methylated alkene after proton elimination. Furthermore,
such an active intermediate could also be trapped by nucleo-
philic functional groups before elimination. In 2012, Buch-
wald and co-workers reported an intramolecular oxytrifluo-
ACHTUNGTRENNUNGroACHTUNGTRENNUNGmethylation of unactivated alkenes using carboxyl or hy-
droxyl groups as nucleophilic traps.[7] Very recently, Sodeo-
ka[8] and Liang[9] expanded this method to aryl-, amine-, and
oxime-tethered alkenes (Scheme 1). However, the a-CF3-
alkyl active species trapped by strongly electron-deficient
groups has not been accomplished yet. Herein, we report
a trifluoromethylation/annulation reaction of electron-defi-
cient, aromatic-ring-tethered alkenes, providing not only
a practically useful methodology for trifluoromethylation,
but also a straightforward access to medically valuable tri-
fluoromethylated 1,2-benzothiazinane dioxide derivatives
(Scheme 1).
At the outset of this investigation, we optimized the reac-
tion conditions by testing different catalysts, ligands, and sol-
vents (Table 1). First, we screened various copper salts using
2,2’-biquinoline (L1) as ligand and acetonitrile as solvent
(Table 1, entries 1–7). Several cuprous catalysts, such as
2-carboxylate (CuTC), could catalyze this reaction, with the
desired product 2a formed in 6–15% yields (Table 1, en-
tries 1–5); [CuACTHNUTRGNE(UNG CH3CN)4]PF6 led to the highest yield (27%)
(Table 1, entry 6). However, CuBr2, a cupric salt, had no cat-
alytic activity in this reaction (Table 1, entry 7). Next, sever-
al ligands were examined (Table 1, entries 8–11). For 2,2’-bi-
pyridine (L2) and 1,10-phenanthroline (L3), only trace
amounts of 2a were afforded (Table 1, entries 8 and 9), simi-
lar to the case in which no ligand was added (Table 1,
entry 11). Di-2-pyridyl ketone (L4) was less effective than
L1, furnishing 2a only in 10% yield (Table 1, entry 10). Fur-
thermore, upon examination of solvent effect, we found that
CH3CN was slightly better than dichloroethane (DCE;
23%) (Table 1, entries 6 and 12, respectively). While in
other solvents, such as toluene, ethanol, and DMF, the reac-
tion did not take place (Table 1, entries 13–15). When
CuCl, CuBr, CuI, CuOTfACHTUNGTRNEGNU(PhH)0.5, and copper(I) thiophene-
Table 1. Optimization of the reaction conditions.[a]
a second portion of [CuACTHNUTRGNE(UNG CH3CN)4]PF6, L1, and Togni re-
agent was added into the reaction mixture after 2 h, the
yield of 2a increased to 43% (Table 1, entry 16). Consider-
ing that the relatively low yield is probably due to the steric
hindrance of the N-protecting isopropyl group, 1b with
a less sterically bulky ethyl group was used as the substrate,
and the yield of the corresponding product 2b was improved
to 50% (Table 1, entry 17). Furthermore, if another portion
of [CuACTHNUGTREN(UNG CH3CN)4]PF6, L1, and Togni reagent was added into
the reaction mixture of 1b, product 2b was formed in 99%
NMR yield (Table 1, entry 18). However, when two portions
of [CuACTHNUGTREN(UNG CH3CN)4]PF6, L1, and Togni reagent were added in
Entry
[Cu]
Ligand
Solvent
Yield [%][b]
1
2
3
4
5
6
7
8
9
CuCl
CuBr
CuI
CuOTf
CuTC
L1
L1
L1
L1
L1
L1
L1
L2
L3
L4
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
DCE
toluene
EtOH
DMF
CH3CN
CH3CN
CH3CN
CH3CN
MeOH
15
11
6
9
13
27
(PhH)0.5
[Cu
N
CuBr2
NR[h]
trace
trace
10
trace
23
NR
NR
NR
43
a one-pot manner, the yield of 2b decreased to 69%
(Table 1, entry 19). As for Na2SO2CF3, a much less expen-
[Cu
[Cu
[Cu
[Cu
[Cu
[Cu
[Cu
[Cu
[Cu
[Cu
[Cu
[Cu
[Cu
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
sive trifluoromethylation reagent,
a
mixture of 1b
10
11
12
13
14
15
16[c]
17[d]
18[c,d]
19[d,f]
20[d,g]
ACHTUNGTRENNUNG
(0.1 mmol, 1.0 equiv), [Cu(CH3CN)4]PF6 (0.15 equiv), L1
AHCTUNGTRENNUNG
[e]
T
–
(0.30 equiv), Na2SO2CF3 (3.0 equiv), tert-butyl hydroperox-
ide (TBHP) (4.0 equiv), and MeOH (1.0 mL) was stirred at
608C under argon atmosphere for 12 h, affording only trace
amount of product on the basis of 19F NMR spectroscopy
(Table 1, entry 20).
With the optimized reaction conditions in hand, we next
investigated the generality of this method and the results
are summarized in Table 2. As for substrate 1a, the corre-
sponding product 2a was isolated in 52% yield when anoth-
N
L1
L1
L1
L1
L1
L1
L1
L1
L1
N
ACHTUNGTRENNUNG
G
R
G
50
99
69
trace
N
T
A
[a] Reaction conditions: Cu cat. (0.15 equiv), ligand (0.30 equiv), Togni
reagent II (1.0 equiv), 1a (0.1 mmol, 1.0 equiv), solvent (1 mL), 608C,
argon atmosphere. [b] The yields were determined by 19F NMR using
er two portions of [CuACTHNUTRGNEUN(G CH3CN)4]PF6, L1, and Togni reagent
were added into the reaction mixture (Table 2, entry 1). To
our delight, when less sterically bulky ethyl, methyl, or hy-
drogen were introduced as N-protecting groups, such as sub-
strates 1b–1d, the desired products 2b–2d were obtained in
excellent yields (89–93%), respectively (Table 2, entries 2–
4). N-Tethered functional groups: OH, Br, and NHTs, as
well as three-membered epoxide and cyclopropyl rings, were
PhCF3 as an internal standard. [c] A second portion of [CuACHTNUTRGNENUG(CH3CN)4]PF6
(0.15 equiv), L1 (0.30 equiv), and Togni reagent II (1.0 equiv) was added
after 2 h. [d] 1b was used as the substrate instead of 1a. [e] No ligand
was added. [f] [Cu
reagent II (2.0 equiv) were added in a one pot manner. [g] Reaction con-
ditions: [Cu(CH3CN)4]PF6 (0.15 equiv), L1 (0.30 equiv), Na2SO2CF3
ACHTUNGTRENNUNG(CH3CN)4]PF6 (0.30 equiv), L1 (0.60 equiv), and Togni
ACHTUNGTRENNUNG
(3.0 equiv), TBHP (4.0 equiv), 1b (0.1 mmol, 1.0 equiv), MeOH (1.0 mL),
608C, argon atmosphere, 12 h. [h] NR=no reaction.
&
2
&
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!