Organic Letters
Letter
iodosuccinimide (NIS) as the electrophile, and PhSO2H as the
nucleophile (Scheme 2). During the substrate scope, we
reaction was rather miscellaneous with lower yield (Table 1,
entries 2 and 3). Furthermore, when NFSI was employed, the
reaction was sluggish, and the corresponding product 2a was
obtained in <5% yield (Table 1, entry 4). After NIS was
identified as the best additive for this transformation, we
turned our attention to the solvent screening. It was gratifying
that except for H2O that gave only 20% yield (Table 1, entry
10), both polar and nonpolar solvents gave high yields (Table
1, entries 5−9). Finally, CH2Cl2 was determined to be the best
solvent for this reaction. Moreover, the NIS loading was also
investigated (Table 1, entries 11−13). It was found that 1.0
equiv of NIS was essential for complete conversion, and the
yield was nearly consistent with the equivalent of NIS used in
the reaction. Finally, we conducted the reaction within a
shorter period and found that this transformation went rapidly
and was completed within 1 h (Table 1, entry 14).
Scheme 2. Our Previous Work and the Present Work
With the optimized reaction conditions in hand, we next
investigated the scope of different 2-alkynylnaphthols. First,
substituents on the ortho-position of naphthalene (R1 groups)
were tested (Scheme 3, 2b−2f). As shown in Scheme 3, the
phenyl group with para-methyl substituent gave an excellent
yield (95%, 2b), whereas the phenyl group with a para-phenyl
substituent gave a relatively lower yield (91%, 2c). Next, a
series of ether substituents was investigated (2d−2f), and these
substrates gave moderate yields ranging from 71 to 80%.
Furthermore, substrates with a methyl substituent on ring A or
ring B were found to be employable, giving the corresponding
products in 82 and 89% yields (2g, 2h). Halogen functional
groups F, Cl, and Br were also introduced to the substrates,
and all of the tested substituents at different positions of 2-
alkynylnaphthols were well-tolerated and afforded the desired
bridged polycyclic compounds (2i−2p) in good yields (82−
99%). Finally, the substrate with a substituent at the naphthol
moiety was tested, and the reaction went smoothly, affording
2q in 95% yield (Scheme 4).
To further demonstrate the synthetic practicality of this
transformation, a gram-scale synthesis was carried out to give
2a in 98% yield. Next, the carbonyl group of 2a was easily
reduced to hydroxyl by L-selectride with high stereoselectivity
in 93% yield. Besides, treatment of 2a with meta-
chloroperbenzoic acid in CH2Cl2 underwent a Baeyer−Villiger
oxidation and gave the corresponding lactone 4 in 86% yield.
After that, the lactone 4 was reduced to hemiketal by using
lithium aluminum hydride as reductant and afforded 5 in 83%
yield. The exact structure of 5 was determined by single crystal
X-ray crystallographic analysis. Next, we carried out an
enantioselective transformation using a chiral spirocyclic
phosphoric acid catalyst.9 Preliminary results (20% ee, 92%
yield) showed that this transformation also proceeded
enantioselectively.
happened to find that a trace of byproduct was observed in
some specific cases. When we changed the reaction condition
by removing PhSO2H and catalyst, this byproduct was formed
in almost complete conversion. Subsequent structural identi-
fication suggested that the product was likely generated from a
dimerization6 of the substrate. The exact structure was later
determined to be a bridged polycyclic compound containing a
bicyclo[3.2.1]octane moiety7 by single crystal X-ray crystallo-
graphic analysis. Because this scaffold was so unique and the
transformation involved the alkene site of the VQM
intermediate which has never been explored, we decided to
conduct a circumstantial study and further exploit the versatile
reactivity of the VQM intermediate.
We commenced our investigation with 1a as the substrate
and NIS as the promoter8 in the presence of dichloromethane.
To our delight, the reaction went smoothly to give 2a as a
single product in almost quantitative yield (Table 1, Entry 1).
Next, we tested a set of N-halosuccinimides such as NCS (N-
chlorosuccinimide) and NBS (N-bromosuccinimide); under
the same reaction conditions, NCS promoted the reaction,
giving 2a in only 34% yield, whereas the NBS-promoted
a
Table 1. Optimization of Reaction Conditions
entry
additive (equiv)
solvent
time (h)
yield (%)
1
2
3
4
5
6
7
8
NIS (1.0)
NCS (1.0)
NBS (1.0)
NFSI (1.0)
NIS (1.0)
NIS (1.0)
NIS (1.0)
NIS (1.0)
NIS (1.0)
NIS (1.0)
NIS (0.1)
NIS (0.2)
NIS (0.5)
NIS (1.0)
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
toluene
Et2O
6
6
6
6
6
6
6
6
6
6
6
6
6
1
99
34
12
<5
96
95
98
97
95
20
8
To gain an insight into the reaction mechanism, two control
experiments were performed (Scheme 5). First, the naph-
thalene substituent of the substrate was replaced by a phenyl
group (substrate 6). Under the standard conditions, it went
through an intramolecular 5-endo-dig electrophilic cycliza-
tion,10 affording iodonaphtho[b]furan as the only product,
while the dimerization product was not detected. This result
indicated that the steric hindrance arising from the bulky
substituent plays a crucial role in forming the bridged
polycyclic compounds through dimerization of VQM. Next,
when the hydroxyl group on the substrate 1a was protected
(substrate 8), the reaction did not occur at all. Based on our
previous studies on VQM-mediated reactions, we speculated
THF
EA
9
CH3CN
H2O
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
10
11
12
13
14
19
48
99
a
Conditions: 1a (0.1 mmol), additive and solvent (1.0 mL) at rt.
B
Org. Lett. XXXX, XXX, XXX−XXX
Tan, Yu
