Communications
oxidation (e.g., KBrO3/H2SO4, NaOCl/H2O-TBABr, NaIO4/
H2O, MeCO3H/H2O, H2O2/H2O-Ac2O, Oxone/H2O, TBA-
OX/MeSO3H-CH2Cl2, DMDO/acetone or CH2Cl2; TBA =
amount of the epoxidized ortho-quinol rac-7a (8% yield),
and the observation of another minor product (vide infra).
Although the outcome of this reaction confirmed the
catalytic action of 1 f, which is oxidized in situ by m-CPBA
into some oxygenating iodane species (vide infra), m-CPBA
was also capable of trapping the initially formed ortho-quinol
intermediate through epoxidation of its D-4,5 bond. The use
of equimolar amounts of 1 f and m-CPBA did not improve the
reaction outcome, but a twofold excess of 1 f increased the
yield of 6a to the detriment of the epoxidation event (Table 1,
entries 2 and 3). On the contrary, only the epoxide 7a was
isolated in 47% yield when a large excess of m-CPBA
(10 equiv) and a catalytic amount of 1 f (0.1 equiv) were
added to a less concentrated solution of 5a in acetone, a
better solvent than CH2Cl2 for dissolving m-CPBA; the
higher dilution was applied with the aim of limiting the
dimerization process (Table 1, entry 4). The exclusive regio-
and diastereoselectivity observed for 7a (and other related
epoxides described below) in favor of a cis epoxidation of the
D-4,5 bond, relative to the orientation of the OH group at C6
of the orthoquinol intermediate, is a directivity gift offered by
this allylic hydroxy group.[9]
To additionally evaluate the scope and limitations of the
in situ generation of HPD-mediating iodanes, two additional
phenols (i.e., 5b and 5c) and 2-methylnaphthol (5d) were
utilized as starting materials (Table 2). Under catalytic
conditions, 2,4,6-trimethylphenol (5b) led to a complex
mixture from which the racemic dimer 6b[4b] and epoxide
7b could each be isolated in low yields. Traces of the ortho-
quinol m-chlorobenzoate 7b’ were also detected (Table 2,
entry 1). The latter compound, resulting from a surprising
(only observed when using 5b) participation of m-chloroben-
zoic acid in the dearomatization process, could be isolated in
11% yield from a reaction in which a twofold excess of 1 f was
used (Table 2, entry 2). Again, the use of such an excess of the
iodoarene prevented the epoxidation event, but the expected
formation of 6b in high yield did not occur.
tetra-n-butylammonium,
TBA-OX = tetra-n-butylammo-
nium oxone, DMDO = dimethyldioxirane),[6] but in most
cases satisfactory conversions were not observed.
The only bits of encouragement were the oxone-mediated
oxidation of 1a into the l5-iodane 4a in 62% yield and the
DMDO-mediated oxidation of 1d into the l3-iodane 3d in
yields ranging from approximately 50 to 90%, together with
the isolation of a white solid whose structure could be
attributed to the l5-iodane 4d (little structural evidence was
obtained to confirm this attribution). To test the oxygenating
and asymmetry-induction capabilities of these materials, (À)-
4a was engaged in a HPD reaction using 2,6-dimethylphenol
(5a) in THF at À788C. The expected ortho-quinol dimer 6a
(vide infra)[4b,d] was obtained in quantitative yield, but in only
20% ee. Under the same reaction conditions, the l3-iodane 3d
was totally inefficient, but each of its optical antipodes could
oxidize methyl p-tolylsulfide into the corresponding sulfox-
ides in good yields (ca. 50 to 70%), albeit in poor enantio-
meric excesses (7–8% ee).[7] Finally, each optical antipode of
the presumed l5-iodane 4d successfully mediated the HPD of
5a into 6a in yields of 70–75% with opposite but less than
10% ee (see the Supporting Information).
The overall difficulty in oxidizing our starting iodoarenes
into iodanes led us to look for another approach. We turned
our attention toward methodologies based on the in situ
generation of iodanes from iodoarenes in the presence of a
co-oxidant, a process having the possibility of proceeding with
only catalytic amounts of iodoarenes.[8] We therefore verified
the potential of this process in an HPD reaction, again using
2,6-dimethylphenol (5a) as a test substrate in various
combinations with 2-iodophenyl acetic acid (1 f) and m-
CPBA, which was recently introduced by Kita and co-
workers[8e] and Ochiai and co-workers[8d] as a terminal co-
oxidant of choice in related iodoarene-catalyzed reactions. By
using 0.1 equivalent of 1 f and 1.0 equivalent of m-CPBA in
CH2Cl2 at room temperature for 2 hours (Table 1, entry 1), 5a
was partially converted into the expected racemic dimer 6a,
which was isolated in a 55% yield. Some starting 5a was also
recovered (16% yield), together with the isolation of a small
Under catalytic conditions, carvacrol (5c) was converted
into biscarvacrol (6c)[2a,4b] and epoxide 7c, together with
significant amounts of the para-quinone 7c’ (Table 2, entries 3
and 4). We presume that 7c’ resulted from a 1,2-hydride shift
(also known as an NIH shift)[10] of the epoxide 7c in the acidic
medium used (see arrows in Table 2). In fact, the same
transformation also occurred with the disubstituted epoxide
7a, but only traces (up to less than 10% as estimated by NMR
analysis) of the corresponding quinone were observed (vide
supra). This epoxide chemistry was again of no concern when
using a twofold excess of 1 f, which led to the isolation of
exclusively 6c in 41% yield (Table 2, entry 5).
Table 1: Preliminary evaluation of reaction conditions for in situ
generation of HPD-mediating iodanes from iodoarenes.[a]
The reaction outcome was much simpler when using 2-
methylnaphthol (5d), the hydroxylative dearomatization of
which led to a nondimerizing ortho-quinol rac-6d and/or to its
epoxide derivative rac-7d (Table 2, entries 6–9). Most
remarkably, the ortho-quinol 6d was obtained in 82% yield
when using a twofold excess of the iodoarene 1 f with an
equimolar amount of m-CPBA (Table 2, entry 8), whereas
only the epoxide 7d was generated in high yield when using a
larger excess of m-CPBA (here limited to 2.5 equiv) and a
catalytic amount of 1 f (Table 2, entry 9).
Entry
1 f [equiv]
m-CPBA
[equiv]
Recovered
5a [%]
6a [%]
7a [%][b]
1
2
0.1
1
2
1
1
1
10
16
18
13
n.d.
55
58
64
n.d.
8
3
n.d.
47
3
4[c]
0.1
[a] The reactions were conducted in CH2Cl2 for 2 h, [5a]=0.33m, unless
otherwise noted. [b] Relative stereochemistry determined by NOESY
experiments. [c] [5a]=0.05m in acetone. n.d.=not detected.
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 4605 –4609