A convenient synthesis of substituted 3-bromotetrahydrofurans from homoallylic alcohols

Article abstract of DOI:10.1016/j.tetlet.2004.10.013

Substituted 3-bromotetrahydrofurans were prepared from homoallylic alcohols via bromination and cyclization in methanol in the presence of potassium carbonate.

Full text of DOI:10.1016/j.tetlet.2004.10.013

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 8811–8813
A convenient synthesis of substituted 3-bromotetrahydrofurans from
homoallylic alcohols
Marina V. Chirskaya,a Andrei A. VasilÕev,b,* Natalia L. Sergovskaya,a
Sergey V. Shorshneva and Sergey I. Sviridova
aChemBridge Corporation Ltd, Box 424, 119048 Moscow, Russian Federation
bN.D.Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, Leninsky pr.47, 119991 Moscow, Russian Federation
Received 2 August 2004; revised 22 September 2004; accepted 1 October 2004
Abstract—Substituted 3-bromotetrahydrofurans were prepared from homoallylic alcohols via bromination and cyclization in
methanol in the presence of potassium carbonate.
Ó 2004 Elsevier Ltd. All rights reserved.
The tetrahydrofuran fragment is not uncommon among
biologically active and natural compounds. Oxidative
cyclization of homoallylic alcohols is a traditional strat-
egy to access tetrahydrofurans with functional groups in
position 3.1 However, such a transformation under elec-
trophilic conditions, in particular in the case of allyl-
carbinols 1, is kinetically unfavorable and therefore has
to be performed indirectly, for example, via preliminary
oxidative functionalization of the double bond. Some
progress in the preparation of 3-hydroxytetrahydro-
furans has been reported recently. For example, the
reaction of homoallylic alcohols 1 with the NaIO4/
NaHSO3 system allows a direct synthesis of 3-hydr-
oxytetrahydrofurans in moderate yields.2 Epoxidation
of 1 followed by treatment with MgBr2 or MgI2 prov-
ides 3-hydroxytetrahydrofurans in satisfactory yields and
sometimes very good diastereoselectivity.3 Iodoetherifi-
cation leading to 3-iodotetrahydrofurans has been tho-
roughly investigated and is restricted to chemotypes of
crotyl (and homologous higher alk-2-enyl) carbinols
with 1,2-disubstituted double bonds, the reaction being
driven via a secondary carbocation.4 Analogous iodo-
etherification of allylcarbinols such as 1 with bis(sym-
collidine)iodine(I) perchlorate leads to four-membered
2-(iodomethyl)oxetanes.5
The equivalent 3-bromotetrahydrofurans 2 represent a
challenge for their straightforward preparation from 1
via bromination followed by cyclization of dibromoal-
cohols 3 by the action of appropriate bases.6–9 However,
KOH in ether, reported previously for this purpose,6
showed very moderate chemoselectivity in our experi-
ments, and the cyclization was accompanied by forma-
tion of a significant fraction of vinylic bromides
4, especially in the cases of tertiary alcohols (R1 and
R2 5 H). Other procedures involving pyridine,7 quino-
line8 and n-BuLi9 are not convenient for large scale
preparations.
In the present work, we have tested a broad variety of
base systems and discovered that bases of moderate
strength in protic solvents are expedient to achieve a
favorable ratio between bromotetrahydrofurans 2 and
vinylic bromides 4 (Table 1). Alkali carbonates in
Br
Br
Br
Br
3
Br2
base
R1
R1
R2
OH
R1
+
R2
R1
OH
R2
OH
O
R2
2
1
4
Keywords: Tetrahydrofurans; Bromination; Homoallylic alcohols; Cyclization.
*
Corresponding author. Fax: +7 095 1355328; e-mail: vasiliev@ioc.ac.ru
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.10.013

8812
M.V.Chirskaya et al./ Tetrahedron Letters 45 (2004) 8811–8813
Table 1. Cyclization of 1-(2,3-dibromopropyl)cyclohexanol [3a, R1 + R2 = (CH2)5] into 3-bromo-1-oxaspiro[4.5]decane 2a using various bases
Entry
Reagent systema
Temperature, °C
Time, h
6
Conversion,b
%
2/4 ratiob
1
2
Li2CO3/MeOH
Na2CO3/MeOH
20–60
2024
6018
20
ꢁ0
—
—
6
99
92/8
100
3
4
5
6
7
8
9
K2CO3/MeOH
Rb2CO3/MeOH
Cs2CO3/MeOH
K3PO4/MeOH
MeONa/MeOH
EtONa/EtOH
K2CO3/EtOH
6–24
90/10
890/11 206
850/15 206
10
10
20
20
6
100
100
90/10
90/10
82/18
0.5
0.5
20
206
100
88/12
47
10
807/163012
2
10DBU/THF
60
0
2012
54/4610
11
12
13
14
15
16
KOH/Et2O
8304/66
K2CO3/DMSO
K2CO3/acetone
K2CO3/MeCN
K2CO3/[bmim]BF4
t-BuOK/THF
210/79 506
5012
10
10
86
12/88
100
50
60/94 206
20
18
10/90
15/85
0.1
100
a 5equiv carbonates or 1.2–1.3equiv of other bases, 0.22M concentration of 3a.
b GC and/or NMR data.
methanol, except for Li2CO3, which failed to give any
product, provided >85% chemoselectivity (entries 1–5),
an equally good result was observed with K3PO4 (entry
6). The use of sodium methoxide (entry 7) was also suc-
cessful, but its application required careful attention to
stoichiometry to prevent further HBr elimination from
2. Ethanol as a solvent provided similar but somewhat
poorer yields (entries 8 and 9). All other base systems
in aprotic media tended to give mainly vinylic bromides
4 (entries 10–16). The best selectivity (94%) for bromo
olefin 4 was achieved in the case of K2CO3 in the ionic
liquid [bmim]BF4 (entry 15); similar results were
observed in entries 12–16. These reagent systems can
be recommended for the preparation of vinylic bromides
4, mainly as (E)-isomers. Another possible regioisomer
of vinylic bromide, possessing a C(Br)@CH2 fragment,
was not usually formed (sometimes it was detected in
negligible amounts).
To evaluate the scope and limitations of our two-step
protocol, we have tested several homoallylic alcohols
(Table 2).10 Clean bromination of starting alcohols 1
was found to be crucial: cyclization should be carried
out with the crude material, since dibromo alcohols 3
would decompose during attempted purification by vac-
uum distillation or column chromatography. Homoally-
lic alcohols 1a–e were brominated relatively cleanly in
CH2Cl2 at ꢀ30°C, and after the cyclization, the target
Table 2. Preparative syntheses of 3-bromotetrahydrofurans (for details, see Ref. 10)
Starting olefinic alcohol
Product
2/4 ratio
Yield of 2,a
%
O
OH
1a
1b
2a
2b
90/10
80
Br
O
OH
93/7
78
82
Br
O
OH
1c
2c
ꢁ100/0
Br
Et
Et
O
OH
Et
Et
1d
1e
5
2d
2eb
6
84/16
80/20
—
72
68
Br
O
OH
Ph
Ph
Br
O
OH
ꢁ90
Br
a Isolated yields after two steps.
b 9/10diastereomeric mixture.

M.V.Chirskaya et al./ Tetrahedron Letters 45 (2004) 8811–8813
8813
10. Representative procedure. A solution of bromine (1.15–
1.25mL, 22–24mmol) in CH2Cl2 (5mL) was added
dropwise to a stirred solution of 1-allylcyclohexanol 1a
(2.80g, 20mmol) in CH 2Cl2 (20mL) at ca. ꢀ30°C under
an argon atmosphere. The addition of bromine was
stopped when its color persist. The mixture was then
treated with Na2S2O3–NaHCO3 (aq), the organic layer
was separated, dried (MgSO4) and concentrated in vacuo.
The crude dibromoalcohol 3a was dissolved in MeOH
(50mL), freshly powdered K2CO3 (8.3g) was added, and
the mixture stirred for 4–24h at ambient temperature until
the consumption of dibromide 3a was complete (TLC, GC
or NMR control). Most of volatiles were distilled off, the
residue treated with water, and the organic components
extracted with ether. The extracts were dried (CaCl2) and
concentrated in vacuo. The residue was subjected to
column chromatography (gradient 0! 3% EtOAc in
hexane) to afford 3.5g (80%) of 3-bromo-1-oxaspiro-
[4.5]decane 2a. Further elution to 5% EtOAc in hexane
gave 0.175g (4%) of (E)-1-(3-bromoprop-2-en-1-yl)cyclo-
hexanol 4a. Compound 2a, bp 123–5°C (20Torr), aD20
bromotetrahydrofurans 2 were separated from alcohols
4 by column chromatography.11 The amount of vinyl
bromide side products 4 is to some extent substrate
dependent (Table 2).
The attempted synthesis of analogous 3-bromotetrahyd-
ropyran derivatives in the same manner was unsuccess-
ful. Homologous olefinic alcohol 5 afforded directly
2-bromomethyltetrahydrofuran derivative 6 during the
bromination step (with less than 10% of other compo-
nents), which was not unexpected.1,12
To conclude, we have developed a practical protocol for
the conversion of homoallylic alcohols into 3-bromo-
tetrahydrofurans that allows good access to various 3-
derivatized tetrahydrofurans. All the steps are easy to
perform, do not require expensive reagents and may tol-
erate certain sets of functional groups.
1
1.5092. H NMR (CDCl3, 400MHz): d 1.28–1.46 (m, 5H,
CH2), 1.50(m, 1H, H from CH 2), 1.58–1.78 (m, 4H, CH2),
2.07 (dd, 1H, J = 13.7, 5.9Hz, 1H, H from 4-CH2), 2.30
(dd, 1H, J = 13.7, 7.6Hz, H from 4-CH2), 3.92 (dd, 1H,
J = 9.8 and 5.9Hz, H from 2-CH2), 4.13 (dd, 1H, J = 9.8,
5.6Hz, H from 2-CH2), 4.32 (m, 1H, CHBr). 13C NMR
(CDCl3, 100MHz): d 23.5, 23.6, 25.3, 37.4 and 37.7 (CH2
from cyclohexane), 45.1 (CH2CHBr), 47.3 (CHBr), 74.0
(OCH2), 83.3 (OC). Compound 4a, bp 96°C (1Torr), aD20
Acknowledgements
A.A.V. acknowledges greatly financial support from the
Russian Science Support Foundation and ChemBridge
Corporation.
1
1.5237. H NMR (CDCl3, 400MHz): d 1.16–1.63 (several
References and notes
peaks, 11H, CH2 and OH), 2.15 (dd, 1H, J = 8.0, 1.1Hz,
allylic CH2), 6.05 (dt, 1H, J = 13.6, 1.1Hz, @CHBr), 6.24
(dt, 1H, J = 13.6, 8.0Hz, @CH). 13C NMR (CDCl3,
100MHz): d 21.9 (3- and 5-CH2 of cyclohexane), 25.5 (4-
CH2 of cyclohexane), 37.2 (2- and 6-CH2 of cyclohexane),
45.5 (allylic CH2), 70.9 (C-OH), 106.5 (@CHBr), 133.4
(@CH). Anal. Calcd for C9H15BrO (219.12) (%): C, 49.33;
H, 6.90; Br, 36.47. Found(%): C, 49.80; H, 7.04; Br, 36.20.
The structures of other products 2b–e were confirmed by
spectral data. CAS registry numbers: 2a 1195-85-3; 2d
1920-14-5.
1. Larock, R. C. Comprehensive Organic Transformations.A
Guide to Functional Group Preparations; VCH: NY, 1989.
2. Okimoto, Y.; Kikuchi, D.; Sakaguchi, S.; Ishii, Y.
Tetrahedron Lett. 2000, 41, 10223–10227.
3. Karikomi, M.; Watanabe, S.; Kimura, Y.; Uyehara, T.
Tetrahedron Lett. 2002, 43, 1495–1498.
4. (a) Lipshutz, B. H.; Barton, J. C. J.Am.Chem.Soc. 1992,
114, 1084–1086; (b) Mihelech, E. D.; Hite, G. A. J.Am.
Chem.Soc. 1992, 114, 7318–7319; (c) Bedford, S. B.; Bell,
K. E.; Fenton, G.; Hayes, C. J.; Knight, D. W.; Shaw, D.
Tetrahedron Lett. 1992, 33, 6511–6514; (d) Barks, J. M.;
Knight, D. W. Tetrahedron Lett. 1994, 35, 7259–7262.
5. Evans, R. D.; Magee, J. W.; Schauble, J. H. Synthesis
1988, 862–868.
6. (a) YurÕev, Yu. K.; Voronkov, M. G.; Gragerov, I. P.;
KondratÕeva, G. Ya. Zh.Obsch.Khim. 1948, 18, 1804–
1810; (b) Ou, K.-H. Ann.Chim. 1940, 13, 175–241.
7. Normant, H. Compt.Rend. 1948, 226, 1734–1736.
8. Huet, J. Bull.Soc.Chim.Fr. 1964, 2677–2680.
9. Prostakov, N. S.; Varlamov, A. V.; Kozm, K.; Aliev, A.
E.; Fomichev, A. A. Khim.Geterotsikl.Soedin. 1989, 5,
11. Branched 1-(1-methylprop-2-en-1-yl)cyclohexanol, 1-(1-
phenylprop-2-en-1-yl)cyclohexanol as well as homologous
1-allylcycloheptanol and in particular 1-allylcyclooctanol
were found to form large amounts of side products during
bromination, and, therefore, the corresponding final
materials 2 were heavily contaminated even after double
purification (chromatography + distillation). We have
found, however, that the material thus obtained can be
safely used in nucleophilic substitution reactions of
bromine, for the impurities remain unchanged. The yields
of products in these cases are at the 50% level.
12. An authentic sample of compound 6 was prepared
cleanly by treatment of 5 with N-bromosuccinimide in
CH2Cl2.
670–696, [Chem.Heterocycl.Compd.(Engl.Trans)l.
1989, 5, 581–585].
,