4464 J . Org. Chem., Vol. 66, No. 13, 2001
Amantini et al.
Sch em e 1
iodocarboxylic acid 4a present at the end of the reaction,
is correlated with the pH of the aqueous medium.
The InCl3 catalyzed and uncatalyzed reactions did not
work at pH 7.0: only 2-3% conversion was observed after
24 h. At pH 4.0, the InCl3 uncatalyzed reaction was
complete after 64 h, and only anti-â-hydroxy-R-iodocar-
boxylic acid 5a was quantitatively isolated. At pH 4.0,
the InCl3 catalyzed reaction was complete in only 2.5 h
and afforded 78% 4a , 20% anti-R,â-dihydroxyhexanoic
acid (diol),13 and only 2% of 5a . At pH 1.5, the Lewis acid
uncatalyzed reaction was fast (86% conversion after 3h)
but poorly regioselective (4a /5a ) 36/64), while the InCl3
catalyzed iodolysis afforded 4a exclusively in 0.5 h. At
pH 0.0, both InCl3 catalyzed and uncatalyzed reactions
occurred very rapidly (0.3 and 0.7 h, respectively) but
were unregioselective (4a /5a ) 65/35 and 43/57, respec-
tively). The poor regioselectivity of InCl3 catalyzed io-
dolysis was due to the competition with Brønsted acid
catalysis. Summing up, the regioisomers 4a and 5a were
isolated quantitatively working in water at pH 1.5 in the
presence of 10 mol % InCl3 and at pH 4.0 in the absence
of Lewis acid catalyst, respectively.
ness of Lewis acid catalysts on these reactions have been
previously investigated.
Continuing our research on organic synthesis in aque-
ous medium9a-e and encouraged by the fact that Lewis
acid catalyzed azidolysis of R,â-epoxycarboxylic acids in
pure water occurs with higher yields and selectivity than
when it is carried out on the corresponding esters,9f we
decided to investigate the bromolysis and iodolysis of R,â-
epoxycarboxylic acids in pure water by using NaBr and
NaI as the sources of nucleophiles and InBr3 and InCl3
as the catalysts, respectively.
InCl3 has received much attention in recent years, and
its effectiveness in many organic reactions was recently
reviewed.10 To date, InBr3 has not been used much.11
Both indium salts have been used mainly in organic
solvents and rarely in pure water.10,11
The investigation was then extended to a variety of
mono- and bisubstituited R,â-epoxycarboxylic acids 1. The
results are illustrated in Table 2.
Resu lts a n d Discu ssion
In exclusively aqueous medium at pH 1.5 and in the
presence of InCl3 (10 mol %), the iodolysis was fast, and
in every case, the â-iodo derivatives 4 were the sole
reaction products and were isolated with 88-95% yields.
At pH 1.5, in the absence of InCl3, the reaction is much
slower and generally unregioselective, indicating that the
competition of Brønsted acid catalysis in the course of
Lewis acid catalyzed reaction is modest.
First of all, we investigated the bromolysis of R,â-
epoxycarboxylic acids 1 (Scheme 1). The results of the
reactions to compounds 1a ,b,d ,e with NaBr, carried out
in pure water at 40 °C at pH 2.0 in the presence of InBr3
(5 and 10 mol %), are illustrated in Table 1. The reactions
were also carried out in the absence of InBr3 at the same
pH value to evaluate the importance of Brønsted acid
catalysis.
At pH 4.0, in the absence of InCl3, the reactions were
generally slow with little conversion. A sole R-regiose-
lectivity was observed for the iodolysis of trans-monoalky-
lated epoxyacids 1a ,b (entries 3 and 6), which gave
exclusively the anti-â-hydroxy-R-iodocarboxylic acids 5a,b,
respectively. The cis-R,â-epoxycarboxylic acid 1c gave a
prevalence of R-iodo derivative (entry 9), and the R,â-
epoxycyclohexane carboxylic acid (1e) did not give a
reverse regioselectivity with respect to that observed at
pH 1.5, either in the absence or in the presence of InCl3
(entries 14-17). The epoxyacids 1d ,f,g at pH 4.0 gave a
complex mixture of products, and a similar behavior was
observed for the iodolysis of â-methyl-R,â-epoxybutyric
acid and 2-methyl-1-oxaspiro[5.2]octane-2-carboxylic acid,
as already found in the iodolysis of analogous carboxylic
esters.6c
InBr3 efficiently catalyzed the bromolysis and promoted
the production of the anti â-adducts 2. However, it was
difficult to develop a protocol for a wide range of R,â-
epoxycarboxylic acids, and sometimes, it was difficult to
isolate the reaction products from the reaction mixture.
Thus, we turned to the iodolysis reaction.
The trans-R,â-epoxyhexanoic acid 1a was taken as a
model, and it was treated in pure water with NaI12 (5
mol/equiv) in the presence and in the absence of InCl3
(10 mol %), keeping the pH of the reaction medium
constant for the entire reaction time. The control of the
pH is fundamental for the success of the reaction (see
Experimental Section). The results of experiments carried
out at different pH values (7.0, 4.0, 1.5, and 0.0) are
illustrated in Figure 1 where the regioselectivity of the
reaction, expressed as percentage of anti-R-hydroxy-â-
The behavior of diastereoisomeric trans- and cis-R,â-
epoxyhexanoic acids 1a ,c gives some indications about
the reactive species that regulate the stereochemistry of
iodolysis. In the absence of InCl3, on going from pH 4.0
to pH 1.5, the percentage of nucleophilic attack on the
â-carbon of the oxirane ring increases for both diastere-
oisomers (Table 2, entries 1,3 and 7,9), while in the
presence of InCl3 at pH 1.5, only â-selectivity was
observed. This means that when the hydrogenionic
concentration increases, the Brønsted acid catalysis tends
to give the same regioselectivity as Lewis acid catalyzed
reactions. The same thing occurred for 1b. The reason
(8) Sharghi, H.; Naeimi, H. Bull. Chem. Soc. J pn. 1999, 72, 1525-
1531.
(9) (a) Fringuelli, F.; Piermatti, O.; Pizzo, F. Organic Synthesis in
Water; Grieco, P. A., Ed.; Blackie Academic and Professional: London,
1998; pp 223-249 and 250-261. (b) Fringuelli, F.; Piermatti, O.; Pizzo,
F.; Vaccaro, L. J . Org. Chem. 1999, 64, 6094-6096. (c) Fringuelli, F.;
Pizzo, F.; Vaccaro, L. Synlett 2000, 311-314. (d) Fringuelli, F.; Pizzo,
F.; Vaccaro, L. Synthesis 2000, 646-650. (e) Fringuelli, F.; Piermatti,
O.; Pizzo, F.; Vaccaro, L. Eur. J . Org. Chem. 2001, 439-455. (f)
Fringuelli, F.; Pizzo, F.; Vaccaro, L. J . Org. Chem. 2001, 66, 3544-
3548.
(10) Babu, G.; Perumal, P. T. Aldrichimica Acta 2000, 33, 16-22.
(11) (a) Marshall, J . A.; Grant, C. M. J . Org. Chem. 1999, 64, 8214-
8219. (b) Ceschi, M. A.; de Araujo Felix, L.; Peppe, C. Tetrahedron Lett.
2000, 41, 9695-9699.
(12) The use of LiI, KI, and MgI2 was also investigated. NaI is the
salt that, when the reaction is carried out in the absence of catalyst at
pH 1.5 and 40 °C, gives the lowest reaction conversion. All salts
investigated, when used in the absence of Lewis acid catalyst, gave
mixtures of â- and R-iodohydrins (30-70%).
(13) The formation of diol is not the result of hydrolysis of iodohydrin
4a . It comes from the hydrolysis of epoxyacid 1a . Indeed, by submitting
4a to the same reaction conditions used for the iodolysis of 1a , no diol
was detected.