Vol. 68, No. 3 (2020)
Chem. Pharm. Bull.
289
oxime (2). Conjugate addition to the α,β-unsaturated ketone 12 and 13). However, toluene afforded not only the desired
and subsequent protonation of the resulting Pd(II) enolate 3-(4-methoxyphenyl)-2-cyclohexen-1-one O-methyl oxime (4a)
typically occurs.10,14,15) However, in the O-methyl oxime, it is but also a mixture of undesired β-arylated O-methyl oximes,
thought that β-hydrogen elimination occurred (Chart 3). The 3-tolyl-2-cyclohexen-1-one O-methyl oximes (4e–g),32) result-
reaction yield was low (11% yield), but this series of reac- ing from starting O-methyl oxime (3) and toluene instead of
tions, oximation of ketone (1) followed by β-arylation of O- 4-iodoanisole. Finally, Pd catalysts were investigated (entries
methyl oxime (3), had the synthetic advantage of giving α,β- 14 and 15). As a result, replacing Pd(OAc)2 with Pd(PPh3)2Cl2
unsaturated O-methyl oxime (4a) which could not be obtained afforded the desired product (4a) in 48% yield (entry 15).
by β-arylation of the ketone (1) and subsequent oximation.
The resulting product (4a) was only the E-isomer, which was
To improve the yield of the resulting O-methyl oxime (4a), determined by a 2D nuclear Overhauser effect spectroscopy
we examined the reaction conditions (Table 1). Using Pd(OAc)2 (NOESY) NMR experiment. Along with the product (4a), only
and AgOAc, the yield was twice as much as that using the tri- the starting O-methyl oxime (3) was recovered (46% recov-
fluoroacetate counterion (entry 2). Cu(OAc)2, AgTFA, Cs2CO3 ery). O-Methyl oxime (2) was not observed. From the ratio
and NaOtBu were less effective as additives (entries 3–6). It of E- and Z-isomers of recovered starting O-methyl oxime
was thought that the less electron-rich PPh3 would be more (3) (existing as approx. 2:3 E/Z mixture), it was considered
suitable as the phosphine ligand (entries 8 and 9), but the pres- that there was no possibility of isomerization of the O-methyl
ence or absence of PPh3 was not crucial in this reaction (entry oxime under this reaction conditions.33)
9 vs. entry 10). In contrast to the phosphine ligands, AgOAc
With the optimized conditions in hand, the substrate scope
was essential (entry 11). Solvent effects were also surveyed. of the aryl iodides was investigated (Table 2). Substitutions
Only 1,4-dioxane without 1,1,1,3,3,3-hexafluoro-2-propanol on the aryl group at the ortho, meta, or para positions were
(HFIP) resulted in a decreased yield (entry 7), thus HFIP all tolerated (4a–c, 4e–h). Aryl iodides with both electron-
was important. Toluene also proved to be efficient (entries donating and electron-withdrawing groups participated to give
the corresponding β-arylated O-methyl oximes (4i–l). Under
this reaction conditions, a variety of functional groups were
tolerated, including nitriles (4i), esters (4k), and nitro groups
(4l). In all cases, E-isomers were obtained preferentially over
Z-isomers.34) The E- and Z-isomers were easily separable by
silica gel column chromatography. The exact reason for the
different ratios of resulting E- and Z-isomers based on the
Chart 3. Proposed Mechanism for β-Arylation of O-Methyl Oxime (3)
Table 2. Substrate Scope of Aryl Iodidesa)
Table 1. Optimization of Reaction Conditionsa)
Pd catalyst
(10mol%)
Ligand
(20mol%)
Additive
(2 equiv.)
Yield
(%)
Entry
1
2
Pd(TFA)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(dppf)Cl2
Pd(PPh3)2Cl2
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
PCy3
P(iPr)3
PPh3
None
PPh3
PPh3
None
None
None
AgTFA
AgOAc
Cu(OAc)2
AgTFA
Cs2CO3
NaOtBu
AgOAc
AgOAc
AgOAc
AgOAc
None
11
22
6
3
4
2
5b)
6b)
7b)
8
8
7
9
28
34
30
0
9
10
11
12c)
13c)
14
15
AgOAc
AgOAc
AgOAc
AgOAc
37
32
36
48
a) All the reactions were run with 3 (0.2mmol) and 4-iodoanisole (0.2mmol) in
1.0mL solvent for 18h. b) HFIP was not added. c) Toluene was used instead of
1,4-dioxane.
a) All the reactions were run with 3 (0.2mmol) and aryl iodide (0.2mmol) in
1.0mL solvent for 18h.