Communication
paved the way for the development of new reactions, which
are otherwise not easily accessible by either catalyst independ-
ently.[8] In addition, the Sanford group nicely demonstrated
a high-yielding aromatic CÀH arylation reaction which involves
a Pd-catalyzed CÀH activation and a visible-light-induced pho-
toredox catalysis (Scheme 1b).[9] Shortly thereafter, the Rovis
group disclosed an efficient catalytic asymmetric a-acylation of
tetrahydroisoquinolines with aldehydes, in which the photoca-
talyst [Ru(bpy)3Cl2] and chiral N-heterocyclic carbene catalyst
have been elegantly employed (Scheme 1c).[10] Furthermore,
MacMillan and co-workers have developed a photocatalytic a-
amino CÀH arylation (Scheme 1d),[11] in which the single pho-
tocatalyst Ir(ppy)3 works in a bifunctional catalytic mode;
therefore no extra co-catalyst, reductive or oxidative reagent is
required. Quite recently, by combining two SET processes, the
Rueping group successfully developed a photocatalytic Mi-
chael addition/aromatization/CÀC bond cleavage cascade reac-
tion, which enabled the formation of indole-3-carboxaldehyde
derivatives in high yields.[5h] Shortly after, the Cho group re-
ported a hydrotrifluoromethylation of alkynes through a photo-
catalytic iodotrifluoromethylation and deiodination sequen-
ce.[5i]
In addition to the SET process, the Yoon group[12c] and our
laboratory[12d] have independently exploited an energy transfer
(ET) process[12] for visible-light-photocatalytic [2+2] cycloaddi-
tions. As part of our ongoing projects on photocatalysis[13] and
tandem catalysis,[14] we wondered whether we could develop
a phototandem catalysis that combines two mechanistically
distinct processes (e.g. ET and SET). It is well known that some
azides can undergo a synthetic transformation through an ET
pathway in the presence of photosensitizers under irradiation
of visible light.[12e,13f] We assumed that structurally similar diaz-
oamides might be activated through an ET process as well.
Based on this hypothesis, we have recently developed a visi-
ble-light-induced phototandem catalysis that relies on a single
iridium photocatalyst. This unprecedented process allows an
efficient transformation of diazoamides[15] into biologically and
synthetically important 3-ester-3-hydroxy-2-oxindole deriva-
Table 1. Optimization of reaction conditions for the first cyclization
step.[a]
Entry
Solvent
Catalyst
t [h] Yield [%][b]
1
2
3
4
5
6
7
8
CH2Cl2
EtOH
MeCN
DMF
CHCl3
DCE
(CF3)2CHOH
CF3CH2OH
CF3CH2OH
CF3CH2OH
CF3CH2OH
CF3CH2OH
CF3CH2OH
CF3CH2OH
CF3CH2OH
CF3CH2OH
[Ir(ppy)2(dtbbpy)]PF6
[Ir(ppy)2(dtbbpy)]PF6
[Ir(ppy)2(dtbbpy)]PF6
[Ir(ppy)2(dtbbpy)]PF6
[Ir(ppy)2(dtbbpy)]PF6
[Ir(ppy)2(dtbbpy)]PF6
[Ir(ppy)2(dtbbpy)]PF6
[Ir(ppy)2(dtbbpy)]PF6
–
48
48
48
48
48
48
48
48
48
48
48
48
6
72
<5
<5
<5
73
71
84
90
20
<5
16
33
94
43
57
27
9
10[c]
11[d]
12[e]
13
14
15
16
–
[Ir(ppy)2(dtbbpy)]PF6
[Ir(ppy)2(dtbbpy)]PF6
[Ir(dF(CF3)ppy)2(dtbbpy)]PF6
.
Ru(bpy)3Cl2 6H2O
48
48
48
Eosin Y
Ir(ppy)3
[a] Unless otherwise noted, the reaction was carried out with 1a
(0.2 mmol), photocatalyst (2 mol%), solvent (4.0 mL), 18W CFL, at room
temperature. [b] Determined by 1H NMR analysis using 1,3,5-trimethoxy-
benzene as an internal standard. [c] With 3W red LED strip as light
source. [d] Without a light source. [e] Without degasings. ppy=2-phenyl-
pyridine; dtbbpy=4,4’-di-tert-butyl-2,2’-bipyridine; dF(CF3)ppy=2-(2,4-di-
fluorophenyl)-5-(trifluoromethyl)pyridine; bpy=2,2’-bipyridine; DMF=
N,N-dimethylformamide; DCE=1,2-dichloroethane. CFL=compact fluo-
rescent lamp. LED=light-emitting diode.
sence of photocatalyst. It was found that no cyclized product
2a was formed even after 48 h (Table 1, entry 10). In addition,
the reaction did not occur in the presence of photocatalyst
without irradiation of visible light (Table 1, entry 11). These re-
sults showed that both visible light and photocatalyst were
critical to the cyclization. Degassing was necessary, as exposure
of the reaction to the air only gave rise to 2a in 33% yield
(Table 1, entry 12). Catalyst screening disclosed that [Ir(dF-
(CF3)ppy)2(dtbbpy)]PF6 was optimal in terms of the reaction ef-
ficiency (Table 1, entry 13 vs entries 8, 14–16).[17]
tives through
(Scheme 1e).[16]
a
cyclization/aerobic oxidation cascade
We initially chose diazoamides 1a as the model substrate to
examine the feasibility of the first cyclization step. To our de-
light, irradiation of 1a with 18W white LED afforded 3-ester-2-
oxindole (2a) in 72% yield in the presence of [Ir(ppy)2-
(dtbbpy)]PF6 in dichloromethane at room temperature. A brief
screen of reaction media identified CF3CH2OH to be the best
solvent of choice, and the cyclization product 2a was obtained
in 90% yield (Table 1, entry 8). Interestingly, a 20% yield (as de-
termined by NMR spectroscopy) of 2a was observed in the ab-
sence of photocatalyst after 48 h, when a solution of 1a was
subjected to 18W compact fluorescent light source (Table 1,
entry 9). UV/Vis spectra of 1a indicated that 1a could absorb
UV light (lmax =242 and 266 nm). As we noted some adsorp-
tion in the UV range under illumination with 18W compact flu-
orescent light, the reaction might proceed through a traditional
UV light-mediated cyclization. Thus, we have performed the
control experiment with a 3W red LED (l=617 nm) in the ab-
With the optimal reaction conditions in hand for the first
cyclization, we turned our attention to optimize the reaction
conditions for the tandem cyclization/oxidation sequence with
oxygen as a terminal oxidant in a one-pot fashion (Table 2). Ini-
tially, upon complete consumption of 1a, the reaction mixture
was exposed to air to initiate further oxidation (Table 2,
entry 1). Unfortunately, no hydroxylated product was detected
after 12 h. Inspired by the recent base-mediated a-hydroxyl-
ation of oxindoles,[18] we briefly examined several inorganic
and organic bases and it was found that the addition of base
was indeed essential to the photocatalytic oxidation step
(Table 2, entries 2–6). NaHCO3 was determined to be the best
choice, affording the desired product 3a in 68% overall yield
(Table 2, entry 2).[19] Notably, further optimization revealed that
this tandem reaction with only 1 mol% catalyst loading and
0.25 equivalents of NaHCO3 could proceed smoothly to deliver
&
&
Chem. Asian J. 2014, 9, 1 – 6
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!