Organic Letters
Letter
a
NaNO2-catalyzed aerobic dearomatization of simple phenolics
at room temperature (Scheme 1c).
Table 1. Optimization of the Reaction Conditions
Our initial attempts focused on the catalytic dearomatization
of oxime ketone 1a, which was prepared from inexpensive and
commercially available 4-(4-methoxy-phenyl)-2-butanone
(raspberry ketone methyl ether, which is a component of the
essential oil).13 We were pleased to discover that the desired
spiro-cyclohexadienone isoxazoline 2a was indeed isolated in
8% yield in the presence of NaNO2 (1.2 equiv) and HCl (1.2
equiv). More interestingly, the extra use of a stable radical
scavenger, 2,2,6,6-tetramethylpiperidin-N-oxyl (TEMPO),
proved critical to improve the dearomatization efficiency.
Based on the above preliminary results, a possible reaction
pathway could be proposed as shown in Scheme 2. At first,
b
NaNO2
(equiv)
H2O
yield
(%)
entry
TEMPO
(equiv)
solvent
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
10 mol %
10 mol %
10 mol %
10 mol %
10 mol %
10 mol %
10 mol %
10 mol %
10 mol %
10 mol %
10 mol %
5 mol %
15 mol %
15 mol %
30 mol %
25 mol %
25 mol %
1.1 equiv
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
−
−
−
−
−
DCE
CHCl3
dioxane
THF
MeCN
EtOH
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
43
0
0
0
0
−
0
30
50
60
70
100
60
60
60
60
60
60
60
46
52
56
34
49
50
84
90
84
89
0
Scheme 2. Proposed Reaction Mechanism
c
16
0.2
−
−
cd
,
17
18
cd
,
0
a
A mixture of oxime (1a, 0.1 mmol), TEMPO, NaNO2, H2O in
solvent (1 mL) was treated with concentrated HCl (1 equiv) at −5
°C, and the reaction was allowed to warm to room temperature for 36
b
c
h. HPLC yield with diphenyl sulfide as the internal standard. Under
d
O2 (1 atm). Without HCl.
stoichiometric amount of TEMPO (Table 1, entries 17 and
18).
ketone oxime 1a is oxidized to form radical cation I by NO+,14
produced in situ from NaNO2 under acidic conditions. The
resulting radical cation I could cyclize and lose a proton to
furnish radical intermediate II. Without any radical capturing,
the intermediate II could easily revert to its previous state,
preventing its further production to the stable spiroisoxazoline.
However, steered by the persistent radical effect,15 the radical
II couples with TEMPO to afford intermediate III. Under
acidic conditions, a fragmentation of enol ether intermediate
III could produce IV and TEMPOH. Meanwhile, both
TEMPO and NO+ could be regenerated by O2, along with
HCl.16 Hydrolysis of IV eventually delivers the final
spiroisoxazoline 2a. Based on this postulated mechanism,
extensive evaluation of various nitrite salts, additives, and
solvents was therefore performed, with the objective of
improving the reaction efficiency (see Table 1).
Using a catalytic amount of TEMPO, and oxime ketone 1a
as the substrate, we first checked the solvent effect, which
revealed that dichloroethane (DCE) was a better choice than
chloroform (CHCl3), dioxane, tetrahydrofuran (THF),
MeCN, and EtOH (Table 1, entries 1−6). The addition of
water further increased the reaction efficiency (Table 1, entries
7−11). As it turned out to accelerate the hydrolysis, 60 equiv
of H2O were found to be optimal (Table 1, entry 9). To our
delight, employing 25 mol % of TEMPO could significantly
improve the yield to 90% (Table 1, entries 12−15).
Furthermore, the amount of NaNO2 could be reduced to 0.2
equiv in the presence of O2 without affecting the reaction
performance (Table 1, entry 16). In contrast, in the absence of
NaNO2 or HCl, no reaction occurred, even with a
With the optimized reaction conditions in hand, we moved
to the reaction scope study (Scheme 3). First, we explored the
scope of various aryl methyl ethers (Scheme 3A). Ketone
oxime 1a proceeded smoothly to afford 2a in 88% isolated
yield. Gratifyingly, 2a could be obtained on a gram scale in
83% yield. The introduction of an additional R1 substituent at
the benzylic position was also tolerated. For example, aerobic
oxidative dearomatization of ketone oxime 1b afforded 2b in
74% yield. The applicability of this dearomatization was further
demonstrated by varying R substituent in ketone oximes,
affording 2c−2r in good to high yields. Aldehyde and ester
functional groups are compatible with our conditions, with 2c
and 2d forming in 82% and 91% yield, respectively. Pleasingly,
various para-substituents on the phenyl ring were tolerated,
regardless of electron-donating or electron-withdrawing
groups. Moreover, the substituents could be at the para
(2g), meta (2h), or ortho (2i) positions. Notably, naphthyl
(2p), 1,3-benzodioxole (2q), and thiophenyl(2r) were also
amenable to this dearomatization to deliver the desired
spiroisoxazolines in high yields. In addition, variation of R2
substituents in oximes was also possible. High yields of 2s−2x
were observed with electron-donating groups such as methoxy
(2s, 2t, 2w, 2x) and methyl (2u), and also electron-
withdrawing groups such as chloro (2v) at different positions.
It is noteworthy that spiroisoxazoline 2x can be prepared
directly from natural product loureirin A with this protocol.
Unfortunately, no desired reaction was observed with phenols
or phenyl ethers bearing strong electron-withdrawing groups
(−CO2Me or NO2), and this protocol did not show any
B
Org. Lett. XXXX, XXX, XXX−XXX