Acid- and iridium-catalyzed tandem 1,3-transposition/3,1-hydrogen shift/chlorination of allylic alcohols

Article abstract of DOI:10.1021/cs501618h

A method for the selective synthesis of α-chlorocarbonyls from allylic alcohols is presented. The reaction occurs through an acid- and iridium-catalyzed tandem process that combines a 1,3-transposition, a 3,1-hydrogen shift, and a chlorination process, and can be applied to a wide range of α-aromatic and heteroaromatic secondary allylic alcohols. Saturated non-chlorinated ketones or other side-products derived from overchlorination were not detected.

Full text of DOI:10.1021/cs501618h

Research Article
pubs.acs.org/acscatalysis
Acid- and Iridium-Catalyzed Tandem 1,3-Transposition/3,1-Hydrogen
Shift/Chlorination of Allylic Alcohols
Ana Vazquez-Romero,†,‡ Antonio Bermejo Gomez,†,‡ and Belen Martín-Matute*,†,‡
́
́
́
‡
†Department of Organic Chemistry and Berzelii Center EXSELENT on Porous Materials, The Arrhenius Laboratory, Stockholm
University, Stockholm, 106 91, Sweden
S
* Supporting Information
ABSTRACT: A method for the selective synthesis of α-chlorocar-
bonyls from allylic alcohols is presented. The reaction occurs through
an acid- and iridium-catalyzed tandem process that combines a 1,3-
transposition, a 3,1-hydrogen shift, and a chlorination process, and
can be applied to a wide range of α-aromatic and heteroaromatic
secondary allylic alcohols. Saturated non-chlorinated ketones or other
side-products derived from overchlorination were not detected.
KEYWORDS: hydrogen transfer, chlorination, allylic alcohols, isomerization, iridium, 1,3-transposition
However, due to the reversibility of the reaction, mixtures of
constitutional isomers were still formed.
INTRODUCTION
■
Molecular rearrangements are important transformations in
organic synthesis that give access to structural isomers of the
original molecules.1 An example that has attracted a lot of
attention in recent years is the transposition of allylic alcohols,
where the positions of the hydroxyl group and the double bond
The design of new methods to drive this transformation
toward a single isomer is still a challenge. A limited number of
strategies that result in highly regioselective Re-catalyzed
transpositions have been reported.2h−j,m,3a−c Initial examples
by Grubbs and co-workers relied on the use of allylic alcohols
that after isomerization gave conjugated alkenes.2h A second
method allowed the thermodynamically disfavored regioisomer,
a primary allylic alcohol, to be obtained through a selective
silylation.2h Two related approaches by Lee were based on
shifting the equilibrium toward the desired direction by
introducing certain functionalities within the molecule able to
stabilize the resulting transposed allylic alcohol. In one case, the
oxophilicity of boron was exploited and a boron−oxygen bond
could be formed only in one of the isomeric allylic alcohols.2i In
the second strategy, a ring contraction was used to control the
regioselectivity.2m Zakarian and co-workers shifted the equili-
brium by using hydroxyl-functionalized allylic alcohols, trapping
the transposed compounds as acetals.2j Other recently reported
examples, also catalyzed by rhenium, involve cascade reactions
where functional groups such as epoxides, enones, or ketals were
used to displace this equilibrium.3a−c
are switched to give isomeric allylic alcohols (Figure 1a).2,3
A
direct application of this rearrangement would be to use it as a
straightforward method for the synthesis of the less accessible
isomer. Several methods for the transposition of allylic alcohols
have been reported.2,3a−c Early examples used very harsh
reaction conditions, high temperatures, and superstoichiometric
quantities of strong acids,2a and gave the rearranged regioisomer
in moderate yields. In some instances, the lower yields could be
partly explained by the intermediacy of allylic cations, highly
reactive species that can lead to the formation of side-products.
Additionally, in such reactions, the conversion stops when
thermodynamic equilibrium is reached, and thus a mixture of the
two regioisomers is usually obtained. The use of transition-metal
catalysts has overcome some of these limitations by promoting
the reaction through a [3,3]-sigmatropic rearrangement. For
instance, good yields and selectivities were obtained with
catalysts based on oxo-vanadium and oxo-molybdenum
complexes, as reported by Chabardes2b and Fujita,2c although
the reactions required high temperatures (150−160 and 200 °C,
respectively). Later, Takai used oxo-metal vanadium and
molybdenum complexes that were able to rearrange allylic
alcohols at room temperature with good selectivity.2d Although
other kinds of catalysts, such as those based on boronic and
benzoic acids, have been reported,2k,l one of the most efficient
methods is the rhenium-catalyzed rearrangement developed by
Osborn:2f readily available trioxorhenium complexes [(Re-
O3(OSiR3)] and rhenium(VII) oxide catalyze the rearrangement
of a wide range of allylic alcohols under mild conditions.2g
A related, yet different, rearrangement of allylic alcohols is
their irreversible transition-metal-catalyzed isomerization into
carbonyl compounds (Figure 1b).4−6 Our research has been
focused on the development of efficient allylic alcohol isomer-
izations and their application in organic synthesis.6,7 Pioneering
works on the isomerization of allylic alcohols in water have been
reported by different groups.5 Gimeno, Cadierno, and co-
Received: October 20, 2014
Revised: December 5, 2014
© XXXX American Chemical Society
708
DOI: 10.1021/cs501618h
ACS Catal. 2015, 5, 708−714

ACS Catalysis
Research Article
Figure 1. (a) 1,3-Transposition of allylic alcohols. (b) Transition-metal-catalyzed redox isomerization. (c) Tandem isomerization (1,3-hydrogen shift)/
halogenation of allylic alcohols catalyzed by Ir(III) complexes (X = F, Cl, or Br). (d) This work: Tandem 1,3-transposition/3,1-hydrogen shift/
chlorination.
a
Table 1. Isomerization of Allylic Alcohols 1a−c by Ru(IV) and Ir(III) Catalysts
b
entry
allylic alcohol
catalyst (mol %)
solvent
H2O
T (°C)
time (h)
conversion (%)
>99
1′:2:2′
c
e
1
1a
1a
1b
1b
1b
1b
1c
1c
1c
[Ru(η3:η3-C10H16)Cl2]2 (0.1)
[Cp*IrCl2]2 (1.0)
[Ru(η3:η3-C10H16)Cl2]2 (2.5)
[Ru(η3:η3-C10H16)Cl2]2 (0.5)
[Cp*IrCl2]2 (1.0)
[Cp*IrCl2]2 (1.0)
[Ru(η3:η3-C10H16)Cl2]2 (0.5)
[Cp*IrCl2]2 (1.0)
75
75
75
75
75
75
75
75
r.t.
1
0.5
1
-:>99:-
-:>99:-
-:>99:-
d
2
THF/H2O
H2O
>99
c
f
f
3
>99
27
4
5
6
7
8
9
H2O
1
76:24:-
-:>99:-
-:>99:-
d
THF/H2O
THF/H2O
H2O
0.5
16
16
16
48
47
d
>99
f
>99
>99:-:-
8:46:46
-:58:42
d
THF/H2O
>99
d
[Cp*IrCl2]2 (1.0)
THF/H2O
90
a
b
Reactions were carried out with 0.4 mmol of allylic alcohol 1a−c (concentration 0.2 M) for 16 h. The ratio 1′:2:2′ was determined by 1H NMR
c
d
e
1
spectroscopy. See ref 5e. Conversions determined by H NMR spectroscopy using 1,2,4,5-tetrachloro-3-nitrobenzene as internal standard. Only
compound 2a was observed. Decomposition was observed (ca. 10%).
f
workers used a very efficient Ru(IV) complex that was able to
achieve impressive turnover numbers (TON = 106) at 75 °C.5e In
2010, we reported a rhodium complex that catalyzes the reaction
in water and at ambient temperature.6b Interestingly, these
isomerizations in water are particularly efficient for allylic
alcohols bearing aliphatic substituents on the alcohol carbon.
For examples with an aromatic substituent, harsher reaction
conditions and considerably higher catalyst loadings had to be
used.5a,d−f,i,j
In the past few years, we have reported the synthesis of α-
haloketones from allylic alcohols in aqueous solvents.7 The
method formally combines an allylic alcohol isomerization (1,3-
hydrogen shift) with a fluorination,7a chlorination,7b or
bromination7c catalyzed by Ir(III) complexes (Figure 1c). The
corresponding α-halocarbonyls were obtained as single constitu-
tional isomers in good to excellent yields. Just as for the
isomerization of allylic alcohols into ketones in water, substrates
bearing aromatic groups at the alcohol carbon required longer
reaction times and/or higher catalyst loadings to achieve high
conversions than those bearing aliphatic substituents.7a,b
Steric effects have been considered to explain the lower
reactivity of α-aryl allylic alcohols compared to that of α-alkyl
ones in the allylic-alcohol-to-carbonyl isomerizations (Figure 1b)
in water.5a,j However, one must also consider that these two
families of substrates are substantially different from an electronic
point of view. In this article, we show that α-aryl allylic alcohols
may undergo cleavage of the C−OH bond when subjected to
transition-metal-catalyzed isomerization into carbonyl com-
pounds in water. This competing process can account for the
lower yields, the longer reaction times, and the higher catalyst
loadings needed for this type of substrates. Furthermore, as
described above, cleavage of the C−OH bond can result in the
formation of a regioisomeric mixture of allylic alcohols. We use
the irreversible transition-metal-catalyzed isomerization/chlori-
nation process as a strategy to control the regioselectivity in the
1,3-transposition of allylic alcohols. Thus, we report the first
tandem 1,3-transposition/3,1-hydrogen shift/chlorination of
allylic alcohols (Figure 1d). This methodology gives access to
certain α-chloroketones from readily available allylic alcohols,
avoiding long and tedious synthetic routes.
RESULTS AND DISCUSSION
■
We started by comparing the reactivity of aliphatic and aromatic
allylic alcohols in their isomerization into carbonyl compounds
catalyzed by [Cp*IrCl2]2, an excellent catalyst for redox
isomerization/halogenation tandem reactions,7 and the highly
active Ru(IV) complex reported by Gimeno (Table 1).5e
The redox isomerization reaction of 1-octen-3-ol (1a) into 3-
octanone (2a) with either [Cp*IrCl2]2 (1 mol %) or [Ru(η3:η3-
C10H16)Cl2]2 (0.1 mol %)5e took place with full conversion and
complete selectivity in less than 1 h at 75 °C (Table 1, entries 1
and 2). In contrast, the isomerization of α-aryl allylic alcohols
709
DOI: 10.1021/cs501618h
ACS Catal. 2015, 5, 708−714

ACS Catalysis
Research Article
required higher catalyst loadings or longer reaction times, and the
outcome of the reaction was dependent on the catalyst used.
Thus, the isomerization of 1b was achieved using the procedure
described in the literature, in which 2.5 mol % of the Ru(IV)
dimer was used5e (Table 1, entry 3 vs entry 1). When we carried
out the same reaction using 0.5 mol % of the Ru(IV) catalyst,
only a 27% conversion was achieved, and a mixture of transposed
allylic alcohol 1′b and ketone 2b in a 76:24 ratio was obtained
after 1 h at 75 °C (Table 1, entry 4). Thus, when lower catalyst
loadings are used, cleavage of the C−O bond clearly competes
with the allylic-alcohol-to-ketone isomerization reaction. It
should be noted that the transposed alcohol 1′b was not further
isomerized into 2′b. The reaction was also carried out using
[Cp*IrCl2]2 (1 mol %) in a THF/H2O mixture (1:1) at 75 °C
(Table 1, entry 5). After 30 min (time required for the
isomerization of 1a, entry 2 in Table 1), the conversion was low,
and ketone 2b was obtained in 27% yield. However, when the
reaction was stirred at 75 °C overnight, full conversion into 2b
was achieved (Table 1, entry 6). For the more substituted
aromatic allylic alcohol 1-phenylbut-2-en-1-ol (1c), using
[Ru(η3:η3-C10H16)Cl2]2 (0.5 mol %) in H2O at 75 °C, afforded
1′c in quantitative yield after 16 h (Table 1, entry 7). Thus, the
transposition of a secondary allylic alcohol into its isomeric
secondary and conjugated allylic alcohol is clearly favored under
these conditions. Interestingly, carbonyl products derived from
redox isomerization were not detected (neither 2c nor 2′c). The
reaction of 1c catalyzed by [Cp*IrCl2]2 (1 mol %) in THF/H2O
(1:1) at 75 °C for 16 h promoted the 1,3-transposition.
Surprisingly, and in contrast with the results obtained with the
Ru(IV) catalyst, led also to the formation of redox isomerization
products from both allylic alcohols 1c and 1′c (Table 1, entry 8).
When the same reaction was carried out at room temperature, a
mixture of redox isomerization products was formed (90%
conversion, 2c:2′c = 58:42), the remainder being unreacted 1c
(Table 1, entry 9). Thus, the 1,3-transposition also occurred at
room temperature when [Cp*IrCl2]2 was used as the catalyst,
but under these milder reaction conditions, a lower conversion
was obtained.
Formation of 1′b-c could be attributed to the intermediacy of
carbocationic species produced upon cleavage of the C−OH
bond. Then, the nucleophilic attack of a molecule of water at
either end of the allylic cation affords the two isomeric allylic
alcohols 1b-c or 1′b-c. These could undergo the transition-
metal-catalyzed redox isomerization process to form the ketones,
2b-c and 2′b-c. In the case of the less substituted α-aryl allylic
alcohol 1b, products derived from this tandem sequence were
not obtained (or were only detected in trace amounts). The
formation of carbocations may also occur with this substrate
(1b), but the secondary carbon bears more of the positive charge
than does the primary carbon, and the nucleophilic attack by
water is faster there. The competing cleavage of the C−OH bond
explains the lower yields, higher catalyst loadings, and longer
reaction times required for α-aryl allylic alcohols compared to
those bearing α-aliphatic substituents in the transition-metal-
catalyzed redox isomerization in aqueous solvents.
chlorination7b reaction with the 1,3-transposition of allylic
alcohols. This tandem process would give access to α-
chlorocarbonyl compounds from allylic alcohols that are
commercially available or that may be synthesized through a
straightforward synthetic route.
In our initial experiments, we investigated the 1,3-trans-
position/3,1-hydrogen shift/chlorination of 1-phenylbut-2-en-1-
ol (1c). When allylic alcohol 1c was treated with 1 mol % of
[Cp*IrCl2]2 in the presence of a slight excess of NCS (1.2 equiv)
at room temperature, a mixture of chloroketones 3c and 3′c
(1:4) was obtained in 96% yield (Scheme 1). To improve the
Scheme 1. Formation of Two α-Chloroketones from 1-
Phenylbut-2-en-1-ol (1c)
selectivity toward 3′c, we first tried to identify the reactant
responsible for the 1,3-transposition (Table 2). Stirring 1c
a
Table 2. 1,3-Transposition of 1c
b
entry
catalyst (mol %)
reagent (equiv)
--
1c:1′c
1
2
3
4
5
6
7
[Cp*IrCl2]2 (1)
[Cp*IrCl2]2 (1)
>99:<1
>99:<1
1:1.4−10
<1:>99
<1:>99
<1:99
Succinimide (1.1)
c
--
--
--
--
--
NCS (1.2)
d
0.1 M HCl
p-TsOH (1)
e
BF3·Et2O (1)
Salicylic acid (1)
46:54
a
Unless otherwise noted, the reactions were carried out with 0.15
mmol of allylic alcohol 1c for 16 h r.t. in a mixture THF/H2O (1:1, 0.2
b
c
1
M). Ratio determined by H NMR spectroscopy. Different batches
of NCS (1.2 equiv) were used. Using a mixture THF/HCl (aq., 0.1
d
e
M). 19% yield of compound 1′c and 72% of unidentified compounds
determined by 1H NMR spectroscopy using 1,2,4,5-tetrachloro-3-
nitrobenzene as internal standard.
overnight in the presence of a catalytic amount of [Cp*IrCl2]2 (1
mol %) at r.t. gave only traces of the transposed allylic alcohol 1′c
(Table 2, entry 1). Additional control experiments were carried
out in the presence of succinimide (1.1 equiv; Table 2, entry 2), a
by-product formed during the chlorination step, and also in the
presence of NCS (1.2 equiv), the latter without the iridium
catalyst (Table 2, entry 3). We found that succinimide did not
promote the 1,3-transposition. In contrast, the reaction did occur
with NCS, although the results were highly dependent on the
batch of NCS used. Since the reaction can be catalyzed by acid,8,9
we concluded that traces of hydrochloric acid9 in the NCS
reagent were responsible for the observed reactivity. When the
reaction was carried out using a mixture of THF and HCl (aq.,
0.1 M) instead of THF/H2O, complete conversion of 1c into 1′c
was achieved (Table 2, entry 4). We also evaluated the effect of
other acidic additives such as p-toluenesulfonic acid (p-TsOH),2e
BF3·Et2O, and salicylic acid.2l In the presence of a stoichiometric
amount of p-TsOH, after 16 h at r.t., 1′c was obtained as a single
Although α-aryl allylic alcohols with a disubstituted double
bond (i.e., 1c) easily underwent a tandem 1,3-transposition/3,1-
hydrogen shift with the iridium complex, the selectivity obtained
was low, which makes this process impractical. We therefore
turned our attention to another reaction process that would have
a significantly different reaction rate for the transposed allylic
alcohol, and would thus improve the selectivity. Thus, we
attempted to combine the tandem 1,3-hydrogen shift/
710
DOI: 10.1021/cs501618h
ACS Catal. 2015, 5, 708−714

ACS Catalysis
Research Article
isomer (Table 2, entry 5). BF3·Et2O afforded only 19% of 1′c
after 16 h, and it also gave decomposition side-products (Table 2,
entry 6). Salicylic acid did not improve the outcome of the
reaction, and afforded a 1:1 mixture of 1c and 1′c (Table 2, entry
7).
Table 3. Optimization of the Tandem 1,3-Transposition/3,1-
Hydrogen Shift/Chlorination of 1c
a
To obtain further insight into the kinetics of the acid-catalyzed
1,3-transposition, we studied the reactions by 1H NMR
spectroscopy. A set of electronically different α-aryl allylic
alcohols [1c (p-H), 1d (p-OMe), and 1e (p-Cl)] was subjected
to the reaction conditions (1 or 0.5 equiv of p-TsOH in D2O/
THF-d8), and the reactions were monitored by 1H NMR
spectroscopy (Figure 2). The 1,3-transposition of allylic alcohol
b
c
d
entry
acid (equiv)
t1 (min) 3c + 3′c (%) ratio of 1′c:3c:3′c
e
f
1
HCl
-
-
-
-
41−56
96
7−25:-:56−75
-:16:84
1:13:86
-:13:87
-:12:88
-:9:91
2
BF3·Et2O (0.1)
Salicylic acid (0.1)
p-TsOH (0.1)
p-TsOH (0.2)
p-TsOH (0.5)
p-TsOH (1)
3
96
4
92
5
93
6
-
-
96
7
96
-:3:97
8
p-TsOH (1)
1
95
-:3:97
9
p-TsOH (1)
10
30
60
60
60
60
93
-:2:98
10
11
12
13
14
p-TsOH (1)
95
-:-:>99
-:-:>99
-:9:91
p-TsOH (1)
95
p-TsOH (0.1)
p-TsOH (0.2)
p-TsOH (0.5)
96
97
-:5:95
>99
-:1:99
a
Unless otherwise noted, 1c (0.54 mmol) and the acid additive were
dissolved in THF/H2O (1:1, total volume 2.7 mL). NCS (0.65 mmol)
and [Cp*IrCl2]2 (1 mol %) were added, and the mixture was stirred
b
for 12 h (full conversion) at r.t. A solution of 1c (0.54 mmol) and p-
TsOH (0.1, 0.2, 0.5, 1 equiv) in THF/H2O (1:1, total volume 2.7 mL)
was stirred at r.t. for t1. Then, NCS (0.65 mmol) and [Cp*IrCl2]2 (1
mol %) were added and the mixture was stirred for 12 h (full
c
conversion) at r.t. Determined by 1H NMR spectroscopy using
d
1,2,4,5-tetrachloro-3-nitrobenzene as internal standard. Determined
e
by 1H NMR spectroscopy. Reactions were carried out on a 0.14
mmol scale in a 1:1 THF/HCl (0.1 M) mixture (total volume 680
μL). 3′c was formed together with polymeric material.
Figure 2. 1,3-Transposition of allylic alcohols 1c−e promoted by p-
TsOH.
f
1c using 1 equiv of p-TsOH was complete after 52 min, whereas a
conversion of only 66% was achieved with 0.5 equiv of p-TsOH
after the same reaction time. For the more electron-rich aryl
substrate 1d, full conversion into 1′d was achieved in less than 2
min using either 1 or 0.5 equiv of p-TsOH. On the other hand,
the transposition rate was much lower for the p-chloro-
substituted alcohol 1e: after 60 min, only 41% or 28% of 1′e
was observed using 1 or 0.5 equiv of p-TsOH, respectively.
Therefore, the 1,3-transposition of α-aryl allylic alcohols
promoted by p-TsOH in aqueous media is favored by the
presence of electron-rich substituents on the aromatic moiety.
Next, we attempted the tandem chlorination process by using
iridium catalysis under acidic conditions. In the presence of HCl
(aq., 0.1 M) (Table 3, entry 1), 1c was consumed to give
chloroketone 3′c as the sole product; however, the yields were
only moderate (41−56%) (see Supporting Information).10
Excellent yields of chloroketones and good selectivities (3c:3′c
= 16:84−10:90) were obtained using 0.1−0.2 equiv of either
salicylic acid,2l p-TsOH,2e or BF3·Et2O (Table 3, entries 2−5 and
Supporting Information Table S2). Further screening revealed
that increasing the number of equivalents of p-TsOH to 0.5 or 1
considerably improved the selectivity (Table 3, entries 6 and 7).
Decreasing or increasing the iridium loading did not improve the
outcome (Supporting Information Table S2). To minimize the
formation of unwanted chloroketone 3c, we allowed 1c to react
with p-TsOH for a given time (t1) (Table 2, entries 8−11 and
Supporting Information Table S2) before adding the remaining
reagents (NCS and the iridium catalyst). The best results were
obtained when t1 = 30 or 60 min, and under these conditions 3′c
was formed in excellent yields (95%) and with excellent
selectivities (>99%) (Table 3, entries 10 and 11). Good to
excellent selectivities and yields were also obtained using catalytic
amounts of p-TsOH (0.1−0.5 equiv) (Table3, entries 12−14).
The scope of the tandem reaction was evaluated (Scheme 2)
under the optimized reaction conditions (Table 3, entry 11)
using p-TsOH (1 equiv) and [Cp*IrCl2]2 (1 mol %) at ambient
temperature. Since the 1,3-transposition is highly dependent on
the electronic properties of the aryl substituent (vide supra,
Figure 2), a t1 of 1 h was used for all the allylic alcohols (Scheme
2). A variety of functional groups on the aromatic rings, including
p-OMe, p-Cl, p-Br, and m-OMe, were well tolerated (Scheme 2,
1c-g). Alkyl groups and disubstituted aromatic substrates also
gave the corresponding α-chloroketones 3′ in good to excellent
yields (Scheme 2, 1h-j). Substrates containing 2-naphthyl (1k)
and 2-thiophenyl (1l) moieties gave moderate yields. For 1k,
decomposition reactions were observed, and 3′k was obtained in
a moderate 44% isolated yield. The reaction of the α-
heteroaromatic allylic alcohol 1l proved to be strongly dependent
on the solvent mixtures used (Supporting Information Table
S3). In a 2:1 THF/H2O mixture, the heteroaromatic α-
chloroketone 3′l was isolated in 63% yield (Scheme 2).
711
DOI: 10.1021/cs501618h
ACS Catal. 2015, 5, 708−714

ACS Catalysis
Research Article
a
Scheme 2. Scope: 1,3-Transposition/Redox Isomerization/Chlorination of α-(Hetero)Aromatic Allylic Alcohols
a
Unless otherwise noted: reactions were carried out with 1 mmol of allylic alcohol in THF/H2O (1:1, concentration 0.2 M) with p-TsOH (1 equiv)
at r.t. After 60 min, NCS (1.2 equiv) and [Cp*IrCl2]2 (1 mol %) were added at r.t, and the mixtures were stirred for 12 h (full conversion). Yields of
1
isolated products are shown (in parentheses yields determined by H NMR spectroscopy using 1,2,4,5-tetrachloro-3-nitrobenzene as internal
b
c
standard). Reactions carried out with 10 mol % of p-TsOH. 35% of transposed allylic alcohol 1e′ was remaining and decomposition was observed.
d
e
f
g
A mixture THF/H2O (2:1) was used. 3 equiv of BF3·Et2O was used. 3 equiv of p-TsOH was used. 2 equiv of p-TsOH was used.
Substrates 1m and 1n were synthesized with acetal protecting
aromatic core, was transformed selectively into 3′o in high yield
(Scheme 2) as a result of a 1,3-transposition/3,1-hydrogen shift/
chlorination on the more substituted allylic alcohol and a 1,3-
hydrogen shift/chlorination on the less substituted one.
Substrate 1p, symmetrically substituted with two identical allylic
alcohol moieties, underwent a clean double tandem chlorination
involving two 1,3-transpositions on the allylic alcohol moieties
using 2 equiv of p-TsOH to give bis(α-chloroketone) 3′p in high
yield (Scheme 2). For substrates with longer-chain alkyl
substituents on the double bond of the allylic alcohol (1q-r),
very good results were obtained, and the corresponding
chloroketones 3′q and 3′r were both isolated in 88% yield.
Importantly, saturated ketones or other side-products derived
from overchlorination (i.e., chlorination at the benzylic or
aromatic positions) were not detected for any of the substrates,
and neither were chlorinated products derived from the direct
groups, in order to avoid undesired reactions on the ketone or
aldehyde functional groups. As the deprotection of acetals is
usually accomplished in aqueous acidic media, we tried to
combine it with the tandem chlorination reaction. These two
substrates did not give good results under the optimized reaction
conditions, and variable amounts of the corresponding α,β-
unsaturated ketones were detected. Different reaction conditions
were therefore tested, including higher loadings of p-TsOH,
different acidic additives, and THF/H2O mixtures (Supporting
Information Table S3). The best result for 1m was obtained
using 3 equiv of BF3·Et2O, which gave 3′m as a single
regioisomer in 68% isolated yield (Scheme 2). For the protected
aldehyde 1n, the best result was obtained using 3 equiv of p-
TsOH (Scheme 2), which gave 3′n in 45% isolated yield.
Substrate 1o, which contains two types of allylic alcohols in the
712
DOI: 10.1021/cs501618h
ACS Catal. 2015, 5, 708−714

ACS Catalysis
Research Article
redox isomerization/chlorination pathway for the allylic alcohols
moieties (3) having a disubstituted double bond.
ACKNOWLEDGMENTS
■
This project was supported by the Knut and Alice Wallenberg
Foundation, the Swedish Research Council (VR) and the
Swedish Governmental Agency for Innovation Systems
(VINNOVA) through the Berzelii Center EXSELENT on
Porous Materials. B.M.-M. was supported by VINNOVA
through a VINNMER grant. A.V.-R. thanks the Wenner-Gren
Foundation for a postdoctoral grant.
The selective tandem reaction could also be achieved using
only a catalytic amount of p-TsOH, but the reactivity was
strongly influenced by the electronic properties of the α-aryl
group of the allylic alcohol, as noted above (Scheme 2, 1c-e).
Using 0.1 equiv of p-TsOH and the iridium catalyst, allylic
alcohol 1e, containing a Cl substituent at the para position of the
aromatic ring, reacted slowly (Scheme 2, 1e), and after 16 h, a
small amount of the desired α-chloroketone 3′e was observed
(13%) along with the corresponding transposed allylic alcohol
1′e in 35% yield. In contrast, when an electron-donating group
(p-OMe) was present in the aromatic ring (Scheme 2, 1d), the
corresponding α-chloroketone 3′d was obtained in quantitative
yield (99%) after 16 h using 0.1 equiv of p-TsOH. For the
substrate with an unsubstituted phenyl ring (Scheme 2, 1c), α-
chloroketone 3′c was obtained in high yield (91%) when using
catalytic amount of p-TsOH.
The versatility of this method therefore gives access to variety
of α-chlorocarbonyls in good to excellent yields. These
compounds can be transformed into functionalized molecules
through cross-coupling11 and substitution reactions.12 Very
commonly, they are used as building blocks in the synthesis of
heterocyclic compounds, which are important scaffolds in
medicinal chemistry.7b,13,14
REFERENCES
■
(1) Reviews of pericyclic and sigmatropic rearrangements: (a) Varela,
J. A.; Saa,
́
C. Chem.Eur. J. 2006, 12, 6450−6456. (b) Li, C. J. Chem.
Rev. 2005, 105, 3095−3165. (c) Lutz, R. P. Chem. Rev. 1984, 84, 205−
247. (d) Ziegler, F. E. Chem. Rev. 1988, 88, 1423−1452. (e) Enders, D.;
Knopp, M.; Schiffers, R. Tetrahedron: Asymmetry 1996, 7, 1847−1882.
(f) Ito, H.; Taguchi, T. Chem. Soc. Rev. 1999, 28, 43−50. (g) Chai, Y.;
Hong, S. P.; Lindsay, H. A.; McFarland, C.; McIntosh, M. C. Tetrahedron
2002, 58, 2905−2928. (h) Martín Castro, A. M. Chem. Rev. 2004, 104,
2939−3002. (i) Lee, E. E.; Batey, R. A. J. Am. Chem. Soc. 2005, 127,
14887−14893.
(2) Examples of methods for the 1,3-transposition of allylic alcohols:
(a) Letourneux, Y.; Lo, M. M. L.; Chaudhuri, N.; Gut, M. J. Org. Chem.
1975, 40, 516−518. (b) Chabardes, P.; Kuntz, E.; Vargnat, J.
Tetrahedron 1977, 33, 1775−1775. (c) Hosogai, T.; Fujita, Y.;
Ninagawa, Y.; Nishida, T. Chem. Lett. 1982, 11, 357−360.
(d) Matsubara, S.; Okazoe, T.; Oshima, K.; Takai, K.; Nozaki, H. Bull.
Chem. Soc. Jpn. 1985, 58, 844−849. (e) Narasaka, K.; Kusama, H.;
Hayashi, Y. Tetrahedron 1992, 48, 2059−2068. (f) Bellemin-Laponnaz,
S.; Gisie, H.; Le Ny, J. P.; Osborn, J. A. Angew. Chem., Int. Ed. 1997, 36,
976−978. (g) Jacob, J.; Espenson, J. H.; Jensen, J. H.; Gordon, M. S.
Organometallics 1998, 17, 1835−1840. (h) Morrill, C.; Grubbs, R. H. J.
Am. Chem. Soc. 2005, 127, 2842−2843. (i) Hansen, E. C.; Lee, D. J. Am.
Chem. Soc. 2006, 128, 8142−8143. (j) Herrmann, A.; Saito, T.; Stivala,
C.-E.; Tom, J.; Zakarian, A. J. Am. Chem. Soc. 2010, 132, 5962−5963.
(k) Zheng, H.; Lejkowski, M.; Hall, G. H. Chem. Sci. 2011, 2, 1305−
1310. (l) McCubbin, J. A.; Voth, S.; Krokhin, O. V. J. Org. Chem. 2011,
76, 8537−8542. (m) Volchkov, I.; Park, S.; Lee, D. Org. Lett. 2011, 12,
3530−3533. (n) Li, P.-F.; Wang, H.-L.; Qu, J. J. Org. Chem. 2014, 79,
3955−3962. For recent reviews, see: (o) Bellemin-Laponnaz, S.; Le Ny,
J. P. C. R. Chimie 2002, 5, 217−224. (p) Volchkov, I.; Lee, D. Chem. Soc.
Rev. 2014, 43, 4381−4394.
CONCLUSIONS
■
A mild and efficient methodology based on a tandem 1,3-
transposition/3,1-hydrogen shift/chlorination of α-aryl secon-
dary allylic alcohols to synthesize a wide range of α-
chloroketones has been reported. The reactions are run under
an atmosphere of air, using low loadings of iridium(III), at room
temperature, and in a THF/H2O mixture (1:1). α-Chloroke-
tones with different aryl substituents have been isolated in high
yields as single constitutional isomers without any evidence of
the formation of saturated ketones derived either from redox
isomerization of the starting allylic alcohols or from a 1,3-
transposition/redox isomerization process without chlorination.
Although the general procedure requires a stoichiometric
amount of inexpensive and readily available p-TsOH, catalytic
amounts of this acid are able to promote the selective
transformation for several substrates. Diverse electron-donating
and electron-withdrawing groups as well as reactive function-
alities have been shown to be compatible with the reaction
conditions. This simple and practical method appears to be an
attractive and powerful way to synthesize a very important and
common type of synthetic intermediates widely used in organic
synthesis.
(3) Recent applications of allylic transpositions to the synthesis of
complex natural products: (a) Xie, Y.; Floreancig, P. E. Chem. Sci. 2011,
2, 2423−2427. (b) Xie, Y.; Floreancig, P. E. Angew. Chem., Int. Ed. 2013,
52, 625−628. (c) Xie, Y.; Floreancig, P. E. Angew. Chem., Int. Ed. 2014,
53, 4926−4929. Some applications of molecular rearrangement in the
construction of complex molecules: (d) Ilardi, E. A.; Stivala, C. E.;
Zakarian, A. Chem. Soc. Rev. 2009, 38, 3133−3148. (e) Overman, L. E.
Tetrahedron 2009, 65, 6432−6446.
(4) Selected reviews on the scope and mechanism of the isomerization
of allylic alcohols: (a) Van der Drift, R. C.; Bouwman, E.; Drent, E. J.
Organomet. Chem. 2002, 650, 1−24. (b) Cadierno, V.; Crochet, P.;
Gimeno, J. Synlett 2008, 8, 1105−1124. (c) Mantilli, L.; Mazet, C. Chem.
Lett. 2011, 40, 341−344. (d) Lorenzo-Luis, P.; Romerosa, A.; Serrano-
Ruiz, M. ACS Catal. 2012, 2, 1079−1086. (e) Ahlsten, N.; Bartoszewicz,
A.; Martín-Matute, B. Dalton Trans. 2012, 41, 1660−1670.
ASSOCIATED CONTENT
■
S
* Supporting Information
The following file is available free of charge on the ACS
(5) (a) Cadierno, V.; García-Garrido, S. E.; Gimeno, J. Chem. Commun.
Publications website at DOI: 10.1021/cs501618h.
2004, 2, 232−233. (b) Csabai, P.; Joo,
5640−5643. (c) Fekete, M.; Joo, F. Catal. Commun. 2006, 7, 783−786.
(d) Crochet, P.; Díez, J.; Fernandez-Zumel, M. A.; Gimeno, J. Adv.
́
F. Organometallics 2004, 23,
́
Experimental details, characterization data, and NMR
spectra of organic compounds (PDF)
́
́
Synth. Catal. 2006, 348, 93−100. (e) Cadierno, V.; García-Garrido, S. E.;
́
Gimeno, J.; Varela-Alvarez, A.; Sordo, J. A. J. Am. Chem. Soc. 2006, 128,
AUTHOR INFORMATION
1360−1370. (f) Cadierno, V.; Francos, J.; Gimeno, J.; Nebra, N. Chem.
■
Commun. 2007, 24, 2536−2538. (g) Campos-Malpartida, T.; Fekete,
Corresponding Author
*E-mail: belen@organ.su.se.
́ ́ ́
M.; Joo, F.; Katho, A.; Romerosa, A.; Saoud, M.; Wojtkow, W. J.
Organomet. Chem. 2008, 693, 468−474. (h) Gonzal
́
ez, B.; Lorenzo-Luis,
́
̋
Notes
P.; Serrano-Ruiz, M.; Papp, E.; Fekete, M.; Csep
́
ke, K.; Osz, K.; Katho,
́
́
The authors declare no competing financial interest.
A.; Joo,
́
F.; Romerosa, A. J. Mol. Catal. A: Chem. 2010, 326, 15−20.
713
DOI: 10.1021/cs501618h
ACS Catal. 2015, 5, 708−714

ACS Catalysis
Research Article
(i) Díez, J.; Gimeno, J.; Lledos
2012, 2, 2087−2099. (j) García-Alvarez, R.; Suar
Crochet, P.; Cadierno, V.; Antinolo, A.; Fernandez-Galan
Hermosilla, F. Organometallics 2012, 31, 8301−8311. (k) Voronova, K.;
Purgel, M.; Udvardy, A.; Benyei, A. C.; Katho, A.; Joo, F. Organometallics
2013, 32, 4391−4401.
(6) (a) Martín-Matute, B.; Bogar
́
, A.; Suar
́
ez, F. J.; Vicent, C. ACS Catal.
ez, F. J.; Díez, J.;
, R.; Carrillo-
́
́
́
́
̃
́
́
́
́
, K.; Edin, M.; Kaynak, F. B.; Backvall,
̈
J.-E. Chem.Eur. J. 2005, 11, 5832−5842. (b) Ahlsten, N.; Lundberg,
H.; Martín-Matute, B. Green Chem. 2010, 12, 1628−1633. (c) Bermejo
Gomez, A.; Holmberg, P.; Backvall, J.-E.; Martín-Matute, B. RSC Adv.
́
̈
2014, 4, 39519−39522.
(7) (a) Ahlsten, N.; Martín-Matute, B. Chem. Commun. 2011, 47,
8331−8333. (b) Ahlsten, N.; Bermejo Gom
Angew. Chem., Int. Ed. 2013, 52, 6273−6276. (c) Bermejo Gom
Erbing, E.; Batuecas, M.; Vazquez-Romero, A.; Martín-Matute, B.
́
ez, A.; Martín-Matute, B.
́
ez, A.;
́
Chem.Eur. J. 2014, 20, 10703−10709.
̌
(8) Leleti, R. R.; Hu, B.; Prashad, M.; Repic, O. Tetrahedron Lett. 2007,
48, 8505−8507.
(9) Lan, H.; Hsuan, Y.; Kaufman, R. J. U. S. Patent Appl. PCT/US03/
15839, 2003.
(10) For other attempts in acidic media, see Supporting Information.
(11) (a) Malosh, C. F.; Ready, J. M. J. Am. Chem. Soc. 2004, 126,
10240−10241. (b) Liu, C.; He, C.; Shi, W.; Chen, M.; Lei, A. Org. Lett.
2007, 9, 5601−5604. (c) Liu, C.; Deng, Y.; Wang, J.; Yang, Y.; Tang, S.;
Lei, A. Angew. Chem., Int. Ed. 2011, 50, 7337−7341.
(12) De Kimpe, N.; Verhe, R. The Chemistry of α-Haloketones, α-
́
Haloaldehydes and α-haloimines; John Wiley & Sons: New York, 1988.
(13) Erian, A.; Sherif, S.; Gaber, H. Molecules 2003, 8, 793.
(14) Bermejo Gom
Matute, B. Org. Synth. 2014, 91, 185−200.
́
ez, A.; Ahlsten, N.; Platero-Prats, A. E.; Martín-
714
DOI: 10.1021/cs501618h
ACS Catal. 2015, 5, 708−714