1366
SHORT PAPER
Synthesis of Cycloocta-3,5-dien-1-ol and Cycloocta-3,5-dien-1-one: SeO2/O2
Oxidation of Dienes
Elena S. Koltun, Steven R. Kass*
Department of Chemistry, University of Minnesota, Minneapolis, MN 55455, USA
Fax +1(612)6267541; E-mail: kass@chem.umn.edu
Received 2 February 2000; revised 20 April 2000
oxidation of cycloocta-1,3-diene. As it turns out, the prod-
Abstract: A useful synthesis of cycloocta-3,5-dien-1-ol was devel-
uct ratio changes with the amount of oxygen in the reac-
oped starting from cycloocta-1,3-diene. Oxidation of the diene with
tion mixture as shown in the Table. When the reaction is
carried out under nitrogen, the allylic alcohol 2 is the ma-
selenium oxide in refluxing acetic anhydride affords the homoallyl-
ic and allylic acetates in a 19:1 ratio and a 36% overall yield when
O2 is bubbled through the reaction mixture. Subsequent reduction jor product. When we reproduced the literature conditions
with LAH affords the homoallylic alcohol (95%), which can be
and carried out the oxidation under air, the LiAlH4 reduc-
readily oxidized via TPAP/N-MMO (73%) to the corresponding
tion yielded 1 and 2 in a 2.5: 1 ratio, respectively. To in-
non-conjugated dienone. This route represents the most efficient
crease the concentration of oxygen in the reaction
method for the preparation of cycloocta-3,5-dien-1-ol and cyclooc-
mixture, air was bubbled through the refluxing acetic an-
ta-3,5-dien-1-one reported to date, and may provide a general ap-
hydride solution and after the subsequent reduction, the
ratio of the products was 5.5:1, respectively.9 The propor-
tion of homoallylic alcohol 1 was significantly improved
when oxygen was bubbled through the reaction mixture.
Only two isomers, 1 and 2, were isolated and they were
obtained in a 19:1 ratio, respectively (Scheme 2). There-
fore, a SeO2/O2 combination allows homoallylic alcohol 1
to be prepared from an inexpensive starting material, cy-
cloocta-1,3-diene, in 33% overall yield and 95% purity.
proach to the synthesis of homoallylic alcohols and non-conjugated
enones.
Keywords: selenium oxide, oxygen, homoallylic alcohol, non-con-
jugated enone, oxidation
Synthesis of cycloocta-3,5-dien-1-ol (1) has been attempt-
ed several times via reduction of vinyl epoxides or seleni-
um oxide oxidation of alkenes or dienes.1-7 These
methods result in mixtures of allylic and homoallylic
products typically with the latter as the minor component.
Even though cycloocta-3,5-dien-1-ol (1) and cycloocta-
2,4-dien-1-ol (2) can be separated via column chromatog-
raphy on small scale, it is almost impossible to separate
large amounts by distillation. Therefore, we developed an
efficient procedure for the preparation of cycloocta-3,5-
dien-1-ol (1) in two steps from an inexpensive starting
material, cycloocta-1,3-diene.
Homoallylic alcohol 1 was oxidized with tetrapropylam-
monium perruthenate (TPAP)/4-methylmorpholine N-ox-
ide (N-MMO) without migration of the double bonds into
conjugation with the ketone (Scheme 2).10 This method,
therefore, could be an efficient way for the synthesis of
b,g-unsaturated alcohols and ketones. Further investiga-
tions into the generality of the SeO2/O2/Ac2O reaction
conditions for the synthesis of a variety of homoallylic al-
cohols and b,g-unsaturated ketones is currently underway.
HO
OH
Cycloocta-3,5-dien-1-ol (1)
For a 10 min period O2 was bubbled through a vigorously stirred
mixture (with a teflon coated stirring bar) of cycloocta-1,3-diene
(11.3 g, 0.1 mol), SeO2 (11.3 g, 0.1 mol) and Ac2O (75 mL). After
the solution was saturated with O2, it was heated to reflux for 12 h.
O2 was bubbled through the reaction mixture over this entire period
and the mixture changed color from red to deep black within
30 min; if any red color persists it is important to promptly raise the
temperature of the heating mantle or oil bath, otherwise a significant
amount of the allylic product forms. The mixture was allowed to
1
2
Hanold and Meier reported the synthesis of 1 via oxida-
tion of cycloocta-1,3-diene with selenium oxide in reflux-
ing acetic anhydride followed by reduction of the cool to r.t. and filtered through a plug of silica gel. The residue was
rinsed with Et2O (150 mL) and the combined organic layers were
resulting acetates with lithium aluminum hydride
(LiAlH4).1 This reaction sequence gave a mixture of prod-
ucts, cycloocta-3,5-dien-1-ol (1), cycloocta-2,4-dien-1-ol
washed with H2O (2 × 50 mL), 2 N NaOH (50 mL) and brine
(100 mL). The resulting solution was dried (MgSO4) and concen-
trated under reduced pressure. Vacuum distillation (0.3 Torr) af-
(2), and cyclooct-3-en-1-one (6)in 7.5:1.5:1 ratio, respec-
forded acetates 3 and 4 in a 95:5 ratio, respectively (6 g, 36%); bp
tively (Scheme 1).8
54-58 °C (Lit.1 bp 63-65 °C/2 Torr). The resulting acetate mixture
(5 g, 0.03 mol) was dissolved in Et2O (50 mL) and a 1 M LiAlH4
(0.04 mol, 40 mL) solution in THF was slowly added. The mixture
was refluxed (36-40 °C) under N2 for 2 h and cooled to 0 °C. To the
resulting solution was added slowly sat. NH4Cl solution (20 mL)
In an attempt to prepare gram quantities of homoallylic al-
cohol 1, we investigated the effect of oxygen on the prod-
uct ratio in the selenium oxide and acetic anhydride
Synthesis 2000, No. 10, 1366–1368 ISSN 0039-7881 © Thieme Stuttgart · New York