH
(i.e. an upfield shift of the 29Si signal was not observed). This does not
preclude, however, that a pentavalent intermediate was involved in
the migration when excess DMF was added (entry 4, Table 4). The
reaction in the presence of DMF was too fast to observe any intermedi-
ates by NMR.
(48) Bassindale, A. R.; Taylor, P. G. Reaction Mechanisms of
Nucleophilic Attack at Silicon. In The Chemistry of Organic Silicon
Compounds; Patai, S., Rappoport, Z., Eds.; J ohn Wiley and Sons: New
York, 1989; Part 1, Chapter 13, pp 839-892.

8748 J . Org. Chem., Vol. 62, No. 25, 1997
Bures et al.
3-(Hyd r oxym eth yl)fu r a n (10). To a stirred mixture of
lithium aluminum hydride (7 g, 184 mmol) and dry diethyl
ether (100 mL) at 0 °C was added 3-furoic acid (20 g, 178 mmol)
over a 30 min period. After being stirred for 6 h at rt, the
solution was cooled to 0 °C and treated dropwise with water
(7 mL), 15% aqueous NaOH (7 mL), and water (21 mL). The
solution was filtered through Celite and the solvent evaporated
in vacuo to leave an oil which was purified by distillation to
afford 10 (13.5 g, 77%) as an oil: bp 98-100 °C/20 Torr (lit.53
ter t-Bu tyld im eth ylsilyl 3-F u r oa te (2). Compound 2 was
prepared in 84% yield according to general procedure 3 using
3-furoic acid (1.2 g, 10.7 mmol): bp 40-45 °C/0.08 Torr; IR
1
(NaCl) 1713, 1163 cm-1; H NMR 0.35 (s, 6H), 0.97 (s, 9H),
6.72 (d, 1H, J ) 1.8 Hz), 7.41 (t, 1H, J ) 1.8 Hz), 7.97 (d, 1H,
J ) 1.8 Hz); 13C NMR -4.9, 17.6, 25.5, 110.0, 121.0, 143.6,
148.0, 162.9; MS m/e 226 (M+), 169 (M+ - t-Bu); HRMS calcd
for C7H9O3Si 169.0321, found 169.0325.
Gen er a l P r oced u r e 4: [1,4] O f C Silyl Migr a tion of
Silyl Ester s 2 a n d 54-60. To a mixture of LDA (1.2 equiv)
and HMPA (1.2 equiv) in THF (1 mL/mmol) at -78 °C was
added the silyl ester (1.0 equiv) in THF (1 mL/mmol). After
15 min at -78 °C, chloroform was added and the mixture
transferred to a separatory funnel. HCl (5 mL of 10%) was
added, the separatory funnel was shaken vigorously, and the
aqueous layer was removed as fast as possible to prevent
decomposition of the furan ring. The chloroform layer was
dried (Na2SO4), filtered, and removed to leave an oil. Chro-
matography on silica gel provided 2-silylated 3-carboxylic acid
of furan or thiophene.
1
bp 79-80 °C/17 Torr); IR (neat) 3390 cm-1; H NMR 2.80
(bs, 1H, exchanges with D2O), 4.45 (s, 2H), 6.37 (d, 1H, J )
1.1 Hz), 7.34 (m, 2H); 13C NMR 56.2, 109.7, 125.0, 139.8,
143.3; MS m/ e 98 (M+).
3-(Hyd r oxym eth yl)th iop h en e (42). To a stirred mixture
of lithium aluminum hydride (4.5 g, 117 mmol) and dry diethyl
ether (60 mL) at 0 °C was added 3-thiophenecarboxaldehyde
(12 g, 107 mmol) over a 30 min period. After being stirred 6
h at rt, the solution was cooled to 0 °C and treated dropwise
with water (4.5 mL), 15% aqueous NaOH (4.5 mL), and water
(13.5 mL). The solution was filtered through Celite and the
solvent evaporated in vacuo to leave an oil which was purified
by distillation to afford 42 (10.1 g, 83%) as an oil: bp 95-98
2-(ter t-Bu tyld im eth ylsilyl)-3-fu r oic Acid (4). Compound
4 was prepared in 72% yield according to general procedure 4
starting with compound 2 (0.31 g, 0.91 mmol). 3-Furoic acid
was also recovered from the mixture (12%). Column chrom-
atography: petroleum ether:EtOAc (20:1); mp 83-88 °C; IR
°C/20 Torr (lit.54 bp 86-88 °C/10 Torr); IR (neat) 3347 cm-1
;
1H NMR 4.51 (s, 2H), 4.70 (s, 1H, exchanges with D2O), 7.1-
7.3 (m, 3H); MS m/ e 114 (M+).
Gen er a l P r oced u r e 1: Silyla tion of a n Alcoh ol. To a
solution of imidazole (84 mmol) in DMF (20 mL) at 0 °C was
added appropriate silyl chloride (40 mmol). After 10 min the
alcohol (40 mmol) was added and the mixture stirred 12 h at
rt. Saturated sodium chloride (20 mL) and diethyl ether (20
mL) were added, and the ether layer was separated, dried
(Na2SO4), and removed in vacuo to leave an oil. The oil was
purified by distillation.
(KBr) 3250-2250 (br), 1685, 1293 cm-1 1H NMR 0.37 (s,
;
9H), 0.97 (s, 9H), 6.80 (d, 1H, J ) 1.8 Hz), 7.63 (d, 1H, J ) 1.8
Hz); 13C NMR -5.9, 18.0, 26.7, 111.0, 128.0, 146.5, 168.0,
170.3; MS m/e 169 (M+ - t-Bu). Anal. Calcd for C11H18O3Si:
C, 58.37; H, 8.02. Found: C, 58.61; H, 8.15.
Gen er a l P r oced u r e 5: [1,4] C f O Silyl Migr a tion of
Com p ou n d s 3, 28-35, 37, a n d 68-73. To a solution of an
alcohol (0.27 mmol) in DMF or THF (5.4 mL) at rt was added
NaH (1.36 mmol), and the mixture was stirred for the time
shown in Table 3. Ether (6 mL) was added followed by the
slow addition of a saturated brine solution (CAUTION: violent
evolution of H2). The ether layer was separated, washed with
saturated brine (6×’s, only when DMF was used), and dried
(Na2SO4). The ether was removed in vacuo to leave an oil
which was distilled.
3-(((ter t-Bu tyld im eth ylsilyl)oxy)m eth yl)fu r a n (1). Us-
ing general procedure 1, compound 1 was prepared in 95%
1
yield: bp 106-109 °C/20 Torr; IR (neat) 1063 cm-1; H NMR
-0.04 (s, 6H), 0.81 (s, 9H), 4.54 (s, 2H), 6.42 (d, 1H, J ) 1.1
Hz), 7.51-7.55 (m, 2H); 13C NMR
-2.9, 18.2, 25.8, 57.5,
109.7, 125.9, 139.4, 143.1; MS m/e 212 (M+). Anal. Calcd for
C11H20O2Si: C, 62.21; H, 9.49. Found: C, 62.34; H, 9.43.
Gen er a l P r oced u r e 2: [1,4] O f C Silyl Migr a tion . A
solution of freshly distilled silylated alcohol (3.3 mmol) and
HMPA (3.6 mmol, dried over CaH2, distilled and stored over
molecular sieves) in dry THF (10 mL) was cooled to -78 °C
under argon and treated with n-BuLi (1.43 mL of 2.5 M in
hexanes, 3.6 mmol). The solution was allowed to come to rt
over 6 h and stirred at rt overnight. Saturated ammonium
chloride was added and the solution extracted with diethyl
ether. The organic layer was washed three times with
saturated copper sulfate and dried (Na2SO4), and the solvent
was removed in vacuo to afford after distillation a 2-silylated-
3-hydroxymethyl heterocycle.
Compounds 3, 28-34, and 68 provided compounds 1, 15-
21, and 10, respectively. Compounds 35 and 37 provided
compounds 22 (or 10) and 24 (or 10), respectively. The spec-
tral data from 1, 10, and 15-22 matched those reported
above. The migration of compound 71 has been reported
previously.7b
4-(((t er t -Bu t yld im e t h ylsilyl)oxy)m e t h yl)-3-fu r a ld e -
h yd e (74). General migration procedure 5 was performed on
furan 6955 (0.1 g, 0.42 mmol) for 5 min to produce, after
distillation, furan 74 as a clear colorless liquid in 88% yield:
1
2-(ter t-Bu tyldim eth ylsilyl)-3-(h ydr oxym eth yl)fu r an (3).
Using general procedure 2, compound 3 was prepared in 87%
bp 60-63 °C/0.05 mmHg (air bath); IR (KBr) 1680 cm-1; H
NMR 0.08 (s, 6H), 0.90 (s, 9H), 4.83 (s, 2H), 7.41 (d, 1H, J )
1.4 Hz), 7.99 (d, 1H, J ) 1.4 Hz), 9.91 (s, 1H); 13C NMR -5.5,
18.3, 25.8, 57.8, 125.6, 126.4, 142.0, 152.7, 185.1; MS m/e 182
(M+ - t-Bu). Anal. Calcd for C12H20O3Si: C, 59.96; H, 8.39.
Found: C, 60.12; H, 8.36.
1
yield: bp 75-78 °C/0.02 Torr; IR (KBr) 3319, 1070 cm-1; H
NMR 0.01 (s, 6H), 0.89 (s, 9H), 1.5 (bs, 1H, exchanges with
D2O), 4.57 (s, 2H), 6.46 (d, 1H, J ) 1.8 Hz), 7.57 (d, 1H, J )
1.8 Hz); 13C NMR -5.7, 18.1, 25.7, 57.1, 110.5, 135.9, 146.7,
155.0; MS m/e 212 (M+). Anal. Calcd for C11H20O2Si: C, 62.21;
H, 9.49. Found: C, 62.27; H, 9.47.
4-(((ter t-Bu tyld im eth ylsilyl)oxy)m eth yl)-2-d eu ter iofu -
r a n (81). To a solution of furan 80 (0.13 g, 0.59 mmol) in DME
(0.2 mL) was added D2O (53 uL). After 15 min the solution
was passed through a plug of sodium sulfate directly into DMF
(5 mL) containing excess NaH. After 5 min the mixture was
worked up according to general procedure 5 to produce, after
distillation, furan 81 as a clear colorless liquid in 82% yield:
Gen er a l P r oced u r e 3: P r ep a r a tion of Silyl Ester s 2
a n d 54-60. A mixture of either 3-furoic acid or 3-thiophen-
ecarboxylic acid (1.0 equiv), silyl chloride (1.2 equiv), and
imidazole (2.5 equiv) in DMF (2 mL/g of acid) were heated at
60 °C for 48 h. The reaction mixture was cooled to rt, ether
(2 mL/g of acid) was added, and the organic phase was washed
with a saturated brine solution (3 × 4 mL). The organic phase
was dried (Na2SO4), filtere,d and removed in vacuo to provide
the silyl esters which were purified by distillation.
1
bp 61-63 °C/0.03 mmHg (air bath); IR (neat) 1082 cm-1; H
NMR 0.07 (s, 6H), 0.90 (s, 9H), 4.58 (s, 2H), 6.35 (d, 1H, J )
1.4 Hz), 7.35 (d, 1H, J ) 1.4 Hz); 13C NMR -5.3, 18.0, 25.9,
57.4, 109.6, 125.7, 139.3 (three lines equal intensity), 143.0;
MS m/e 156 (M+ - t-Bu).
(52) Still, W. C.; Kahn, M.; Mitra, A. J . Org. Chem. 1978, 43,
2923.
(53) Dictionary of Organic Compounds, 2nd Supplement, 5th ed.;
Chapman and Hall: London, 1984; pp 215.
(54) Aldrich Chemical Catalog. 1994-5, pp 1347.
(55) For the preparation of this compound, see the following paper
in this issue: Bures, E.; Nieman, J . A.; Yu, S.; Bontront, J .-L. J .; Hunt,
I. R.; Rauk, A.; Keay, B.A. J . Org. Chem. 1997, 62, 8750.

Regioselective Preparation of Substituted Furans
J . Org. Chem., Vol. 62, No. 25, 1997 8749
Su p p or tin g In for m a tion Ava ila ble: Complete spectral
data for 15-25, 28-40, 54-60, 61-67, 75, 77, and 78 (12
pages). This material is contained in libraries on microfiche,
immediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
Ack n ow led gm en t. We thank the Natural Sciences
and Engineering Research Council of Canada (NSERC),
the University of Calgary, and the University of Wind-
sor for financial support. We also thank NSERC for
scholarships (to C.R. and I.R.H.) and the Royal Society
and Alberta Foundation for Medical Research for a
scholarship (to I.R.H.). We also thank the referees for
helpful suggestions.
J O971097J