Practical synthesis of β-oxo benzo[d]thiazolyl sulfones: Scope and limitations

Article abstract of DOI:10.1039/c1ob06510f

In this paper, we discuss our new synthetic approach towards functionalized benzo[d]thiazolyl (BT) sulfones, based on the reunion of alkyl BT sulfones and various electrophiles (e.g. R-CO-X, RO-CO-X, RS-CO-X, Ts-X...). All important aspects of this coupling reaction, including relevant and undesirable side reactions, are evaluated by means of calculations and competitive experiments. The scope and limitations of this method are established. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c1ob06510f

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2012, 10, 1225
www.rsc.org/obc
PAPER
Practical synthesis of b-oxo benzo[d]thiazolyl sulfones: Scope and limitations†
Jirˇ´ı Posp´ısˇil,* Raphae¨l Robiette, Hitoshi Sato‡ and Kevin Debrus
Received 2nd September 2011, Accepted 2nd November 2011
DOI: 10.1039/c1ob06510f
In this paper, we discuss our new synthetic approach towards functionalized benzo[d]thiazolyl (BT)
sulfones, based on the reunion of alkyl BT sulfones and various electrophiles (e.g. R–CO–X,
RO–CO–X, RS–CO–X, Ts–X . . . ). All important aspects of this coupling reaction, including relevant
and undesirable side reactions, are evaluated by means of calculations and competitive experiments.
The scope and limitations of this method are established.
Introduction
Over the past 5 years, the use of b-carbonyl heteroaryl sulfones in
organic synthesis, particularly in the domain of organocatalysis,
has dramatically increased.1 Especially, the use of these type of
compounds as nucleophiles in the context of organocatalysis
led to the synthesis of new classes of previously inaccessible
compounds and structural motives. For example, the use of b-oxo
benzo[d]thiazolyl sulfones or b-oxo phenyltetrazolyl sulfones in
combination with a,b-unsaturated aldehydes and ketones has led,
in the presence of a prolinol-based catalyst, to the stereoselective
formation of both b-alkynylated and b-alkenylated carbonyl
derivatives.1i The drawback of this method was the long and
expensive synthesis of the starting material, b-oxo heteroaryl
sulfones.2 Additionally, only a small library, from a structural
diversity point of view, of b-oxo heteroaryl sulfones could be
prepared using the standard synthetic protocols.
Scheme 1 Previous approaches to b-oxo heteroaryl sulfones.
approach to sulfones 4 was not only inefficient but also not general
enough for our purposes.
In this context, we have recently reported a practical synthesis
of b-acyl and b-alkoxycarbonyl heterocyclic sulfones 4 based on
the pairing of sulfone 5 with electrophile 6 (Scheme 2).4
In general, this class of compounds is prepared in two steps,
starting from a-bromo carbonyl compounds and corresponding
heteroaryl sulfides (Scheme 1, eq. 1). To achieve a reasonable
conversion of starting material to product 4, prolonged reaction
times are required (days). Unfortunately, sulfones 4 often slowly
degrade under the given reaction conditions. This drawback
was recently solved by Jørgensen et al.,3 by introducing a new
oxidation protocol of b-oxo hetoroaryl sulfides 3 to sulfones 4.
This procedure allows the isolation of the desired products, not
only in excellent yields but also in very short reaction times (10–45
min) (Scheme 1, eq. 2).
Scheme 2 An alternative approach to b-oxo heteroaryl sulfones 4.
Recently, we started several synthetic ventures based on the use
of b-oxo benzo[d]thiazolyl sulfones 4 and we were searching for
a more versatile access to this class of compounds. The standard
Herein, we wish to present a full report of our experimental
and computational studies, which focused on the evaluation of
the reaction between benzo[d]thiazolyl (BT) sulfone a-anions
and various electrophiles. Firstly, the behaviour of BT sulfones
and targeted b-oxo BT sulfones under non-nucleophilic basic
conditions is discussed. Based on these results, the reactivity
of a BT sulfonyl anions with various electrophiles is evaluated,
with a special mention to carbonyl electrophiles.5 The scope and
limitations of these coupling reactions are also discussed.
Institute of Condensed Matter and Nanosciences, Universite´ catholique de
Louvain, Place Louis Pasteur 1 box L4.01.02, B-1348, Louvain-la-Neuve,
Belgium. E-mail: jiri.pospisil@uclouvain.be; Fax: +32 (0)10 47 29 19;
Tel: +32 (0)10 47 29 19
† Electronic supplementary information (ESI) available: detailed ex-
perimental and computational procedures and data. See DOI:
10.1039/c1ob06510f
‡ Visiting researcher from Tohoku University, Japan.
This journal is
The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1225–1234 | 1225
©

View Article Online
Table 1 Dimerization of sulfone 5a under basic conditionsa
Yield (%)c
Entry
Base (equiv)
Time at -78 ◦C (min)
Conv. of 5ab (%)
5a–D
7
7–D
1d
2g
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
LDA (1.1)
LDA (1.1)
120
120
120
120
120
120
0.5
2
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
n.a.
n.a.
<5e
<5e
<5e
<5e
97
91
85
61
48
52
61
53f
n.a.
n.a.
LiN(TMS)2 (1.1)
LiN(TMS)2 (4.0)
KN(TMS)2 (1.1)
KN(TMS)2 (4.0)
LiN(TMS)2 (5.0)
LiN(TMS)2 (5.0)
LiN(TMS)2 (5.0)
LiN(TMS)2 (5.0)
LiN(TMS)2 (5.0)
LiN(TMS)2 (5.0)
KN(TMS)2 (5.0)
KN(TMS)2 (5.0)
KN(TMS)2 (5.0)
KN(TMS)2 (5.0)
KN(TMS)2 (5.0)
KN(TMS)2 (5.0)
<5e
82f
49
<5e
<5e
<5e
<5e
<5e
<5e
<5e
<5e
<5e
<5e
<5e
<5e
<5e
91
<5e
<5e
7
16
24
31
2
20
31
63
81
80
5
15
30
60
0.5
2
21
90
76
53
5
15
30
60
26
<5e
<5e
a Conditions: (i) base, THF, -78 ◦C, (ii) stirred at -78 ◦C for given time, (iii) DCl in MeOD added. b Based on 1H NMR spectra of crude reaction
mixure. c Isolated yield. d Taken from lit9 e No traces of product observed (1H NMR) in crude reaction mixture. f Mixture of mono- and di-deuterated and
non-deuterated products 7-D and 7, respectively. g The reaction was quenched using HCl in MeOH.
1
D could be detected by H NMR in the crude mixture. Only the
Results and discussion
dimers of 5a, compounds 7 and/or 7-D were observed.
Preliminary studies
To our great surprise, the self-condensation of 5a to 7 occurred
even if a large excess of base was used (Table 1, entries 4 and 6).
This observation suggested that either (a) deprotonation of 5a is
not quantitative and the dimerization proceeds via the addition
of anion 5a-Li to non-deprotonated sulfone 5a, or (b) the anion
5a-Li reacts with itself (Scheme 3).
First, we evaluated the reliability of our synthetic approach to b-
oxo BT sulfones, 4, based on the addition of the sulfone 5 a anion
onto the electrophile, 6. From the literature it is known that sulfone
5 undergoes rapid self-condensation under basic conditions (Table
1, entry 1).6 Of course, if this self-condensation reaction was too
fast it would compete with the addition to electrophile 6 and
our newly designed approach to the synthesis of 4 would be
compromised.
To evaluate this risk, we first reproduced the literature results
and reacted sulfone 5a with a slight excess of LDA (Table 1,
entry 2). Similar to the literature, we obtained the product of
self-condensation 7 in 61% yield.7 Next, we decided to use the less
nucleophilic [M]N(TMS)2 bases that are successfully used in the
Sylvestre Julia olefination reaction to deprotonate BT sulfone 5a,7
even though the pKa values of 5a and HN(TMS)2 (conjugate acid)
are presumably very close.8 These pKa values, of course, raised the
question whether the deprotonation of sulfone 5a by [M]N(TMS)2
base is quantitative. To answer the question, we decided to perform
series of experiments in which sulfone 5a would react with either
a slight (1.1 equiv) or a large excess (4.0 equiv) of LiN(TMS)2 and
KN(TMS)2 (Table 1, entries 3–6).
Scheme 3 Self-condensation reaction of sulfone 5a.
To determine which of these hypotheses is correct, sulfone 5a
was reacted with a large excess (5.0 equiv) of LiN(TMS)2 and the
conversion of 5a and the formation of its dimer 7 was monitored
(Table 1, entries 4 and 7 - 12).
These experiments showed that, in the presence of an excess
of the base, the deprotonation of the acidic hydrogen a to a
sulfone group is very fast (less than 30 s, Table 1, entry 7)9 and
the concentration of anion 5a-Li is slowly decreasing with time.
Correspondingly, the formation of dimer 7-D is increasing.
In all cases, complete conversion of sulfone 5a was observed
after 2 h. Interestingly, neither unreacted sulfone 5a nor, after the
reaction work-up with DCl/MeOD, its deuterated equivalent 5a-
1226 | Org. Biomol. Chem., 2012, 10, 1225–1234
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
Table 2 Computed charges (NBO) at C-1 and C-2 carbons in sulfone 5a
and its derivativesa
Entry
R
[M]
C-1
C-2
1
2
3
H
H
H
Li
Li
-0.211
-0.191
-0.184
n.d.
-1.061
-0.802
Scheme 4 Possible self-condensation of keto sulfone 4a.
To experimentally evaluate this information, keto sulfone 4a
was first stirred with an excess of LiN(TMS)2 (Scheme 5, eq. 1).
No degradation of 4a was observed. Once the stability of 4a under
basic conditions was determined, the resistance of enolate 4a-Li
against the attack of anion 5a-Li was tested (Scheme 5, eq. 2).
Again, no reaction of 4a-Li was observed.
4
5
Li
Li
-0.185
-0.180
-0.844
-0.856
6
Li
-0.194
-1.140
a For details, see ESI.†
These results indicate that deprotonation of sulfone 5a is rapid
(presumably due to high kinetic acidity of the hydrogens a to
sulfone group) and that dimerization of sulfone 5a proceeds, even
if this latter is fully transformed into its anion 5a-Li. This suggests
thus that anion 5a-Li reacts with itself.
Similar behaviour was observed if an excess of KN(TMS)2 was
used (Table 1, entries 6 and 13–18). However, in this case we
observed that newly generated anion 5a-K is dimerizing even faster
than 5a-Li, substantially diminishing the concentration of anion
5a-K in the reaction mixture. We believe that faster dimerization
process of anion 5a-K, as compared to anion 5a-Li, is caused by
the higher nucleophilicity of the potassium anion 5a-K.
Scheme 5 Determination of keto sulfone 4a stability in the presence of
base and a sulfonyl anion.
These experiments did show us that self-condensation of sulfone
5a is rather rapid (full conversion within 2 h), whereas keto
sulfones 4 (the desired class of compounds) are stable under basic
conditions and do not react with external nucleophiles, such as
anion 5a-Li.
To shed more light into the reactivity of anion 5a-Li and to
evaluate the possibility that anion 5a-Li might also behave as an
electrophile under the reaction conditions, we decided to evaluate
and compare the electrophilicity of the BT group in 5a and 5a-
Li by computational means.10 Relative electrophilicity of the BT
group has been estimated by computing the charge on the C-1
carbon atom (Table 2).11 The obtained partial charges on C-1
for 5a (-0.211) and 5a-Li (-0.191) were very similar suggesting
that both species should have essentially the same electrophilic
properties.12 These results thus support our hypothesis that dimer
7 is formed by the self-reaction of 5a-Li.
Having described the behaviour of sulfone 5a in the presence of
the base, we focused on the evaluation of the stability of the b-oxo
BT sulfone 4. As a model substrate, we used keto sulfone 4a.13
We feared that compound 4a might undergo a self-condensation
reaction similarly to sulfone 5a (Scheme 4). Indeed, according to
the corresponding computed charges, the C-1 carbon atom of the
enolate of 4b should be (at least) as electrophilic as the one in 5a
and 5a-Li (Table 2, entry 3).
Synthesis of b-keto sulfones 4
Having collected these results, we were able to investigate the
coupling between sulfone 5a and benzoyl chloride 6a (Table 3).
Our primary goal was to find suitable reaction conditions for
this coupling (Scheme 6). We thus had to design a system in which
the addition of a sulfonyl anion of 5a-Li14 to electrophile (6a)
would be faster than its self-condensation to adduct 7 (k1
ꢀ
k2). Additionally, the transformation of the adduct, 4a, into its
enolate, 4a-Li, should be more rapid than the eventual Sylvestre
Julia reaction that could occur between the ketone presented in
4a and sulfone anion 5a-Li (k5ꢀk6). Once all of adduct 4a is
transformed into its enolate, 4a-Li, it should be safe, since we
demonstrated that enolate 4a-Li is stable in the presence of a base
or nucleophile (Scheme 5).
Taking into account these considerations, we started to inves-
tigate the coupling between sulfone 5a and benzoyl chloride 6a.
Initially, generated anion 5a-Li was stirred at -78 C for 15 min
prior to the benzoyl chloride 6a addition (Table 3, entry 1). The
desired product 4a was isolated in 15% yield along with 32% of
the dimerization product 7. As expected, if the pre-metallation
period was longer (Table 3, entry 2), the self-condensation was
the predominant reaction and only trace amounts of products
were formed. On the other hand, the progressive decrease of
◦
On the other hand, lithiated keto sulfone 4a-Li should be
less nucleophilic than sulfone anion 5a-Li due to additional
negative charge stabilization by the carbonyl group (enolate). This
expectation is supported by our computational results (see charges
on C-2, Table 2).
This journal is
The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1225–1234 | 1227
©

View Article Online
Table 3 Optimization of reaction conditions
Yield (%)a,b
Entry
X
Conditions
4a
7c
1
2
3
4
5
6
7
8
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
F
Br
OCOPh
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, 15 min then BzCl (1.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, 30 min then BzCl (1.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, 1 min then BzCl (1.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, then BzCl (1.2 equiv)
15
7
32
48
7
<5
<5
11
17
11
9
12
13
<5
5
<5
<5
<5
<5
65
96
97
82
72
74
78
64
52
95
88
93
97
92
92
BzCl (1.2 equiv), -78 ◦C, THF, then LiN(TMS)2 (2.2 equiv)
NaN(TMS)2 (2.2 equiv), -78 ◦C, THF, then BzCl (1.2 equiv)
KN(TMS)2 (2.2 equiv), -78 ◦C, THF, then BzCl (1.2 equiv)
LiN(TMS)2 (2.2 equiv), 12-crown-4 (4.5 equiv), -78 ◦C, THF, then BzCl (1.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF/HMPA = 6 : 1 (v/v), then BzCl (1.2 equiv)
KN(TMS)2 (2.2 equiv), 18-crown-6 (4.5 equiv), -78 ◦C, THF, then BzCl (1.2 equiv)
KN(TMS)2 (2.2 equiv), -78 ◦C, THF/TDA-1 = 6 : 1 (v/v), then BzCl (1.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, Et2O, then BzCl (1.2 equiv)
9
10
11
12
13
14
15
16
17
LiN(TMS)2 (2.2 equiv), -78 ◦C, DME, then BzCl (1.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, then BzF (1.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, then BzBr (1.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, then Bz2O (1.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, then Bz-Im (1.2 equiv)
18
19
20
CN
OMe
OMe
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, then BzCN (1.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, then BzOMe (1.2 equiv)
BzOMe (1.2 equiv), -78 ◦C, THF, then LiN(TMS)2 (2.2 equiv)
91
<5
<5
<5
63
71
a Overall yields refer to pure, isolated products. b Average of two runs. c Yield recalculated to 100%.
Scheme 6 Considered undesired side reactions that could compete with the b-keto sulfone 4 synthetic pathway (red).
the metallation period led to an increased yield of adduct 4a.
Finally, the highest yield of 4a was obtained when electrophile
6a was added immediately after the base (Table 3, entry 4).15 We
thus tested also the coupling step under the Barbier-type reaction
conditions.16 Not surprisingly, the desired product 4a was isolated
in excellent 95% yield (Table 3, entry 5).
At this stage, we decided not to use the Barbier-type reaction
conditions as our standard reaction protocol since we thought
that they might be applicable only in the case of acylating agents
that possess no a hydrogen atoms. Indeed, we expected that if, for
example, acetyl chloride 6b was used as electrophile, several side
reactions might occur. Some of them are depicted in Scheme 7.
1228 | Org. Biomol. Chem., 2012, 10, 1225–1234
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
Table 4 Barbier-type reaction conditions evaluation
Yield (%)a
Entry
R
Conditions
4
7b
1
2
3
CH3
CH3
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, then RCOCl (1.2 equiv)
RCOCl (1.2 equiv), -78 ◦C, THF, then LiN(TMS)2 (2.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, then RCOCl (1.2 equiv)
4b, 90
4b, 67
4h, 83
<5
14
<5
4
RCOCl (1.2 equiv), -78 ◦C, THF, then LiN(TMS)2 (2.2 equiv)
4h, 54
21
a Overall yields refers to pure, isolated products. b Yield recalculated to 100%.
4c were formed in lower yield than if Barbier-type conditions were
used (Table 4, entries 1 vs. 2, and 3 vs. 4).
Next, NaN(TMS)2 and KN(TMS)2 were tested as bases but,
in both cases, lower reaction yields of 4a were observed (Table 3,
entries 6 and 7). This observation is in agreement with anion 5a-K
stability experiments we made previously (Table 1).
To gain indirect evidence that would provide a link between
the anion 5a-[M] reactivity and the rate of dimerization, we
decided to evaluate the coupling reaction of anions 5a-Li and
5a-K in the presence of selective Li+ and K+ chelating agents (12-
crown-6, 18-crown-6, TDA-1 and HMPA). By this method, we
expected to generate, in situ, a more reactive “naked” anion of
5a (Table 3, entries 8–11). As a consequence, the introduction
of chelating agents increased the speed of sulfone 5a self-
condensation.
We reasoned that the nature of the solvents could have a
similarly large influence. We therefore attempted the coupling
reaction in THF, Et2O and DME (Table 3, entries 4, 12 and 13,
respectively).17 As expected, reactions carried out in Et2O and
THF gave comparable yields, whereas the use of DME increased
the yield of the self-condensation product 7.
Scheme 7 Possible side reactions that could be observed if Barbier-type
reaction conditions are used for the b-keto sulfone 4 synthesis.
Next, we focusedour attention onthe nature ofthe leaving group
in the electrophile (Table 3, entries 14–20). Benzoylating agents
with reactivity similar to that of benzoyl chloride, such as BzF,
BzBr, Bz2O, benzoyl imidazole18 or BzCN, produced the adduct
4a in essentially same yields. On the other hand, methyl benzoyl
ester was found to be unreactive under our reaction conditions
as well as under Barbier-type reaction conditions (Table 3, entry
19 and 20). In both cases, only the product of dimerization 7 was
isolated.
Having devised suitable reaction conditions for b-acyl sulfone
synthesis, the scopes and limitations of this method were explored.
A selection of pertinent results is shown in Table 5.
It was demonstrated that both alkyl and aryl acyl electrophiles
react smoothly with lithiated sulfones 5. In all cases the reaction
yields were greater than 80%, except for when sulfone 5b was
Scheme 8 Determination of bis-sulfone 12b stability in the presence of
base and a sulfonyl anion.
As a consequence, a lower yield of the desired adducts would
be isolated. To verify this hypothesis, the mixing of sulfone 5a
with acetyl chloride (6b) and cyclohexyl acyl chloride (6c) was
investigated (Table 4). As suspected, the desired products 4b and
This journal is
The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1225–1234 | 1229
©

View Article Online
Table 5 Preparation of acyl benzo[d]thiazol sulfones 4, starting from sulfones 5 and acyl-containing electrophiles
Entry
1
Sulfone
Acylating agent
AcCl
Product
Yield (%)a
90
2
3
4
Ac2O
Ac–Im
92
89
83
5
6
7
8
9
92
89
86
78
81
a Overall yields refer to pure, isolated products.
reacted with monoethyl oxalyl chloride (Table 5, entry 8). In
this case the desired product 4f was obtained in 78% yield.
Additionally, even acyl chlorides with enolizable hydrogen atoms
reacted under the given reaction conditions to yield the desired
a-acyl sulfones 4 in very good yields.
After establishing the access to sulfonyl esters, we decided
to extend this methodology to other carboxylic acid derivatives
(Table 7). Interestingly, at this moment the abilities the of
activating/leaving groups presented on electrophile were fully
revealed.
As shown in Table 7, when a Cl- group was used as an
activating/leaving group, ester, thioester and amide BT sulfonyl
derivatives 11 could be easily prepared (Table 7, entries 1–3).
However, in the case of amides, the product of sulfone 5a self-
condensation, compound 7, was also formed, along with the de-
sired adduct, 11m, in 11% yield (Table 7, entry 3). This observation
suggests that in the case of N,N-dialkylamino chlorocarbamates,
the addition of anion 5a-Li to the carbonyl is much slower than in
the case of alkyl chloroformates and thioformates. The difference
in the reactivity between chloroformates and chlorocarbamates
became even more evident when CN- group was used as Cl-
equivalent (Table 7, entries 4–6). In this case, the ester and thioester
derivatives were formed in very good yields (Table 7, entries 4
Synthesis of a-sulfonyl carboxylic acid derivatives
Our new protocol was also applied to the synthesis of a-sulfonyl
carboxylic acid derivatives 11 (Table 6). For this purpose, four
different types of alkoxy carbonylating reagents (bearing Cl,
imidazole,19 OCOR or CN as a leaving group) were tested.
The nature of the leaving group was shown to have little
effect on the reaction yield and all four electrophiles could
be used as coupling partners. Additionally, standard functional
groups are tolerated under the reaction conditions. In all cases,
the desired products, 11, were prepared in good to excellent
yields.
1230 | Org. Biomol. Chem., 2012, 10, 1225–1234
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
Table 6 Preparation of alkoxycarbonyl benzo[d]thiazol sulfones, 11, starting from sulfone 5 and alkoxy carbonyl-containing electrophiles
Entry
1
Sulfone
Electrophile
Cl–COOMe
Product
Yield (%)a
94
2
3
4
Im–COOMe
NC–COOMe
Cl–COOallyl
89
91
95
5
6
Boc2O
Im–COOtBu
94
98
7
Cl–COOMe
88
8
9
10
Im–COOMe
NC–COOMe
Boc2O
94
92
91
11
12
Im–COOtBu
Cl–COOMe
98
88
13
14
15
Im–COOMe
NC–COOMe
Cl–COOMe
94
93
88
16
17
18
Im–COOMe
NC–COOMe
Cl–COOallyl
93
94
89
19
Cl–COOMe
87
20
21
22
Im–COOMe
NC–COOMe
Cl–COOMe
95
94
89
23
24
25
Im–COOMe
NC–COOMe
Cl–COOallyl
93
91
92
a Overall yields refer to pure, isolated products.
This journal is
The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1225–1234 | 1231
©

View Article Online
Table 7 Extending of the methodology to other carboxylic acid derivatives
Yield (%)a
Entry
1
Electrophile
Product
11
7
94
94
63
91
95
<5
2
3
4
5
<5
11
<5
<5
6
7
<5
<5
77
49
a Overall yields refers to pure, isolated products.
and 5), but no amide derivative was formed (Table 7, entry 6). If
dimethyl carbonate was used as electrophile, no adduct 11a of the
addition was observed (Table 7, entry 7).
These results suggest that only sufficiently reactive electrophiles
are capable to react with BT sulfone anions under the coupling
conditions. If the electrophile is not reactive enough (Table 7,
entries 3, 6 and 7), the competitive dimerization reaction starts to
play an important role.
pothesis, we decided to evaluate the bis-sulfone 12b20 stability
under the basic conditions (Scheme 8). No degradation of bis-
sulfone 12b was observed when it was placed under the reaction
conditions. This observation suggests that, at any time, desired bis-
sulfone 12b was not formed under the given reaction conditions.
At this stage we believe that our unsuccessful synthesis of BT
sulfonyl derivatives 12a–c is caused by the lack of reactivity of the
corresponding electrophiles (TMSCl, TsCl and MsCl) towards
nucleophile 5a-Li.
Non-carbonyl-containing electrophiles
Conclusions
At this stage, we decided to extend our methodology to other non-
carbonyl-containing electrophiles such as TMSCl, TsCl and MsCl
(Table 8). Unfortunately, in all cases only the product of sulfone
5a self-condensation, compound 7, was isolated. Interestingly, no
traces of any self-condensation or degradation products that could
arise from potentially formed adducts 12a–c were identified.
This observation suggests that the self-condensation is the
fastest step under our reaction conditions. To verify this hy-
In summary, we have disclosed a short and efficient approach
to a-acyl and a-carboxylic acid-derivative sulfones, 11, starting
from heterocyclic sulfones and appropriate electrophile. Reaction
conditions tolerate various functionalities, such as TBDPS ethers,
phenyl ethers and halogenated or unsaturated alkanes. We believe
that this is a general approach towards this class of C-nucleophiles,
which can be easily used in the context of the formation of olefins
1232 | Org. Biomol. Chem., 2012, 10, 1225–1234
This journal is
The Royal Society of Chemistry 2012
©

View Article Online
Table 8 Coupling reaction using other non-carbonyl-containing electrophiles
Yield (%)a
Entry
E+
Conditions
12
7
1
2
3
4
5
6
TMSCl
TMSCl
TsCl
TsCl
MsCl
MsCl
LiN(TMS)2 (1.2 equiv), -78 ◦C, THF, then TMSCl (1.2 equiv)
TMSCl (1.2 equiv), -78 ◦C, THF, then LiN(TMS)2 (1.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, then TsCl (1.2 equiv)
MsCl (1.2 equiv), -78 ◦C, THF, then LiN(TMS)2 (2.2 equiv)
LiN(TMS)2 (2.2 equiv), -78 ◦C, THF, then TsCl (1.2 equiv)
MsCl (1.2 equiv), -78 ◦C, THF, then LiN(TMS)2 (2.2 equiv)
12a, <5
12a, <5
12b, <5
12b, <5
12c, <5
12c, <5
62
54
55
57
48
60
a Overall yields refer to pure, isolated products.
or alkynes. Unfortunately, the extension of this methodology
to other, non-carbonyl containing electrophiles proved to be
unsuccessful. We believe that it is due to a low reactivity of
these electrophiles. Additionally, the dimerization of BT sulfone 5a
under the basic conditions was evaluated using both, experimental
approach and theoretical calculations.
Finally, we believe that our simple newly developed approach
to b-carbonyl BT sulfones will extend their use beyond the field
of asymmetric organocatalysis. Further development and use of
b-carbonyl heterocyclic sulfones of general structure 4 and 11 is
now under progress in our laboratory and will be reported shortly.
The sulfones 5a–g4 and bis-sulfone 12b20 were prepared accord-
ing to published procedures.
General procedure for the synthesis of b-oxo sulfonyl derivatives
A solution of sulfone (1.0 mmol, 1.0 equiv) in THF (5 mL, 0.20
M) was cooled to -78 ◦C and LiHMDS (1.0 M sol. in THF)
(2.2 mL, 2.2 mmol, 2.2 equiv) was added dropwise. The colour of
the reaction mixture turned from colourless or slightly yellow to
orange/red within approx. 20 to 30 s. Immediately afterwards,
a solution of acylating agent (acyl halide, carboxylic acid an-
hydride, acyl nitrile or acyl imidazole) or alkoxy carbonylating
agent (alkoxy chloroformate, alkoxy imidazoylformate, alkoxy
cyanoformate or Boc2O) (1.1 mmol, 1.1 equiv) in THF (0.5 mL)
was added. The colour of the reaction mixture faded within few
Experimental section
General experimental
◦
minutes. The resulting mixture was stirred at -78 C for 30 min,
allowed to warm to 0 ◦C within 1 h and stirred at 0 ◦C for a further
30 min before sat. aq. sol. of NH4Cl (15 mL) was added. The whole
mixture was extracted with EtOAc (3 ¥ 75 mL) and the combined
organic layers were washed with brine (50 mL), dried over MgSO4,
filtered and the solvents were removed under reduced pressure. The
residue was purified by flash column chromatography on SiO2. See
ESI file for characterization data.†
1H and 13C NMR spectra were recorded on a Brucker AC-300
Avance II (working frequency 300 MHz and 75 MHz, respectively)
at ambient temperature in CDCl3 (Aldrich). Coupling constants (J
value) are reported in hertz. The chemical shifts are shown in ppm
downfield from tetramethylsilane, using residual chloroform (d =
7.27 in 1H NMR) or the middle peak of CDCl3 carbon triplet (d =
77.23 in 13C NMR) as an internal standard. Low resolution mass
spectroscopic data were recorded on a Finigan TSQ 7000. High
resolution mass spectra were acquired at the University College
London Mass spectroscopy facility using the Thermo Finnigan
MAT900xp spectrometer.
Acknowledgements
J. P. is grateful to Prof. Istva´n E. Marko´ (U.C.L.) for his
continuous support. We gratefully acknowledge the Universite´
catholique de Louvain, the Fond de la Recherche Scientific
(Charge´ de Recherche F. R.S.-FNRS to J. P., Chercheur Qual-
ifie´ F.R.S.-FNRS to R. R. and FRFC Project N◦ 2.4502.05),
Socrates-Erasmus exchange program (H. S.) and Tohoku
University G-COE program (IREMC) (H. S.) for financial
support.
Melting points were determined using a Bu¨chi Flawil apparatus
and are uncorrected.
Chemicals were purchased from Acros, Sigma-Aldrich and
Fluka and were used as received. THF was distilled under argon
from sodium benzophenone ketyl. Flash chromatography was
performed on silica gel 60 (40–63 mm) (ROCC). All reactions were
carried out under an atmosphere of argon in flame-dried apparatus
with magnetic stirring, unless otherwise indicated. Brine refers to
a saturated aqueous solution of sodium chloride.
The identity of known products was confirmed by comparison
with literature spectroscopic data. The structure determination
of new compounds was made with a help of 2D-COSY, HSQC,
HMBC, 2D-NOESY and NOEdiff experiments.
Notes and references
1 For reviews see: (a) A.-N. R. Alba, X. Companyo´ and R. Rios, Chem.
Soc. Rev., 2010, 39, 2018; (b) M. Nielsen, C. B. Jacobsen, N. Holub,
M. W. Pixao and K. A. Jørgensen, Angew. Chem. Int. Ed., 2010, 49,
2668For recent leading articles see: (c) C. B. Jacobsen, M. Nielsen, D.
Worgull, T. Zweifel, E. Fisker and K. A. Jørgensen, J. Am. Chem. Soc.,
2011, 133, 7398; (d) A. Landa, A. Puente, J. I. Santis, S. Vera Oiarbide
and C. Palomo, Chem.–Eur. J., 2009, 15, 11954; (e) J. L. Garc´ıa Ruano,
V. Marcos and J. Alema´n, Chem. Commun., 2009, (29), 4435; (f) G. K.
Preparation of imizadole-containing acylating and alkoxycar-
bonylating reagents: Ac–Im and Bz–Im,18 and Im–CO2Me and
Im–CO2tBu.19
This journal is
The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 1225–1234 | 1233
©

View Article Online
S. Prakash, F. Wang, T. Steward, T. Mathew and G. A. Olah, Proc.
Natl. Acad. Sci. U. S. A., 2009, 106, 4090; (g) A. Alba, X. Companyo´,
A. Moyano and R. Rios, Chem.–Eur. J., 2009, 15, 11095; (h) S. Mizuta,
N. Shibata, Y. Goto, T. Furukawa, S. Nakamura and T. Toru, J. Am.
Chem. Soc., 2007, 129, 6394; (i) C. Cassani, L. Bernardi, F. Fini and
A. Ricci, Angew. Chem., Int. Ed., 2009, 48, 5694; (j) A. Landa, M.
Maestro, C. Masdeu, A. Puente, S. Vera, M. Oiarbide and C. Palomo,
Chem.–Eur. J., 2009, 15, 1562; (k) M. Nielsen, C. B. Jacobsen, M. W.
Paixa˜o, N. Holub and K. A. Jørgensen, J. Am. Chem. Soc., 2009, 131,
10581.
2 (a) B. M. Trost, J. D. Chisholm, S. T. Wrobleski and M. Jung, J. Am.
Chem. Soc., 2002, 124, 12420; (b) B. Zajc and S. Kate, Org. Lett., 2006,
8, 4457; (c) P. R. Blakemore, D. H. K. Ho and W. M. Nap, Org. Biomol.
Chem., 2005, 3, 1365; (d) D. R. Williams and L. Fu, Org. Lett., 2010,
12, 808; (e) L. Xu, J. Cheng and L. Trudell, J. Org. Chem., 2003, 68,
5388.
9 Originally we thought that the self-condensation of 5a in basic media
is a consequence of the reaction between already deprotonated sulfone
5a with its parent molecule.
10 Geometry optimisation has been carried out using the Jaguar 7.5
program package (Jaguar 7.5, Schro¨dinger, LLC, New York, NY, 2008)
at the B3LYP/6-31+G*(THF) level. NBO analysis has been performed
at the B3LYP/cc-pVTZ level of theory. See Electronic Supplementary
Information† for full details.
11 Interestingly, optimization of 5a-Li reveals that in the most stable
isomer lithium cation is not linked to the carbon atom but instead
complexed to both the nitrogen atom and one oxygen of the sulfone
function. Since the charge on C-1 carbon atom was found not to depend
highly on the nature of the isomer, results given in Table 2 refer only to
the lowest energy isomer (see ESI† for full results).
12 In fact calculations suggest that the deprotonated sulfone 5a-Li is even
a slightly better electrophile than its protonated form 5a.
13 Prepared for this purpose via standard literature protocol: C. B.
Jacobsen, L. Lykke, D. Monge, M. Nielsen, L. K. Ransborg and K. A.
Jørgensen, Chem. Commun., 2009, (43), 6554.
3 T. Zweifel, M. Nielsen, J. Overgaard, C. B. Jacobsen and K. A.
Jørgensen, Eur. J. Org. Chem., 2011, 2011, 47.
4 J. Posp´ısˇil and H. Sato, J. Org. Chem., 2011, 76, 2269.
5 Sylvestre Julia reaction (reaction of BT sulfones 5 with aldehydes
or ketones) was not the object of this research and therefore simple
carbonyl-containing electrophiles (aldehydes and ketones) were not
used. For more details about Sylvestre Julia reaction see: (a) P. R.
Blakemore, J. Chem. Soc., Perkin Trans. 1, 2002, 2563; (b) C. A¨ıssa,
Eur. J. Org. Chem., 2009, 2009, 1831; (c) J. Posp´ısˇil, Tetrahedron Lett.,
2011, 52, 2348.
6 J. B. Baudin, G. Hareau, S. A. Julia, R. Lorne and O. Ruel, Bull. Soc.
Chim. Fr., 1993, 130, 856–878.
7 Of course, the maximal yield of sulfone 5a self-condensation reaction
is 50% but for comparison purposes the yield was recalculated to 100%
value. Valid for all the yields of self-condensation reaction presented in
this paper.
14 The competitive reaction that might occur with sulfone 5 are not
considered, because we showed that the transformation of sulfone 5a
to its anion 5a-Li, in the presence of the excess of the base, is very fast
(Table 1, entry 7).
15 The solution of BzCl in THF was precooled (-78 ◦C) before addition.
16 A precooled (-78 ◦C) solution of LiN(TMS)2 in THF was added to a
cold (-78 ◦C) solution of sulfone 5a and BzCl in THF.
17 Other solvents such as toluene or DMF (reaction performed at -50 ◦C)
were also tested, but the desired adduct 4a was isolated in significantly
lower yields.
18 Prepared in situ from the corresponding carboxylic acid: (a) C. A.
Ibarra, R. C. Rodriguez, M. C. Fernandez Monreal, F. J. Garcia
Navarro and J. J. Martin Tesorero, J. Org. Chem., 1989, 54, 5620.
19 Prepared according to the literature procedure: (a) S. P. Rannard and
N. J. Davis, Org. Lett., 1999, 1, 933; (b) S. P. Rannard and N. J. Davis,
Org. Lett., 2000, 2, 2117.
8 (a)
The
pKa
values
known
in
the
litterature
(http://www.chem.wisc.edu/areas/reich/pkatable/): for (TMS)2N–
H, pKa = 30 (in DMSO); (b) for BT SO2–CH2–H, pKa value is not
known, but pKa for PhSO2CH2–H is 29 (in DMSO). The value for 5a
is expected to be approximately the same or lower.
20 Prepared for this purpose via standard literature protocol: M. He, A.
K. Ghosh and B. Zajc, Synlett, 2008, 2008, 999.
1234 | Org. Biomol. Chem., 2012, 10, 1225–1234
This journal is
The Royal Society of Chemistry 2012
©