.
Angewandte
Communications
Table 1: Initial studies with stoichiometric [Pd(MeCN)2Cl2].[a]
Table 2: Optimization studies.[a]
Entry
n
Solvent
T [8C]
t [h]
10 [%]
11 [%]
1[b]
2[b]
3[c]
3
4
3
THF
THF
1,4-dioxane
60
60
90
20
20
2.5
12
0
32
14
6
19
[a] Reaction conditions: 9 (210–250 mmol), [Pd(MeCN)2Cl2], anhydrous
solvent (2.5 mL), N2 atmosphere. [b] [Pd(MeCN)2Cl2] (1.1 equiv).
[c] [Pd(MeCN)2Cl2] (1.0 equiv).
Entry
Catalyst
[mol%]
Oxidant
(equiv)
Solvent
T
[8C]
t
13
[%]
1
2
3
4
5
10
10
10
10
10
10
10
10
1
14 (1)
15 (1)
15 (2)
15 (4)
15 (2)
15 (2)
15 (2)
16 (2)
15 (2)
15 (2)
15 (2)
dioxane
dioxane
dioxane
dioxane
DMF
DMF
DMF
DMF
DMF
90
90
90
90
80
5 h
25 h
18 h
22 h
16
17
27
23
46
60
54
61
60
82
81
Initial attempts to render this reaction catalytic with
respect to palladium suffered from the susceptibility of the
furan moiety to undergo conjugate addition to the benzoqui-
none (BQ) oxidant used. The use of non-quinone-based
oxidants unfortunately did not lead to substantial product
formation. A solution was found in the use of oxidants 14–16
featuring increased steric bulk around a quinone core.[9]
As the use of furylpyrrole-based starting materials led to
mixtures of products, an alternative cyclization substrate was
sought. In preliminary studies with the indole 12 (Table 2),
only the fully oxidized product 13 was ever observed, thus
indicating that the oxidation of the intermediate dihydro ring
system (analogous to 10) was more facile than in the case of
pyrroles. 2,6-Dimethylbenzoquinone (14) and 3,5-di-tert-
butyl-o-benzoquinone (15) were both found to effect catalytic
turnover of palladium (Table 2, entries 1 and 2); however,
during reactions with 14 a side product was observed that
indicated addition of the furan moiety to the quinone was still
taking place. As the initial cyclized product (analogous to 10)
was undergoing oxidation, it followed that two equivalents of
15 would be required for complete conversion. An increase in
the amount of oxidant used to two equivalents increased the
yield marginally from 17 to 27% but depleted the amount of
starting material recovered (Table 2, entry 3). A further
increase in the amount of oxidant used did not lead to an
additional increase in product yield (Table 2, entry 4).
A substantial increase in productivity was observed when
the solvent was switched to DMF (Table 2, entry 5) and the
temperature was increased to 1208C (entry 6). Other qui-
nones with sufficient bulk (e.g. 16; Table 2, entry 8) gave the
product in comparable yields. However, the key parameter to
the success of the sequence was a short reaction time, typically
under half an hour, at an elevated temperature. The most
consistent results were obtained when the reaction flask
containing substrate, catalyst, oxidant, and solvent was heated
by rapid immersion in a preheated oil bath. Slow heating of
the reaction mixture up to the required temperature resulted
in poorer yields and mass recovery. The reaction concen-
tration and catalyst loading were also significant, and
optimization studies showed that a concentration of 0.02m
and a 5 mol% loading of the catalyst [Pd(MeCN)2Cl2] led
to consistently high-yielding reactions at the short contact
1 h
6
120
120
120
120
120
120
20 min
20 min
25 min
1.5 h
30 min
25 min
7[b]
8
9[c]
10[c,d]
11[c,e]
10
5
DMF
DMF
[a] Reaction conditions, unless otherwise specified: 12 (125 mmol,
0.1m), anhydrous solvent, N2 atmosphere. [b] The reaction was carried
out in an open flask with reagent-grade dimethylformamide (DMF).
[c] The reaction was carried out with 1 mmol of 12. [d] The reaction was
carried out at a 0.01m concentration of 12. [e] The reaction was carried
out at a 0.02m concentration of 12.
times required at 1208C (Table 2, entry 11). It was pleasing to
note that unlike in our previous carboamination studies,
electron-withdrawing substituents were not necessary for
diene functionalization.
Initial exploration of the scope of the reaction under these
optimized conditions focused on varying the substituent on
the indole ring (Table 3, entries 1–9). In general, electron-rich
and electron-neutral systems performed better than electron-
poor systems. The reaction is tolerant of halogen substituents
(Table 3, entries 4 and 5) as well as esters, nitriles, and
aldehydes (entries 6–8). The vinyl-substituted substrate 31
gave the desired product in moderate yield (Table 3, entry 9).
Extension of the alkyl chain to four CH2 units severely
affected the cyclization, which gave only very small quantities
of the Stemona alkaloid related product 34 (Table 3, entry 10).
The application of this methodology to seven-membered-ring
formation is a limitation that is currently under investigation.
Chain branching in the alkyl tether is tolerated (Table 3,
entry 11), although a gem-dimethyl group adjacent to the
reactive diene center greatly impeded reactivity (entry 12).
Pleasingly, this reaction proved general in terms of the
heterocycles present in the substrate, thus enabling rapid
access to a range of novel structures. Thiophene- and pyrrole-
containing products were readily obtained in good yield
(Table 3, entries 13 and 14), and both furan and indole
moieties can be replaced by pyrrole units, as in 43 (entry 15).
One limitation appears to be the use of phenyl rings as the p-
2
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 1 – 6
These are not the final page numbers!