Table 1. Examination of regioselective diaryl ether formation
OBn
O
Entry Reagents
PhI(OAc)2 EtOH
1.0 equiv)
K2CO3
3.0 equiv)
PhI(OAc)2 CH2Cl2/EtOH (= 5/1)
2.0 equiv)
K2CO3
5.0 equiv)
PhI(OAc)2 CH3CN/EtOH (= 5/1)
2.0 equiv)
K2CO3
5.0 equiv)
PhI(OAc)2 ClCH2CH2Cl/EtOH (= 5/1) 11:12 = 3:2
2.0 equiv)
K2CO3
5.0 equiv)
PhI(OAc)2 PhCl/EtOH (= 5/1)
2.0 equiv)
K2CO3
5.0 equiv)
PhI(OAc)2 PhCl
2.0 equiv)
K2CO3
5.0 equiv)
Solvents
Resultsa
2
1
2
3
4
5
6
11:12 = 0:1
1
1
O
(
I
O
O
Ph
(
11:12 = 1:4
11:12 = 1:2
13a
(
PhI(OAc)2
K CO
aprotic
solvent
EtOH
2
3
4
(
Ph
OEt
I
OBn
O
(
O
12
4
O
(
O
(
13b
(
11:12 = 3:1
(11: 52%, 12: 18%)
Scheme 3. Alternative mechanistic modes of regioselective
etherification.
b
(
(
spectroscopic data of synthetic (+)-3 matched exactly those
of authentic natural aspercyclide C, thereby confirming its
identity. The optical rotation of our synthetic material was found
to be identical to Fürstner’s report.
decomposed
1
(
4
(
In summary, an efficient regioselective total synthesis of
+)-aspercyclide C (3) was accomplished by virtue of a
a
1
b
Determined by H NMR analysis. Isolated yield.
(
versatile, intramolecular, oxidative diaryl etherification tactic.
It should be noted that the direct chemo- and regioselective aryl
CO bond formation was realized without the need to resort to
aryl Chalogens or aryl COtriflates. More interestingly, the
regiocontrol in the macrocyclization step was manifested by
appropriately tuning the polarity and nucleophilicity of the
reaction media. Collectively, our synthetic strategy described
herein represents a new and concise entry to the syntheses of
clization precursor 4 in 93% yield over two steps. The key
intramolecular, oxidative aryl CO bond formation was then
examined using PhI(OAc)2 and K CO3 in ethanol at room
2
8
temperature. In practice, this step proceeded with high chemo-
selectivity but resulted in exclusive formation of the undesired
C4-bonded diaryl ether 12 (Table 1, Entry 1). We thus decided
to investigate solvent effects to alter the regioselectivity by
careful consideration of the potential mechanistic modes of
cyclization (Scheme 3). Thus, the putative aryloxy iodo-
nium(III) intermediate 13b, which would afford undesired 12,
was speculated to equilibrate with a cyclic intermediate (e.g.,
1
3,14
various bioactive natural and unnatural diaryl ethers.
This work was supported financially by a Grant-in-Aid for
Specially Promoted Research from the Ministry of Education,
Culture, Sports, Science and Technology (MEXT) in Japan.
1
3a) in less polar aprotic media, which would lead to the desired
12
cyclic diaryl ether 11. Under this speculation, various aprotic
cosolvents were added to the ethanolic reaction mixture
References and Notes
(
Table 1). Indeed, mixed solvent systems such as CH2Cl2/EtOH
1
a) T. Ishizaka, D. H. Conrad, E. S. Schulman, A. R. Sterk, K.
and CH CN/EtOH started to produce the desired regioisomer
3
1
1, although undesired 12 still predominated (Entries 2 and 3).
While the solvent mixture ClCH2CH2Cl/EtOH favorably gave
1 in an improved ratio (11:12 = 3:2, Entry 4), a 5:1 ratio of
2
1
PhCl/EtOH was eventually found to yield 11 as the major
regioisomer (11:12 = 3:1, Entry 5). By careful thin layer
chromatography (TLC) examination of the reaction, an unstable
intermediate was transiently detected. High-resolution mass
spectrometry (ESI) of the intermediate (m/z found 677.1761)
indicated that the species corresponded to the putative iodonium
3
4
5
6
7
+
species 13a (calcd for C H IO [M + H] 677.1764), which
36
38
5
disappeared as the reaction proceeded to give 11. In the absence
of ethanol, aryl ether products were not produced and decom-
position was observed (Entry 6). Thus the presence of ethanol
was found to be indispensable for the reaction to proceed via the
desired oxidative pathway.
a) J. L. Carr, D. A. Offermann, M. D. Holdom, P. Dusart,
46, 1824. b) J. L. Carr, J. J. P. Sejberg, F. Saab, M. D.
Lastly, cleavage of the benzyl ether of 11 with BCl3
at ¹78 °C gave (+)-aspercyclide C (3) in 87% yield. The
© 2014 The Chemical Society of Japan