8
12
S. D. Sawant et al. / Tetrahedron Letters 55 (2014) 811–814
Table 1
Catalyst screening for optimization of phenol formation from phenylboronic acid
Entry
Catalyst (10 mol %)
Fe
Solvent
Temp (°C)/time (h)
Phenola (% yield)
1
2
3
4
5
6
3
O
4
THF
THF
THF
THF
THF
(a) THF
rt, >5
rt, >5
rt, >5
rt, >5
rt, >5
rt, >5
NR
NR
NR
NR
NR
05
FeO
FeCl
3
Fe(OH)
Fe(acac)
3
3
a
-Fe
2
O
3
(
(
(
(
b) THF:ACN (1:1)
c) DCM
d) ACN
rt, >5
rt, >5
rt, >5
rt, >5
30
20
25
10
e) H
2
O
7
8
a
a
-Fe
-Fe
2
O
O
3
3
THF
THF
Ambient (solar VIS-light irradiation), 3.5 h
rt (dark), >5 h
>95
Trace
2
a
Yields are based on GC–MS results; rt—room temperature; NR—no reaction.
Table 2
Optimization of catalyst quantity
Entry
Substrate
Catalyst
-Fe
Mol % of catalyst
Isolated yieldsa (%)
1
2
3
4
Phenyl boronic acid
a
2
O
3
0.1
1
10
25
Trace
5
95
>95
a
All yields mentioned are isolated yields after column purification.
OH
acid substrates undergo this transformation smoothly. Based on
the literature reports, we hypothesize here that -Fe may play
a similar role like TiO , a widely used photocatalyst for oxidation
OH
B
Photocatalyst α-Fe
2
O
3
a
2 3
O
OH
2
THF; Solar VIS-light irradiation
reactions, which is reported for experiencing an electron-hole pair
1
2
8
process.
As reported by Brown et.al, organoboranes are readily oxidized
by a wide variety of agents including molecular oxygen. Based on
Scheme 1.
2 3
a-Fe O catalyzed formation of phenol from arylboronic acid.
these reports we propose here that the combination of Fe
also works in a similar manner. First the molecular oxygen will
be adsorbed on Fe and then reacted with boron and forms the
expected product. Based on the literature evidences, we assume
that -Fe behaves as an active photocatalyst as previously re-
2 3 2
O –O
in less time and this was found to be the best condition offering
highest yields amongst all experiments conducted (Table 1, entry
2 3
O
9
7
). Under dark conditions, the reaction gave very poor yields
(Table 1, entry 8), suggesting that the reaction is accelerated by
a
2 3
O
8
,9
light. Further we moved our attention to ascertain the participation
of air oxygen, thus we conducted an experiment using isotope
ported in various studies. Adsorption of a photon with energy
greater than the band gap of the iron oxide leads to the formation
1
8
8
+
À
labeled
desired
O
O
2
as oxygen source. The products observed had given
labeled phenols (Supplementary Fig. 1), which could
of an electron hole pair (hvb /ecb ). The valence band hole of
1
+
2
a-Fe O
3
(hvb ) [E
H
= 2.1 V] acts as a powerful oxidant and the
2
18
À
be confirmed by GC–MS analysis. H
the desired product. The experiments with
2
O experiment does not give
conduction band electron (ecb ) is a relatively poor reductant.
The conduction band electron, ecb , does not react with oxygen,
18
18
À
O
2
and H
2
O sug-
gested that the reaction uses air oxygen for the oxidation process.
Moreover, additional oxygen supply in the reaction has given the
products in very less time, which also supports the involvement
of oxygen in the reaction.
2 3
but it may lead to the dissolution of a-Fe O .
Quantitative oxidation of the organoborane to the correspond-
ing alcohol would presumably involve the uptake of oxygen of
organoborane, followed by hydrolysis of the intermediate boron
derivative. The oxygen adsorbed on iron oxide will attack on phen-
ylboronic acid generating phenyl radical. The first carbon–boron
bond is oxidized very rapidly by a radical-chain process. This initial
oxidation produces peroxide, which may either react with a second
mole of oxygen or may undergo an intermolecular redox reaction.
This ultimately produces phenol and boric acid as side product.
However, in the absence of oxygen or under inert conditions the
reaction does not proceed, indicating the reaction requires air oxy-
gen for oxidation. Moreover, as supporting evidence the reaction
without catalyst has not given the product, this indicates that the
catalyst plays an important role in this oxidation reaction, the
plausible mechanism is given as shown in Scheme 2.
We used different solvents as depicted in Table 1 (entry 6a–e),
to see the best conversion. While tetrahydrofuran has given best
results, reaction in other solvents like dichloromethane, acetoni-
trile, water, or in a mixture of water/tetrahydrofuran (1:1) has
not given good results. The quantity of catalyst was optimized by
conducting reactions using various mol % of catalysts; the best re-
sults were obtained with 10 mol % catalyst. Therefore, after optimi-
zation, all reactions were performed using this condition (Table 1,
entry 7) in tetrahydrofuran. To explore the applicability of the
present method on various substrates, different electron
withdrawing and donating groups containing arylboronic acid
substrates were converted to respective phenols at optimized
7
conditions. The experiment for the used Fe
2
O
3
catalyst was car-
Even though there are many methods described in the literature
1
0
ried out which has given the product with excellent yields (90%),
confirmed by GC–MS (Table 3).
The present protocol described herein for the synthesis of phe-
nol offers excellent yields under mild conditions and therefore this
protocol can be widely applied. Various substituted arylboronic
for the conversion of arylboronic acids into phenols, these meth-
ods involve ligand or base mediated transformations. Our method
offers several advantages over present methods. The established
procedure is very simple and the reactions can be conducted in
an ordinary flask under sunlight using
2 3
a-Fe O as a catalyst and