5838 J . Org. Chem., Vol. 65, No. 18, 2000
Notes
Ta ble 2. Rea ction of Diols w ith HBr in Tolu en ea
due to the relatively lower reactivity of bromo alcohols
(compared to diols) under the reaction conditions. The
fundamental reason(s) for this lower reactivity is not
obvious,12 but if one considers that (long-chain) bromo
alcohols might behave like surfactants, then one may
speculate that the bromo alcohols are less reactive due
to the formation of aggregates such as reverse micelles
or water/oil microemulsions. Such aggregates might be
expected to shield the polar hydroxyl groups from re-
agents in the bulk solvent. We have no direct evidence
for aggregation, but it is consistent with observations that
azeotropic removal of water decreases selectivities9 (since
water is sometimes required for nonionic surfactants to
form aggregates13) and that the selectivities are some-
what concentration dependent (at higher concentrations,
slightly lower selectivities are obtained). Whatever the
real reason(s) may be, this method should be very useful
to chemists interested in preparing ω-bromo alcohols as
synthetic intermediates.
ratio
entry
n
time (h)
1:2:3b
yield of 2c
1
2
3
4
5
6
5
6
7
8
10
12
9
13
72
72
48
72
5:94:1
4:95:1
0:99:1
1:97:2
2:97:1
4:94:2
81
79
94
87
87
80
a
All reactions were carried out with 1.2 equiv of aqueous HBr,
no Dean-Stark trap. Additional HBr (0.3-0.5 equiv) was added
if significant amounts of diol were present after 36 h. b Determined
by GC analysis of acetylated crude reaction mixtures. c Isolated
yield of 2 after flash chromatography.
Since toluene (without a Dean-Stark trap) seemed to
be the best solvent for this reaction, we examined diols
with other chain lengths under these conditions (Table
2). Reactions did not always proceed to completion when
only 1.1 equiv of HBr was used, but excellent results were
obtained if the reactions were monitored by TLC and
additional HBr was added as required. In all cases, the
product distributions were excellent and good yields of
bromo alcohols were obtained. It should be noted that
the product distributions reported in Table 2 were
determined by GC analysis of crude reaction mixtures.
This is important since the product distributions could
be easily distorted by fractionation. For example, diol 1d
is not very soluble in cold toluene and easily precipitates
out; sampling of the supernatant would then give inac-
curate compositions. It should also be noted that yields
reported are isolated yields of chromatographed bromo
alcohols and that isolated yields of dibromides and diols
were consistent with GC analyses.
The large polarity difference among dibromides, bromo
alcohols, and diols makes chromatographic separations
trivial. However, cost considerations, particularly on
larger scales, may make it desirable to avoid chromatog-
raphy. Fortunately, since these reactions are quite clean,
reasonably pure bromo alcohols may be obtained by
simple Kugelrohr distillation of the crude reaction mix-
ture. For example, with diol 1d , the desired bromo alcohol
2d (contaminated with 1% 1d and 2% 3d ) was obtained
in 95% yield using this simple procedure.
It is not obvious why these reactions give product
distributions which deviate significantly from the statis-
tical 25:50:25 mixture of 1:2:3 that one might expect. In
the classical continuous extraction chemistry,2 selective
extraction of the bromo alcohol (leaving the more polar
diol to react) effectively protects it from further reaction.
Similarly, other monoprotection/functionalization schemes
for diols are accompanied by reasonable explanations for
the observed selectivities.11 In this case, it can be
reasoned that, for the high yields of 2 that are observed,
the bromo alcohols must be much less reactive than the
diols. And, in fact, treatment of bromo alcohol 2d with
1.1 equiv of HBr in toluene at reflux for 48 h resulted in
only 11% conversion to the dibromide. Thus, we can
superficially explain the nonstatistical mixtures as being
Exp er im en ta l Section
Gen er a l In for m a tion . Diols were obtained from Aldrich
Chemical Co.; hydrobromic acid (47-49%, BDH, AnalaR) and
reagent grade solvents were used without further purification.
Thin-layer chromatography was carried out using Merck 5715
silica gel F254 plates, while flash chromatography employed EM
Science 35-75 µm silica gel 60. Ratios of products (Table 2) were
determined by GC-MS analysis of acetylated (xs Ac2O, pyr, cat.
DMAP) crude reaction mixtures and are not corrected for
response factors; the order of elution using an HP5 column was
dibromide, bromoacetate, and diacetate. All bromo alcohols
prepared are known compounds and showed spectral data
consistent with literature values.3
Typ ica l P r oced u r e. To a mixture of diol 1d (30 g, 0.205 mol)
and toluene (600 mL) was added concentrated HBr [27 mL of a
48% (9 M) aqueous solution, 0.24 mol]. The heterogeneous
mixture was stirred and heated at reflux for 36 h. TLC analysis
(silica, ether as eluent, Rf ) (diol 1d ) 0.05, (bromo alcohol 2d )
0.55, and (dibromide 3d ) 0.95) indicated substantial amounts
of diol 1d still remained. Thus, a further quantity of HBr (10
mL, 0.09 mol) was added, and the mixture was heated at reflux
for a further 36 h, at which time TLC analysis showed no diol
remaining. The reaction mixture was allowed to cool to rt, and
the phases were separated. The organic layer was diluted with
ether and washed with 1 M NaOH, brine, and phosphate buffer
(3 M, pH 7). Drying (Na2SO4) and concentration of the organic
layer gave a yellow oil which was distilled (Kugelrohr, bath temp
110-120 °C, 0.2 Torr) to provide bromo alcohol 2d as a colorless
liquid (42.0 g). GC analysis of this material (after acetylation)
showed a purity of 97% (1% diol, 2% dibromide). Alternatively,
flash chromatography of the crude reaction mixture (500 g of
silica, hexanes-ether, 1:1) provided 37.4 g (87%) of 2d free of
diol or dibromide.
Essentially identical results were obtained when 1200 mL of
toluene (40 mL/g of diol) was used, but use of 10 mL/g of diol
resulted in formation of considerably more dibromide (6%).
Ack n ow led gm en t. We thank the Natural Sciences
and Engineering Research Council of Canada (NSERC)
for financial support and an Undergraduate Student
Research Award (to M.A.H.) and Dr. Fabio Souza for
enlightening discussions.
J O000291U
(12) It was suggested by a reviewer that differences in reactivity
might be due to lower aqueous solubility of bromo alcohols compared
to diols and that experiments with a PTC might provide some insights.
Addition of 10 wt % BnEt3NCl or Aliquat 336 to these reactions did
not affect product distributions, suggesting that aqueous solubility is
not a significant issue here.
(11) See, for example: (a) McDougal, P. G.; Rico, J . G.; Oh, Y.-I.;
Condon, B. D. J . Org. Chem. 1986, 51, 3388-3390. (b) Houille, O.;
Schmittberger, T.; Uguen, D. Tetrahedron Lett. 1996, 37, 625-628.
(13) Organized Assemblies in Chemical Analysis: Reversed Micelles;
Hinze, W. L., Ed.; J AI Press: Greenwich, CT, 1994.