.
Angewandte
Communications
Table 1: Optimization of the perfluoroalkylation reaction.[a]
electronic similarities with I, in situ generated enolates of type
II were considered as suitable donors.[11] Perfluoroalkyl
iodides 1 were selected as electron-accepting substrates,
because sparse literature precedents[12] qualify them as
potential acceptors for facilitating associations of EDA
complexes. The feasibility of our plan was tested by reacting
ethyl a-cyano phenylacetate (2a) with perfluorohexyl iodide
1a under irradiation by a 23 W CFL bulb (RF indicates the
perfluoroalkyl fragment). The reaction was conducted in
MeCN and in the presence of K2CO3 (3 equiv, Figure 1) so as
to favor the formation of the corresponding enolate. Imme-
diately after mixing with the iodide 1a, the solution developed
a marked yellow-orange color, while its optical absorption
spectrum showed a bathochromic displacement in the visible
spectral region, diagnostic of an EDA complex (Figure 2).
Entry
Base
Solvent
t [h]
Yield
[%][b]
Distribution [%]
4[c]
2[c]
2,4[c]
1
2
3
4
5
6
7
none
Cs2CO3
TMG
Cs2CO3
Cs2CO3
Cs2CO3
TMG
MeCN
MeCN
MeCN
DMF
16
16
16
16
16
16
5
0
70
62
45
22
73
83
61
0
–
60
60
65
66
55
58
63
–
–
33
31
32
31
31
32
31
–
–
7
9
3
3
14
10
6
AcOEt
MeCN/HexF
MeCN/HexF
MeCN/HexF
MeCN/HexF
8[d]
9[e,f,g]
TMG
TMG
2
24
–
[a] Reactions were performed on a 0.1 mmol scale using 3 equiv of 1a
and 2 equiv of base, [2a]0 =0.5m, and a 23 W CFL bulb to illuminate the
reaction vessel. [b] Total yield determined by 1H and 19F NMR analysis
using 1-fluoro-2-nitrobenzene as the internal standard. [c] Percent
distribution of the para- (4), ortho- (2), and ortho,para-functionalized
(2,4) products. [d] Reaction performed using a 300 W xenon bulb,
equipped with a cut-off filter at 385 nm. [e] Reaction in the dark.
[f] Reaction in air. [g] Reaction performed in the presence of 2 equiv of
TEMPO. TMG=1,1,3,3-tetramethylguanidine,
HexF =tetradecafluorohexane.
with a minor amount of the ortho,para-bifunctionalized
adduct (less than 10%). We initially confirmed that the
photochemical activity of the enolate, formed in situ upon
deprotonation of 2a, was essential for reactivity, as in the
absence of a base the starting substrates were completely
recovered (entry 1). Further optimization of the standard
reaction parameters showed that the nature of the base and
the reaction medium were the crucial factors for an efficient
system. When a solution of the substrates and Cs2CO3
(3 equiv) in acetonitrile was irradiated for 16 h, the perfluor-
oalkylated product 3a was isolated in 70% yield (entry 2).
The starting substrate 2a (24%) was recovered even after
prolonged irradiation. The inability of the reaction to
progress to completion was rationalized on the basis of
control experiments, which showed how the product 3a
inhibited the process (the enolate of 3a has a strong
absorption in the visible region, details are reported in
Section D of the Supporting Information). To address this
issue, we added tetradecafluorohexane[13] to the reaction
mixture (in a 1:5 ratio to MeCN) to collect the generated
perfluorinated product 3a in a different phase. The use of the
biphasic system, 1,1,3,3-tetramethyl guanidine (TMG,
2.5 equiv), and a high stirring speed improved the overall
yield while shortening the reaction time (entry 7, 81% total
yield of isolated products, 5 h).
Figure 2. Optical absorption spectra recorded in MeCN in quartz
cuvettes (1 cm path) using a Shimadzu 2401PC UV-visible spectropho-
tometer. [2a]=0.01m, [1a]=0.03m; [TMG]=0.02m. While the sub-
strates 1a and 2a are transparent to light, the resulting enolate IIa
showed a weak absorption at about 360 nm (dashed line): its
combination with perfluorohexyl iodide 1a leads to a strong bath-
ochromic shift (black line).
We recognized the formation of the EDA complex as
critical to reaction development, as visible-light irradiation
might induce an electron transfer, giving the anionic contact
radical pair IV. The presence of iodine as a suitable leaving
group within the radical anion partner may induce a rapid
fragmentation event, productively rendering the iodide anion
along with the desired open-shell species, including the
perfluoroalkyl radical V. In analogy to our alkylation of
aldehydes,[10] we anticipated the formation of the a-carbonyl
perfluroalkylated adduct 4 to be favored, by means of
a radical–radical combination or a radical trap by the enolate.
The arene perfluoroalkylation product 3a (para/ortho formed
in a 2:1 ratio) was generated instead, suggesting an HAS
pathway to be highly preferred under these reaction con-
ditions. A control experiment, carried out by performing the
reaction in the dark, did not provide any reactivity, thus
testifying to the photochemical nature of the transformation.
The unanticipated reactivity prompted us toward further
explorative studies to optimize the reaction conditions
(Table 1). All the experiments, conducted in MeCN using
a 23 W CFL bulb, provided both the para- and ortho-
functionalized products 3a in a roughly constant 2:1 ratio,
Further experiments showed how the careful exclusion of
light completely suppressed the process. In addition, the use
of a 300 W xenon bulb, equipped with a cut-off filter at
385 nm, did not significantly alter the reaction efficiency
(entry 8), indicating that the visible-light-induced photoactiv-
ity of the EDA complex is responsible for the reaction that
occurs.[14] The inhibition of the reactivity observed under
aerobic atmosphere was consistent with a radical mechanism.
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!