Enantioselective synthesis, chiroptical properties and absolute configuration of 3-aminosubstituted isothiazole S-oxides

Article abstract of DOI:10.1016/j.tetasy.2009.08.023

Herein we report a mild and efficient method to synthesize chiral 3-aminosubstituted isothiazole sulfoxides taking advantage of (+)- and (-)-((8,8-dichlorocamphoryl)sulfonyl)oxaziridine under microwave irradiation. The determination of the absolute config

Full text of DOI:10.1016/j.tetasy.2009.08.023

Tetrahedron: Asymmetry 20 (2009) 2247–2256
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Enantioselective synthesis, chiroptical properties and absolute configuration
of 3-aminosubstituted isothiazole S-oxides
a
a
Alessandro Casoni a, Giuseppe Celentano a, Francesca Clerici a, , Alessandro Contini , Maria Luisa Gelmi ,
*
Giuseppe Mazzeo b, Sara Pellegrino a, Carlo Rosini b
a DISMAB, Sezione di Chimica Organica ‘A. Marchesini’, Università di Milano, Via Venezian 21, 20133 Milano, Italy
b Dipartimento di Chimica, Università della Basilicata, via N. Sauro 85, 85100 Potenza, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 27 July 2009
Accepted 20 August 2009
Available online 15 September 2009
Herein we report a mild and efficient method to synthesize chiral 3-aminosubstituted isothiazole sulfox-
ides taking advantage of (+)- and (ꢀ)-((8,8-dichlorocamphoryl)sulfonyl)oxaziridine under microwave irra-
diation. The determination of the absolute configuration of the chiral sulfoxide was achieved by theoretical
calculation of the CD spectra. The reason for the observed stereoselectivity was enlightened by means of
analysis of our data using DFT calculations.
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
enantiopure 3-aminoisothiazole derivatives (Scheme 1). Computa-
tional analysis of the oxidation mechanism was performed for a
The importance of the isothiazole nucleus containing com-
pounds appears to have grown over the recent years. New syn-
thetic approaches and unprecedented reactions have been
reported and numerous technical and pharmaceutical applications
have been discovered.1a–e
better understanding of the observed enantioselectivity.
O
O
N
R1
R2
R1
R2
R1
R2
S
S
S
N
N
+
In particular, both the synthesis and the chemical reactivity of
3-aminoisothiazole 1,1-dioxides have been studied by us for many
years, and the important transformations and applications of these
compounds have been found as precursors both to N- and/or S-
containing heterocycles and to open chain compounds through
ring-opening reactions.2 Moreover, it has to be remembered that
several representatives of the 3-aminoisothiazole dioxide series
display interesting biological activity as antiproliferating
agents.3–5 Very recently we described a mild and efficient method
to synthesize chiral racemic 3-aminosubstituted isothiazole sulf-
oxides by taking advantage of arylsulfonyloxaziridines.6 It should
be taken into account that the preparation of chiral compounds
has been an important and challenging area of contemporary syn-
thetic organic chemistry, and that especially the preparation of
new chiral ligands for application in asymmetric catalysis contin-
ues to occupy the research focus these days. New classes of ligands
that might offer new opportunities for applications or provide in-
sight into the fundamental chemical processes are always interest-
ing. One relatively rare class of ligands is that in which the
stereogenicity resides not at carbon atoms, but at heteroatomic
sites such as sulfur atoms.7 This phenomenon prompted us to
study the asymmetric oxidation of the prochiral sulfur to produce
NHR3
NHR3
(R)-2
NHR3
(S)-2
1
Scheme 1. 3-Aminosubstituted-isothiazoles 1 and S-oxides (R)-2 and (S)-2.
Moreover, a study of the possibility to assign the absolute con-
figuration of the new 3-aminosubstituted isothiazole S-oxides by
applying methods based on the quantum chemical calculation of
the chiroptical properties has also been performed.
2. Results and discussion
2.1. Asymmetric oxidation
With the aim of synthesizing optically active sulfoxides, we
considered either the enantioselective oxidation using chiral oxi-
dizing reagents or the diastereoselective approach where the intro-
duction of a stereogenic centre on the starting isothiazole would be
potentially capable of providing asymmetric induction.
2.1.1. Diastereoselective oxidation
This basic process makes use of the proximity of a defined
stereogenic centre to relay stereochemistry to the newly formed
sulfoxide. The 3-(S)-(1-phenylethyl)aminoisothiazole 1a was pre-
pared starting from enantiomerically pure (S)-1-phenylethylamide
2 by a known protocol through the intermediate isothiazolone 3
(Scheme 2).8
* Corresponding author. Tel.: +39 0250314472; fax: +39 0250314476.
E-mail address: francesca.clerici@unimi.it (F. Clerici).
0957-4166/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2009.08.023

2248
A. Casoni et al. / Tetrahedron: Asymmetry 20 (2009) 2247–2256
O
O
O
S
CH3
S
S
1) POCl3
S
25°C, 12h,
CH2Cl2
SO2Cl2
N
SCH2CH2CO NH
Ph
N
N
Ph
+
N
+
PhSO2
N
H
H
2) NH3,
CH3CN
ClCH2CH2Cl
Ph
CH3
NH
NH
NH
O
H
2
H
H
4
3
2
Ph
Ph
Ph
CH3
CH3
CH3
(S)-1a
(R,S)-2a
(S,S)-2a
Total yield 80%,
d.e. 6%
Scheme 2. Synthesis and oxidation of 3-(S)-(1-phenylethyl)aminoisothiazole 1a to (S,S)-2a and (R,S)-2a.
In a previously reported study we have shown that arylsulfonyl-
oxaziridines were able to selectively oxidize 3-aminosubstituted
isothiazoles to the chiral racemic corresponding sulfoxides in a
mild and efficient way.6 We applied such a method to (S)-1a by
using N-phenylsulfonyl-4-phenyl-oxaziridine 4 as the oxidant
but, in spite of the good yields, the reaction, unfortunately, gave
very low diastereoselection (de 6%). This unsatisfactory result
could be explained by arguing that the chiral amino group is too
far from the prochiral sulfur atom to exert a powerful influence
on the stereochemical outcome of the oxidation.
O
X
N
O
N
O2
S
N
S
S
S
X
+
CH2Cl2
NH
N
+
+
S
O
O2
NHBn
(S)-2b
NHBn
NHBn
NHBn
(R)-2b
(-)-6a X=H
(-)-6b X=Cl
7b
1b
Scheme 3. Oxidation of 3-alkylaminoisothiazoles 1b with oxaziridines (ꢀ)-6a,b.
jor enantiomer was the counterpart of the one obtained with (ꢀ)-
6b, that is, the (S)-isomer. The best results in terms of yields and
selectivity were obtained by performing the reaction under micro-
wave irradiation. When a solution of 1b and (ꢀ)-6b (1 equiv) in
CH2Cl2 was irradiated at 400 W (80 °C) enantiomers (R)-2b and
(S)-2b were produced in 82% overall yield and 50 ee%, together
with small amounts of 7b (8%) (Table 1, entry 5).
2.1.2. Enantioselective oxidation
Enantiopure N-sulfonyloxaziridines which are known to be use-
ful reagents for the enantioselective oxidation of a variety of sub-
strates in high and predictable stereoselectivity were
considered.9 Among the different types of the available enantio-
pure oxaziridines, compounds 5 and 6a,b,c,d were evaluated
(Fig. 1).10,11
Table 1
X
N
X
N
Oxidation of 1b with (ꢀ)- and (+)-6a,b
X
X
Entry
Oxidant
React. cond.a
Time (h)
Yield % 7
ee %
NSO2Ph
1
2
3
4
5
6
(ꢀ)-6a
(ꢀ)-6b
(+)-6b
(ꢀ)-6a
(ꢀ)-6b
(+)-6b
A
A
A
B
B
B
240
168
168
7
1
1
62 (20)
60 (18)
60 (15)
24 (10)
82 (8)
10
18
20
24
50
55
S
S
O
O
O
O2
O2
5
(+)-6a X=H
(+)-6b X=Cl
(+)-6c X,X= =O
(+)-6d X= OMe
(–)-6a X=H
(–)-6b X=Cl
80 (7)
a
Reaction conditions: A: CH2Cl2, 25 °C; B: lW, 400 W, 80 °C.
Figure 1. Structures of enantiopure N-sulfonyloxaziridines 5, 6.
For the first experiments we chose N-benzylaminoisothiazole
1b which was reacted with N-sulfonyloxaziridines 5, (+)-6c and
(+)-6d, but the desired isothiazole sulfoxide 2b was not obtained
in appreciable yields or in good chemoselectivity either at room
or higher temperature, or by irradiation with microwaves or soni-
cation. Improved results were obtained with N-sulfonyloxaziri-
dines (ꢀ)-6a and (ꢀ)-6b, which oxidized 3-aminoisothiazole 1b
at room temperature in CH2Cl2, producing a mixture of enantio-
mers (R)-2b and (S)-2b in appreciable yields (60–62%) and with
reasonable chemoselectivity (7b: 15–20%) but over a very long
reaction time (168–240 h) (Scheme 3). Despite the good yields
and selectivity for the sulfo format, disappointing results were ob-
tained in terms of enantioselectivity with both (ꢀ)-6a (ee 10%) and
(ꢀ)-6b (ee 18%) (Table 1, entries 1 and 2). In both cases, one enan-
tiomer, the (R)-enantiomer, was formed in slight excess as shown
by chiral HPLC analyses. The absolute configuration was estab-
lished as described later.
As expected, comparable results were obtained with (+)-6b (Ta-
ble 1, entry 6). It was evident that microwave irradiation enhanced
the rate of the reaction (1 h vs 168 h) and the enantioselectivity
(50% vs 20%). The reaction rate was higher with (ꢀ)-6b (Table 1,
entry 5), as compared to (ꢀ)-6a (Table 1, entry 4) and this is con-
sistent with the fact that electronegative groups attached to the
oxaziridine carbon in N-sulfonyloxaziridines increase the rate of
the oxidation.12
Comparable results were obtained with 1c (Scheme 4, Table 2,
entries 1 and 2).
Finally, the oxidation of (S)-1a with (+)- and (ꢀ)-6b showed
similar behaviour and a mixture of 3-(S)-(1-phenylethyl)aminoiso-
thiazole-(S)-1-oxide (S,S)-2a and 3-(S)-(1-phenylethyl)aminoiso-
thiazole-(R)-1-oxide (S,R)-2a was formed over
a very long
reaction time (240 h) and in unsatisfactory yields (36%) and only
8% de (Scheme 4, Table 2, entry 3). However, considerable improve-
ments in yields (50%) and diastereoselectivity (60%) were obtained
by performing the reaction under microwave irradiation (400 W),
with an equimolar amount of (+)-6b in CH2Cl2 at 80 °C (Table 2, en-
try 4). As expected, analogous results in terms of yield and de% were
Comparable results were obtained by performing the reaction
with (+)-6b, which afforded a mixture of (R)-2b and (S)-2b (Table
1, entry 3). As expected, chiral HPLC analysis revealed that the ma-

A. Casoni et al. / Tetrahedron: Asymmetry 20 (2009) 2247–2256
2249
O
O2
S
N
X
N
O
N
S
S
CH2Cl2
S
X
N
+
1,2,7a R = CH(Me)Ph
1,2,7c R = Me
+
+
N
S
O
NHR
O2
NHR
NHR
NHR
(-)-6b X=Cl
(+)-6b X=Cl
1a,c
7a,c
(R)-2a,c
(S)-2a,c
Scheme 4. Oxidation of 3-alkylaminoisothiazoles 1a,c with oxaziridines (ꢀ)- and (+)-6b.
Table 2
Oxidation of 3-alkylaminoisothiazoles 1c,a with oxaziridines (ꢀ) or (+)-6b
Entry
Compd
R
Oxidant
React. cond.a
Time (h)
Yield % 7
ee %
1
2
3
4
5
1c
1c
1a
1a
1a
Me
Me
CH(Me)Ph
CH(Me)Ph
CH(Me)Ph
(ꢀ)-6b
(+)-6b
(+)-6b
(+)-6b
(ꢀ)-6b
B
B
A
B
B
2.3
2.5
240
2.5
3
70 (15)
60 (10)
36 (10)
50 (14)
56 (16)
43.6
49.6
8b
60b
40b
a
Reaction conditions: A: CH2Cl2, 25 °C; B: lW, 400 W, 80 °C.
Diastereoisomeric excess.
b
obtained by performing the reaction with (ꢀ)-6b, which afforded
the opposite ratio of enantiomers 2a (Table 2, entry 5).
H
H
Br
H
S
1) LDA, -60°C,THF,10 min.
2) BrCN
S
N
N
The presence of the stereocentre on the isothiazole 1a did not
seem to significantly influence the formation of the new stereocen-
tre on the sulfur atom. With this procedure, we were able to isolate
the diastereoisomeric compounds (S,S)-2a and (R,S)-2a in pure
form (98% purity, chiral HPLC analyses) after column chromatogra-
phy and crystallization with H2O/MeOH.
NHBn
NHBn
1b
1f
Scheme 6. Synthesis of compound 1f.
Theoretical and experimental studies support an SN2 type reac-
tion for the oxygen transfer from N-sulfonyloxaziridines to the sul-
fur nucleophiles. It was concluded that molecular recognition was
largely steric in origin as dictated by the substituents on both the
oxaziridines and substrates.13–15 Apparently, the sulfur atom on
the planar isothiazole system of compounds 1a, 1b and 1c, is not
surrounded by a sterically demanding group. Accordingly, we pre-
pared isothiazole derivatives 1d–g substituted at C-4 or C-5 with a
halogen, aiming to evaluate the role of the substituent in the selec-
tivity of the process.
The new compounds, 3-benzylamino-4-chloro-isothiazole 1d
and 3-benzylamino-4-bromo-isothiazole 1e were synthesized by
a known procedure with some modifications (Scheme 5).16–18
To synthesize the previously unreported compound 1f bearing a
bromine atom on C-5, we generated the 5-lithium anion by react-
ing 1b with LDA at ꢀ60 °C, and capturing the anion with BrCN to
afford the desired compound 1f (Scheme 6).19,20
reaction was performed on 1e, with a bromine atom on C-4, we ob-
served the formation of 2e in very good overall yield (80%) and sat-
isfactory reaction time (12 h) with a marked enantiomeric excess
(65%) (Table 3, entry 4). An increase in the ee% (65%) was also ob-
served with the 4-chloro derivative 1d (Table 3, entry 2). In this
last case, a longer reaction time was required. It was apparent that
the presence of a halogen atom on C-5 had no influence on the
enantioselectivity of the oxidation reaction but the presence of a
halogen atom on C-4 significantly increased the formation of one
enantiomer with respect to the other. The observed enantioselec-
tivity for the oxidation reactions was rationalized by performing
DFT calculations in order to analyze the transition structures lead-
ing to (R)- and (S)-enantiomers.
2.2. Configuration assignment
The sulfoxidation of compounds 1d–g was performed by apply-
As stated previously, compounds (S,S)-2a and (R,S)-2a were
obtained in crystal form. Unfortunately, many attempts to grow
crystals suitable for X-ray analysis in order to define the absolute
configuration of the sulfur in the sulfoxide group, failed. In order
to solve this problem, we decided to make use of chiroptical spec-
troscopy. In order to determine the absolute configuration of the
new 3-aminosubstituted isothiazole S-oxides, we decided to apply
the methods based on the quantum chemical calculation of the
optical rotation, obtained for an arbitrarily assumed absolute con-
ing the optimum reaction conditions described above (i.e.,
lW,
400 W, 80 °C) using both (+)-6b and (ꢀ)-6b as the oxidants
(Scheme 7). The results are reported in Table 3. The 5-chloro deriv-
ative 1g afforded the corresponding sulfoxides 2g in very good
overall yields (83%, 12 h) but with very low enantiomeric excess
(Table 3, entry 1).
Similar results were obtained with 1f, in which the bromine
substituent resides at C-5 (Table 3, entry 2). However, when the
R1
R2
R1
R2
S
S
1d: SO2Cl2, dichloroethane;
1) POCl3
2) NH3
N
NBn
O
(SCH2CH2CONHBn)2
1e: SO2Cl2,CH2Cl2,Δ,1h; Br2 18h
NHBn
1d R1=H, R2=Cl
1e R1=H, R2=Br
Scheme 5. Synthetic strategy for compounds 1d,e.

2250
A. Casoni et al. / Tetrahedron: Asymmetry 20 (2009) 2247–2256
The electronic spectrum shows two main regions of absorption:
a low energy one centred at about 330 nm (emax 4000 ca.) and
extending from 380 to 280 nm, and a high energy one which in-
O
O
N
R1
R2
R1
R2
R1
R2
S
S
(+)-6b or (-)-6b
CH2Cl2, μW
S
N
+
N
creases monotonically from 280 nm down to 200 nm, where an
e
NHBn
(R)-2d-g
NHBn
NHBn
value of about 15,000 can be measured. In the ECD spectrum a very
1d-g
(S)-2d-g
1,2 d R1=H, R2=Cl
1,2 e R1=H, R2= Br
1,2 f R1=Br, R2=H
1,2 g R 1=Cl, R2=H
Scheme 7. Oxidation of 4- or 5-substituted 3-benzylaminoisothiazoles with oxaz-
iridines (+)-6b or (ꢀ)-6b.
Table 3
Oxaziridine oxidation of isothiazoles 1d–g
Entry
Compd
R1
R2
Oxaziridinea
Time (h)
Yield %
ee %
1
2
3
4
1g
1f
1d
1e
Cl
Br
H
H
H
Cl
Br
(+)-6b
(+)-6b
(+)-6b
(+)-6b
12
12
24
12
83
70
54
80
6
6.4
65
65
Figure 2. Absorption (UV curve) and circular dichroism (CD curve) of (+)-3-
methylaminoisothiazole 1-oxide 2c in ethanol.
H
a
Compound (ꢀ)-6b was also tested affording comparable results in terms of
yields and ee, obviously in favour of the opposite enantiomer.
broad, negative Cotton effect can be observed between 380 and
280 nm (
positive CD band centred at 250 nm, in turn followed by a third
Cotton effect at 210 nm with
ꢀ42. We can now attempt at
D
emax of the order of ꢀ1), followed by a strong (
De +20)
D
e
figuration and the comparison with the experimental data. This
method was introduced in 199721a,b,22a,b however since then, sev-
eral papers appeared in this field: accounts of these investigations
can be found in Refs. 21c and 23, whilst in Ref. 24 number of appli-
cations to the determination of absolute configurations of new
molecules can be found. Clearly this approach requires that the
simulated values are fully reliable in sign and order of magnitude.
Since the 1997 papers cited above, where the first optical rotation
calculations were carried out using the Hartree–Fock/small basis
set approach, many studies have been carried out at different levels
of theory to set up methods which couple reliability and accuracy
to a reasonable computational effort.23 From this point of view the
investigations of Stephens et al. have demonstrated that Time
Dependent Density Functional Theory (TDDFT) calculations with
the B3LYP hybrid functional and extended basis sets (i.e., aug-cc-
pVDZ or higher) give completely reliable results (with an accept-
able computational effort) in reproducing experimental optical
reproducing the above-mentioned chiroptical data by means of
some TDDFT/B3LYP/aug-cc-pVDZ calculations. The calculations ob-
ject of this investigation have been carried out as described in Sec-
tion 4.1. Assuming an (S)-absolute configuration for the sulfur
stereogenic centre of 3-methylaminoisothiazole 1-oxide 2c, by
means of a sequence of molecular mechanics/DFT calculations
two different conformers have been found in the gas phase, the
structure, relative energy and population of which are reported
in Figure 3.
The most stable conformer (conf. 1) is characterized by having
the methyl group linked to the nitrogen atom on the same side
of the N(2), when the same group is placed near to the vinyl hydro-
gen linked to C(4) (conf. 2) the free energy increases significantly
(+2.07 kcal/mol) leading to populations of 97% and 3%, respec-
tively. The optical rotation prediction (TDDFT/B3LYP/aug-cc-pVDZ
level of theory) gives +71 (conf. 1) and ꢀ119 (conf. 2): these values
afford a weighted average value of +65. This figure, obtained for the
(S)-configured compound, is of the same sign of the experimental
number, even if, from a quantitative point of view, this number
is, at maximum, almost 50% of the experimental value. In order
to have a better agreement between experimental and predicted
values we introduced the solvent effect, by means of the PCM mod-
el both in the optimization of the molecular geometry and in the
calculation of the optical rotation.26 The results are collected in Ta-
ble 4 (ethanol), Table 5 (methanol) and Table 6 (chloroform).
Looking at Tables 4–6, it can be observed that: (i) conformer 1
remains, in all solvents, the most stable, and its population further
increases in all of them; (ii) whilst in the gas phase we saw that the
optical rotation of the most stable conformer was smaller in abso-
lute value than the optical rotation of the second conformer, the
situation is completely reversed in all the solvents; (iii) as a conse-
quence, the theoretical rotation is determined, in practice, by the
most stable conformer, even if the rotations of the two conformers
are opposite in sign; (iv) the weighted average optical rotation (i.e.,
the number to be compared with the experimental value) is always
positive for the (S)-absolute configuration, as observed experimen-
tally for the three solvents: this strongly indicates the correlation
(+)/(S) for 3-methylaminoisothiazole 1-oxide; (v) from a numerical
rotations of at least 100 deg [dm g/cm3]ꢀ1 25
Hence we decided
.
to apply the Stephens protocol to assign the absolute configuration
of these new compounds by predicting their optical rotations and
comparing these data with the experimental values. The neces-
sary experimental data were taken from 3-methylaminoisothiaz-
ole 1-oxide, that is, a compound having a small size (this fact
allows to use the extended aug-cc-pVDZ basis set, suggested
by Stephens, even on a common desktop PC) and a reduced con-
formational freedom, only due to the possible rotations around
the C(3)–N bond: for these reasons a heavy computational effort
can be excluded. Then, 3-methylaminoisothiazole 1c was oxi-
dized by (ꢀ)- and (+)-6b by obtaining the corresponding sulfox-
ide 2c with [
methanol), ee 49.6%. From these numbers we can deduce an
absolute value for [ of 3-methylaminoisothiazole 1-oxide of
a]
D = ꢀ60 (c 1, methanol), ee 43.6% and +69 (c 1,
a]
D
138 in methanol. Changing the solvent from methanol to either
ethanol or chloroform causes some variations in the absolute va-
lue, but, most importantly, no changes of sign are observed: it
becomes +165 (c 1, ethanol) and +225 (c 1, chloroform). In addi-
tion, the electronic circular dichroism (ECD) spectrum has also
been measured in ethanol and it is reported, together with the
absorption spectrum in Figure 2.

A. Casoni et al. / Tetrahedron: Asymmetry 20 (2009) 2247–2256
2251
In Figure 4 the predicted ECD spectra of (S)-3-methylaminoiso-
thiazole 1-oxide as obtained as a weighted average of the ECD
spectra of the two conformers, taking into account the first 30 ex-
cited states in both the gas phase and ethanol, are reported. The
ECD spectra have been calculated in both the length and velocity
formalism, these calculated spectra are almost coincident indicat-
ing the high level of the calculation. For this reason in Figure 4 only
the velocity-form predicted spectra are reported. As far as the gas
phase calculations are concerned, the positions of ECD intensity
maxima in the theoretical spectrum do not perfectly match those
of the theoretical spectrum (the predicted spectrum is consider-
ably red shifted with respect to the experimental one, an effect
of this type has been already observed in TDDFT calculations27
)
Figure 3. Stable conformers of (S)-3-methylaminoisothiazole 1-oxide 2c.
and the predicted ECD intensities are in general smaller than the
experimental data. Nevertheless, the experimental spectrum
shows a sequence of negative, positive, negative Cotton effects.
The same features are reproduced in the theoretical spectrum,
again confirming the (+)/(S) correlation found with the optical
rotation calculations. Including the solvent (ethanol) effect in the
above-mentioned calculation significantly improves the agreement
between the calculated and experimental data, strongly supporting
the (+)/(S) configurational correlation. In fact, the complete se-
quence of negative, positive and negative Cotton effects observed
in the experimental spectrum is very well reproduced by the pres-
ent calculations. In particular, a significant solvent effect can be ob-
served in the intensities of the ECD bands, which are now very well
simulated. It can also be seen that there is a reduction in the
amount of red-shift of the predicted ECD bands with respect to
the experimental ones. However, the remaining red-shift and the
increase in intensity of the ECD bands could be responsible for
the overestimation of the optical rotation data. We can now con-
clude that (+)-3-methylaminoisothiazole S-oxide 2c possesses an
(S) absolute configuration.
Table 4
Relative-free energies with respect to the most stable conformer (kcal molꢀ1),
populations, [
values (deg [dm g/cm3]ꢀ1) obtained at TDDFT/B3LYP/aug-cc-pVDZ
level for each conformer of (S)-methylaminoisothiazole 1-oxide in ethanol
a
]
D
D
G
% Pop
Optical rotation
Calcd
Scaled
Conf. 1
Conf. 2
0
2.7
99
1
+280
ꢀ76
+277
ꢀ0.8
Total
276
Table 5
Relative-free energies with respect to the most stable conformer (kcal molꢀ1),
populations, [
values (deg [dm g/cm3]ꢀ1) obtained at TDDFT/B3LYP/aug-cc-pVDZ
level for each conformer of (S)-methylaminoisothiazole 1-oxide in methanol
a
]
D
D
G
% Pop
Optical rotation
Calcd
Scaled
Conf. 1
Conf. 2
0
2.7
99
1
+290
ꢀ74
+287
ꢀ0.7
2.3. Computational analysis of the oxidation mechanism
Total
286
With the aim to better understand the reason for the observed
enantioselectivities of the oxidation reactions herein reported, we
performed DFT calculations in order to analyze the transition struc-
tures leading to (R)- and (S)-enantiomers. As shown above, best re-
sults in terms of selectivity were obtained for the 4-
halosubstituted isothiazole compounds 1d and 1e (ee: 65%), while
the 5-halosubstituted compounds 1g and 1f show scarce enantiose-
lectivity (ee: 6% and 5%, respectively). For these reasons, the reac-
tions of oxaziridine (+)-6b with 4-bromo-isothiazole 1e and 5-
bromoisothiazole 1f, respectively, are chosen as the model systems.
The model is further simplified by replacing the benzyl substituent
with a computationally affordable CH3 group, as the experimental
evidences show that the substitution pattern of the exocyclic nitro-
gen plays a little role in the stereochemical outcome of the oxidation
reaction. Accordingly, in the following discussion reactants and
products will be referred as M1e–1f and M2e–2f, respectively. As re-
ported in Section 4.1, calculations are performed at both the B3LYP/
6-31G(d) and B3LYP/6-31G(d,p),S(3df) levels of theory and the fol-
lowing discussions are referred to the latter method. Both the endo
and exo approaches of the isothiazoles to the oxaziridine (+)-6b were
evaluated, but only the structures corresponding to the exo transi-
tion state actually converged. This result can be explained by taking
into account the huge steric hindrance of the oxaziridine ethylene
bridge. The obtained transition state geometries, depicted in Figure
5, are in line with the ‘planar’ transition state models for the oxygen
transfer from an oxaziridine to a sulfoxide, ethylene or H2S obtained
at the HF/4-31G level by Bach et al.14 However, concerning the here-
in described transition states, located at both the B3LYP/6-31G(d)
and B3LYP/6-31G(d,p),S(3df) levels, some differences are observable
for the oxygen transfer mechanism.
Table 6
Relative-free energies with respect to the most stable conformer (kcal molꢀ1),
populations, [
values (deg [dm g/cm3]ꢀ1) obtained at TDDFT/B3LYP/aug-cc-pVDZ
level for each conformer of (S)-methylaminoisothiazole 1-oxide in chloroform
a
]
D
D
G
% Pop
Optical rotation
Calcd
Scaled
Conf. 1
Conf. 2
0
2.25
98
2
+213
ꢀ83
+209
ꢀ1.7
Total
207
point of view, it can be observed that the theory predicts similar
values in ethanol and methanol (276 and 286, respectively) whilst
the experimental figures are smaller (165 and 138, respectively). In
other words, it seems that the present solvation model overesti-
mates the effect of the alcoholic solvents and cannot reproduce
the experimental difference between the experimental values in
ethanol and methanol; (vi) by contrast, the agreement in chloro-
form results is excellent. To conclude this part, we can state that
even if the PCM solvation model does not afford a fully quantitative
agreement with the experiment, it reproduces the measured opti-
cal rotations in sign and order of magnitude and that the absolute
configuration of (+)-3-methylaminoisothiazole 1-oxide 2c is (S).
We decided to try to further support the above-mentioned correla-
tion by means of the TDDFT/B3LYP/aug-cc-pVDZ calculation of the
ECD spectrum of this compound in the range 380–200 nm, in both
the gas phase and ethanol.

2252
A. Casoni et al. / Tetrahedron: Asymmetry 20 (2009) 2247–2256
Figure 4. Experimental (black line) ECD spectrum of (+)-3-methylaminoisothiazole 1-oxide and predicted (TDDFT/B3LYP/aug-cc-pVDZ, velocity form) for (S)-3-
methylisothiazole 1-oxide 2c in gas phase (green line) and ethanol (red line).
Indeed, the analysis of geometrical parameters of the planar
transition state described by Bach for the oxidation of H2S is evi-
dence that the C–O and N–O oxaziridine bond cleavage occurs in
a quite synchronous manner, as C–O and N–O distances resulted
in 2.003 and 1.988 Å, respectively. On the other hand, the bond
length reported in Table 7 reveals unequivocally that the oxygen
transfer occurs asynchronously, with an initial N–O and a subse-
quent C–O cleavage. This behaviour could be due either to the ste-
ric and electronic properties of our specific structures, or to the
different chosen theoretical level, as it includes electron correlation
and a more extended basis set, particularly important for sulfur-
containing structures. Seeking for an answer to the above-men-
tioned question we optimized at the B3LYP/6-31G(d,p),S(3df) level
the transition state model for the oxygen transfer from an unsub-
stituted oxaziridine to H2S, starting from both the ‘spiro’ and the
‘planar’ orientations reported by Bach. Both structures converged
to a unique geometry, represented in Figure 6, perfectly compara-
ble with the other transition states herein located, with N–O and
C–O distances of 2.116 and 1.608 Å, respectively. The analysis of
the vibrational motion associated with the unique imaginary fre-
quency further confirmed the observed asynchronicity of this oxi-
dation reaction.
Coming back to the main objective of this computational study,
we will provide a possible explanation for the different enantiose-
lectivities observed within the oxidation of 4- and 5-halosubstitut-
ed isothiazoles.
The activation energies reported in Table 7 show that TS-
(proS)M2e is favoured over TS-(proR)M2e by 2.3 kcal/mol, while
a decidedly lower energy difference is computed between TS-
(proS)M2f and TS-(proR)M2f, as the former is favoured by only
0.2 kcal/mol. We have initially found such results quite surprising,
as we thought that TS-(proR)M2e should have been stabilized by
an attractive interaction between the C5–H and the SO2 moiety
(see Fig. 6). Indeed, HꢁꢁꢁO and C5ꢁꢁꢁO distances and C5–H–O angle
(2.954, 3.668 Å and 123.9 deg, respectively) measured for TS-(pro-
R)M2e are perfectly comparable with the weak hydrogen interac-
tion parameters reported by the literature.28a,b
However, as one can see from both Table 7 and Figure 5, such a
weak hydrogen bond causes the oxaziridine oxygen to approach
the isothiazole sulfur in a decidedly distorted manner (C5–S–O
and N2–S–O angles measure 88.8 and 114.4 deg, respectively) if
it is compared to the ‘ideal’ transition state structure represented
by the simplified model in Figure 6, thus provoking the observed
energetic raise. On the other hand, the oxygen approach in TS-
(proS)M2e is decidedly favoured by the geometrical point of view.
This behaviour further confirms that weak hydrogen interactions
in transition states are sometimes critical in determining the ste-
reo/regiochemical outcome of chemical reactions, as already ob-
served for Diels–Alder and hetero-Diels–Alder cycloadditions as
well as other coupling reactions.29,30 On the other hand, isothiazole
M2f, being substituted at C-5 by an halogen atom, is not able to
interact with the SO2 moiety, thus the oxaziridine oxygen ap-
proaches the sulfur atom in a quite symmetric manner with re-
spect to the isothiazole ring in both TS-(proS) and TS-(proR)M2f,
thus explaining a similar computed energy. It should be noted that
in TS-(proR)M2f the S–O forming bond is slightly distorted if com-
Figure 5. Transition states leading to (S)-M2e, (R)-M2e, (S)-M2f and (R)-M2f.

A. Casoni et al. / Tetrahedron: Asymmetry 20 (2009) 2247–2256
2253
Table 7
Selected geometrical parameters (Å, deg) and imaginary frequencies (cmꢀ1) of located transition states, together with ZPE corrected activation energies (kcal/mol)a
Compd
S–O
C–O
N–O
C5–S–O
N2–S–O
IF
D
Eà
TS-(proS)M2e
TS-(proR)M2e
TS-(proS)M2f
TS-(proR)M2f
1.918
1.931
1.923
1.944
1.467
1.496
1.469
1.477
2.051
2.043
2.049
2.045
92.6
88.8
90.5
91.2
97.4
114.4
97.4
ꢀ359.54
ꢀ398.56
ꢀ371.47
ꢀ381.68
22.7
25.0
22.7
22.9
104.5
a
Results from B3LYP/6-31G(d,p),S(3df) calculations are herein reported. The sum of isolated reactants’ energies is taken as reference.
pared with the (proS) structure, which indeed is favoured in terms
of relative energy, in perfect concordance with the observed exper-
imental results.
4. Experimental section
4.1. Theoretical calculations
Optical rotation analyses were carried out following the proto-
col described below. The starting geometries were generated by
conformational analysis with the SPARTAN02 software,31 using
molecular mechanics (MMFF94s force field) and retaining all
the structures differing by 2 kcal/mol in energy or less. All these
geometries were fully optimized at the B3LYP/6-31G* level,32,33
both in the gas phase and in the proper solvents. In all cases,
the free energy values at T = 298 K were employed to calculate
the population of each conformer, using the Boltzmann statistics.
The geometry optimization of (S)-3-methylaminoisothiazole S-
oxide conformers was repeated using the 6-31G** basis set and
the energy difference of the two structures became 2.13 kcal/
mol (vs 2.07 kcal/mole previously obtained using the 6-31G* basis
set), thus obtaining populations of 97.3% and 2.7% (previously 97%
and 3%). Such a small difference in population does not justify the
use of the larger basis set. The calculations of the optical rotatory
power were carried out at TDDFT/B3LYP/aug-cc-pVDZ level both
in the gas phase and in solution.34 The theoretical values of opti-
cal rotation (to be compared with the experimental ones) were
obtained as weighted averages on the Boltzmann populations cal-
culated in the gas phase or in the selected solvents. All the calcu-
lations of geometries, populations and optical rotatory power for
solvated systems were obtained using a continuum solvation
model, namely the Integral Equation Formalism version of the
Polarizable Continuum Model (IEF-PCM).26 In this framework,
the solvent is represented as a polarizable continuum dielectric
and the solute is assumed to be embedded in a molecular cavity
of proper shape and dimension within such a dielectric. By intro-
ducing a proper surface charge distribution on the cavity surface
representing the response of the solvent to the presence of the
solute, it will be possible to quantify solvent induced changes
both in the solute geometry and in its optical rotatory power.24e
3. Conclusion
In conclusion, we have described results of the oxidation of pro-
chiral sulfur of 3-aminosubstituted isothiazoles affording new chi-
Figure 6. Orthogonal views of the simplified transition state model for the oxygen
transfer, obtained at the B3LYP/6-31G(d,p),S(3df) level. Distances are reported in Å.
ral 3-aminosubstituted isothiazole S-oxides by two different
approaches consisting in the oxidation of (i) an achiral compound
by way of an enantiopure oxaziridine which allowed to obtain the
sulfoxide derivatives with good ee, (ii) an enantiopure compound
which did not give interesting results in terms of diastereoselec-
tion but offered the possibility to isolate the two diastereomers
through chromatographic separation and crystallization. Among
the different oxaziridines tested, the reagent of choice is (+)-
((8,8-dichlorocamphoryl)sulfonyl)oxaziridine (+)-6b, which is
very efficient under microwave irradiation. To the best of our
knowledge, the binomial camphorylsufonyloxaziridine-micro-
wave irradiation has never been used in isothiazole oxidation
until now. The influence on the stereochemical outcome of
the reaction exerted by the substituents was also considered.
We found that the presence of a halogen as substituent on
C-4 of the heterocyclic ring increases the enantiomeric excess,
which does not occur when the halogen is on C-5. The ob-
served differences in terms of enantioselectivity are explained
with the aid of theoretical calculations. Indeed, it is found that
the formation of a weak hydrogen bond between the isothiaz-
ole H-5 and the oxygen of the camphorylsulfonyloxaziridine,
observed only for the transition state leading to the (R)-4-halo-
substituted isothiazole 1-oxide, destabilizes the transition state
itself and thus the (S)-stereoisomer corresponds to experimen-
tal findings. Thus, a new interpretation of the oxidation reac-
tion mechanism has been provided on the basis of B3LYP/6-
31G(d,p),S(3df) calculations. The absolute configuration to the
stereogenic sulfur atom was assigned by a comparison of the
experimental optical activities and ECD measurements with
those theoretically predicted at the TDDFT/B3LYP/aug-cc-pVDZ
level of theory.
The ECD calculations were carried out following
a
known
procedure.35
TS-(proS)M2e,
TS-(proR)M2e,
TS-(proS)M2f and
TS–
(proR)M2f, together with the corresponding reactants and prod-
ucts, were initially optimized in the gas phase using the B3LYP
functional and the 6-31G* basis set. Vibrational frequencies were
computed at the same level of theory in order to define optimized
geometries as minima (no imaginary frequencies) or transition
states (a unique imaginary frequency corresponding to the vibra-
tional stretching of the forming/breaking bonds) and to calculate
ZPVE and thermochemical corrections to electronic energies
(1 atm, 298.15 K). It was reported that an adequate description
of the geometry and electronic properties of sulfur-containing
compounds would require the inclusion of supplementary d
functions in the basis set,36 thus all stationary points were
reoptimized at the B3LYP/6-31G(d,p),S(3df) level and confirmed
by a vibrational analysis at the same level of theory. Geometries
and relative energies within the two theoretical levels were quite
comparable: energy difference between TS-(proS) and TS-(pro-

2254
A. Casoni et al. / Tetrahedron: Asymmetry 20 (2009) 2247–2256
R)M2e resulted ꢀ1.8 and ꢀ2.3 kcal/mol with the former and the
latter more accurate method, respectively, while the relative en-
ergy between TS-(proS) and TS-(proR)M2f resulted ꢀ0.5 and
ꢀ0.3 kcal/mol at the B3LYP/6-31G(d) and B3LYP/6-31G(d,p),S(3df)
levels, respectively. All QM calculations were performed with the
GAUSSIAN 03 package.37
C10H9BrN2S (269.06): C, 44.62; H, 3.37; N, 10.41. Found: C, 44.27;
H, 3.10; N, 10.02.
4.3.3. (S)-3-(1-Phenylethyl)-isothiazol-amine 1a
Yield: 42%. Mp 127–128 °C (white powder from CH2Cl2/Et2O). IR
3300 cmꢀ1 (NH). 1H NMR (CDCl3): d = 1.56 (d, J = 6.6 Hz, 3H; CH3),
4.88 (m, 1H; CH), 4.93 (m, 1H; NH), 6.33 (d, J = 4.8 Hz, 1H; H-4),
7.20–7.41 (m, 5H; ArH), 8.31 (d, J = 4.8 Hz, 1H; H-5) 13C NMR
(CDCl3): d = 24.1 (CH3), 53.8 (CH), 111.6 (C-4), 126.2, 127.3, 128.8
(ArCH), 145.0 (ArC), 147.7 (C-5), 165.5 (C-3). MS (EI) m/z = 205
(M+, 100). Anal. Calcd for C11H12N2S (204.29): C, 64.67; H, 5.92;
N, 13.71. Found: C, 64. 89; H, 5.86; N 13.56.
1H NMR spectra were obtained in the solvents indicated with
200 or 300 MHz instruments. Coupling constants (J) are given in
hertz. Mass spectra were obtained by electron impact ionization
at 70 eV using the direct exposure probe (DEP).
Compounds 1b,c,g16–18,38 4,39 5,10 and 6c–d,11 were prepared
according to the procedure given in the literature. Compounds
6a,b are commercially available.
4.3.4. Synthesis of 2a
4.2. Synthesis of 3-aminosubstituted isothiazoles
Equimolecular amounts (0.56 mmol) of 1a and 4 were dissolved
under stirring in CH2Cl2 (20 mL) at room temperature and the pro-
gress of the reaction was checked by TLC (AcOEt; 12 h). The solvent
was evaporated under reduced pressure at room temperature. The
residue was chromatographed on silica gel (eluent: cyclohexane/
AcOEt 100:0 to 0:100) affording the two diastereoisomers (S,S)-
2a and (R,S)-2a. Total yields 80%.
4.2.1. N-Benzyl-5-bromo-isothiazol-3-amine 1f
Compound 1b (1.04 mmol) was dissolved in anhydrous THF
(15 mL) and the solution cooled at ꢀ60 °C under nitrogen. Next,
LDA (2.10 mL of a 2 M solution in THF) was added and after about
5–10 min. solid BrCN (2.08 mmol) was added and stirred. At the
end of the reaction (TLC cyclohexane/ethyl acetate 3:1), the solvent
was evaporated and the mixture was taken up with CH2Cl2 (30 mL)
and washed with water (3 ꢂ 10 mL). The organic phase was dried
with Na2SO4, filtered and evaporated to dryness affording an im-
pure oil, which was purified by column chromatography (cyclo-
hexane/ethyl acetate 100:0 to 0:100) affording pure 1f. Yield:
51%, mp 97–98 °C (yellow powder from Et2O/pentane). IR
3340 cmꢀ1 (NH). 1H NMR (CDCl3): d = 4.51 (s, 2H; CH2), 4.74 (br
s, 1H; NH), 6.46 (s, 1H; H-4), 7.26–7.36 (m, 5H; ArH) 13C NMR
(CDCl3): d = 47.7 (CH2), 115.3 (C-4), 127.7, 127.8, 128.9 (ArCH),
134.9 (ArC), 139.0 (C-5), 165.1 (C-3). MS (EI) m/z = 269.4 (M+,
100). Anal. Calcd for C10H9BrN2S (269.06): C, 44.62; H, 3.37; N,
10.41. Found: C, 44.45; H, 3.54; N, 10.16.
4.4. General procedure for the oxidation of 1a,b,c.d,e,g with (+)
and (ꢀ)-6a,b
Equimolecular amounts (0.24 mmol) of the opportune 3-amino-
isothiazole and (+)- or (ꢀ)-8a,b were dissolved in CH2Cl2 (5 mL)
and irradiated with microwaves (400 W, 80 °C) for the time indi-
cated in Tables 3 and 4. The solvent was evaporated under reduced
pressure and the residue was chromatographed on silica gel (elu-
ent: cyclohexane/AcOEt 100:0 to 0:100) affording pure sulfoxides
2a,b,c.d,e,g. Compounds (S,S)-2a and (S,R)-2a can be obtained with
98% purity (chiral HPLC analyses) after column chromatography
and crystallization with H2O/MeOH.
Yields, ee% and de % are reported in Tables 3 and 4.
4.3. General procedure for the preparation of isothiazolamines
1a,d,e
4.4.1. 3-(S)-(1-Phenylethyl)aminoisothiazol-(S)-1-oxide (S,S)-
2a40
The opportune isothiazole 3-one16 was dissolved in POCl3
(3 mL for each gram of reagent) under stirring at room temper-
ature. When the reagent was completely consumed (TLC ethyl
acetate, about 2–24 h), the solvent was evaporated under re-
duced pressure. Diethylether was added to the mixture
(2 ꢂ 5 mL) at 0 °C and a deep orange oil was separated. Decan-
tation of the solvent afforded a gum perfectly suitable for the
next step. This gum was suspended in CH3CN (30 mL) and
NH3 gas was bubbled for about 1 h under stirring at 0 °C. The
reaction mixture was brought to room temperature and the sol-
vent was evaporated under reduced pressure. The residue was
purified by column chromatography (AcOEt/cyclohexane 0:100
to 100:0).
Mp 149–151 °C dec (after recrystallization from MeOH/H2O:
98% purity). 1H NMR (CDCl3): d = 1.63 (d, J = 7 Hz, 3H; CH3), 5.26
(m, 1H; CH), 6.39 (m, 1H; NH), 6.74 (d, J = 5.5 Hz, 1H; H-4), 7.26–
7.37 (m, 5-H: ArH), 7.72 (d, J = 5.5, 1H; H-5) 13C NMR (CDCl3):
d = 22.09 (CH3), 53.08 (CH), 126.63, 127.94, 129.01 (ArCH),
129.25 (C-4; CH), 142.32 (ArC), 155.92 (C-5; CH), 164.87 (C-3;
Cq). MS (EI) m/z = 221 (M+, 100) Anal. Calcd for C11H12N2OS
(220.07): C, 59.97; H, 5.49; N, 12.72. Found: C, 59. 75; H, 5.76; N,
12.46. [
a
]
D = ꢀ133.3 (c 1, CHCl3 calculated on a sample with de
74.64%) HPLC 20 lL loop on Chiralcel ODH, (isopropanol/hexane
30:70), flow: 1 mL minꢀ1, retention time 8.89 min.
4.4.2. 3-(S)-1-Phenylethyl)aminoisothiazole-(R)-1-oxide (S,R)-2a
Mp 138–140 °C dec 1H NMR (CDCl3): d = 1.58 (d, J = 7 Hz, 3H;
CH3), 5.24 (m, 1H; CH), 6.43 (t, 1H; NH), 6.70 (d, J = 5.1 Hz, 1H;
H-4), 7.26–7.34 (m, 5-H: ArH), 7.62 (d, J = 5.1, 1H; H-5) 13C NMR
(CDCl3): d = 21.56 (CH3), 53.07 (CH), 126.72, 128.13, 129.02 (ArCH),
128.70 (C-4; CH), 142.12 (ArC), 155.74 (C-5; CH), 164.38 (C-3; Cq).
MS (EI) m/z = 219.2 (M+, 100). Anal. Calcd for C11H12N2OS (220.07):
C, 59.97; H, 5.49; N, 12.72. Found: C, 60.19; H, 5.80; N, 12.40.
4.3.1. N-Benzyl-4-chloro-isothiazol-3-amine 1d
Yield: 40%. Mp oil. IR 3418 cmꢀ1 (NH). 1H NMR (CDCl3): d = 4.65
(d, J = 4.4 Hz, 2H; CH2), 5.07 (br s, 1H; NH), 7.26–7.40 (m, 5H; ArH),
8.24 (s, 1H; H-5). 13C NMR (CDCl3): d = 47.8 (CH2), 110.1 (C-4),
127.9, 128.2, 129.1 (ArCH), 139.1 (ArC), 141.9 (C-5), 160.5 (C-3).
Anal. Calcd for C10H9ClN2S (224.61): C, 53.45; H, 4.04; N, 12.47.
Found: C, 53.68; H, 4.14; N, 12.20.
[
a
]
D = ꢀ353.6 (c 1, CHCl3, calculated on a sample with de 82.30%)
HPLC 20 lL loop on on Chiralcel ODH, (isopropanol/hexane
4.3.2. N-Benzyl-4-bromo-isothiazol-3-amine 1e
30:70), flow: 1 mL minꢀ1, retention time 6.27 min.
Yield: 55%. Mp yellow oil. IR 3415 cmꢀ1 (NH). 1H NMR (CDCl3):
d = 4.63 (d, J = 6.2 Hz, 2H; CH2), 5.25 (br s, 1H; NH), 7.22–7.42 (m,
5H; ArH), 8.40 (s, 1H; H-5). 13C NMR (CD3COCD3): d = 46.9 (CH2),
95.4 (C-4), 127.0, 127.7, 127.4 (ArCH), 140.3 (ArC), 145.5 (C-5),
161.9 (C-3). MS (EI) m/z = 269 (M+, 100). Anal. Calcd for
4.4.3. N-Benzyl-isothiazol-3-amine S-oxide 2b
Mp 94 °C (powder from CH2Cl2/Et2O). IR 3208 cmꢀ1 (NH). 1H
NMR (CD3COCD3): d = 4.70 (d, J = 5.9 Hz, 2H; CH2), 7.04 (d,

A. Casoni et al. / Tetrahedron: Asymmetry 20 (2009) 2247–2256
2255
J = 5.5 Hz, 1H; H-4), 7.30–7.47 (m, 5H; ArH), 7.94 (br s, 1H; NH),
8.02 (d, J = 5.5 Hz, 1H; H-5) 13C NMR (CD3COCD3): d = 45.6 (CH2),
97.2 (C-4), 126.6–127.7 (ArCH), 137.1 (ArC), 157.0 (C-5), 164.2
(C-3). MS (EI) m/z = 207 (M+, 100). Anal. Calcd for C10H10N2OS
(206.26): C, 58.25; H, 4.85; N, 13.59. Found: C, 56.29; H, 5.18; N,
(c 1, CHCl3, calculated on a sample with ee 65%) HPLC separation
of (R)- and (S)-2e: 20 lL loop on Chiralcel ODH, (isopropanol/hex-
ane 30:70), flow: 1 mL minꢀ1, retention time (R)-2e 12.25 min; (S)-
2e 15.64 min.
12.96. [
39.72%); [
a
]
D = +36.6 (c 1, CHCl3, calculated on a sample with ee
D = +50.6 (c 1, CHCl3, calculated on a sample with ee
Acknowledgements
a]
55%); HPLC separation of (R)- and (S)-2a: 20
lL loop on Chiralcel
The authors thank the Ministero dell0Istruzione, dell0Università
e della Ricerca (MIUR) for the financial support (First and PRIN) and
the Centro Interuniversitario Lombardo per l’Elaborazione Autom-
atica (CILEA) for providing computational facilities.
ODH, (isopropanol/hexane 30:70), flow: 1 mL minꢀ1, retention
time (R)-2a 7.74 min; (S)-2a 10.78 min.
4.4.4. N-Methyl-isothiazol-3-amine S-oxide 2c
Mp 119 °C (powder from CH2Cl2/Et2O). IR 3225 cmꢀ1 (NH). 1H
NMR (CD3OD): d = 3.05 (s, 3H; CH3), 6.96 (d, J = 5.3 Hz, 1H; H-4),
7.99 (d, J = 5.3 Hz, 1H; H-5) 13C NMR (CD3OD): d = 28.7 (CH3),
128.7 (C-4), 156.0 (C-5), 167.8 (C-3). MS (EI) m/z = 130 (M+, 100).
Anal. Calcd for C4H6N2OS (130): C, 36.91; H, 4.65; N, 21.52. Found:
References
1. (a) Clerici, F.; Gelmi, M. L.; Pellegrino, S. Isothiazoles. In Comprehensive in
Heterocyclic Chemistry III; Katritzky, A. R., Ramsden, C. A., Scriven, E. F. V., Eds.;
Elsevier: Oxford, UK, 2008; Vol. 4, pp 545–633. Chapter 4.05; (b) Chapman, R.
F.; Peart, B. J. Isothiazoles. In Comprehensive in Heterocyclic Chemistry II;
Katritzky, A. R., Rees, C. W., Scriven, E. F. V., Eds.; Pergamon Press: Oxford, 1996;
Vol. 3. Chapter 3.07; (c) Khalaj, A.; Adibpour, N.; Shahverdi, A. R.; Daneshtalab,
M. Eur. J. Med. Chem. 2004, 39, 699–705; (d) Zani, F.; Vicini, P.; Incerti, M. Eur. J.
Med. Chem. 2004, 39, 135–140; (e) Da Settimo, F.; Primofiore, G.; La Motta, C.;
Sartini, S.; Taliani, S.; Simorini, F.; Marini, A. M.; La Vecchia, A.; Novellino, E.;
Boldrini, E. J. Med. Chem. 2005, 48, 6897–6907.
2. Clerici, F.; Lo Presti, L.; Gelmi, M. L.; Soave, R. Tetrahedron 2002, 58, 5173–5178.
and references cited therein.
3. Clerici, F.; Contini, A.; Corsini, A.; Ferri, N.; Grzesiak, S.; Pellegrino, S.; Sala, A.;
Yokoyama, K. Eur. J. Med. Chem. 2006, 41, 675–682.
4. Clerici, F.; Gelmi, M. L.; Yokoyama, K.; Pocar, D.; Van Voorhis, W. C.; Buckner, F.
S.; Gelb, M. H. Bioorg. Med. Chem. Lett. 2002, 12, 2217–2220.
5. Clerici, F.; Corsini, A.; Ferri, N.; Gelb, M. H.; Yokoyama, K.; Pocar, D. Biochem.
Pharmacol. 2005, 70, 1735–1743.
6. Casoni, A.; Clerici, F.; Contini, A.; Pellegrino, S.; Sala, A. Lett. Org. Chem. 2008, 5,
623–627.
C, 36.57; H, 4.78; N, 21.33. [
+69 (c 1, methanol), ee 49.6%. HPLC separation of (R)- and (S)-2c:
a
]
D = ꢀ60 (c 1, methanol), ee 43.6% and
20 lL loop on Chiralcel OF, (isopropanol/hexane 35:65), flow:
1 mL minꢀ1, retention time (R)-2c 27.9 min; (S)-2c 36.6.
4.4.5. N-Benzyl-5-chloro-isothiazol-3-amine S-oxide 2g
Mp 127–129 °C (powder from CH2Cl2/Et2O). IR 3053 cmꢀ1
(NH). 1H NMR (CDCl3): d = 4.50–4.74 (m, 2H; CH2), 6.74 (s, 1H;
H-4), 7.08 (bs, 1H; NH), 7.28–7.39 (m, 5H; ArCH) 13C NMR
(CDCl3): d = 46.8 (CH2), 122.7 (C-4), 128.4, 129.1 (ArCH), 136.5
(ArC), 163.9, 165.5 (C-5, C-3). MS (EI) m/z = 240 (M+, 100). Anal.
Calcd for C10H9ClN2OS (240.45): C, 49.90; H, 3.77; N, 11.64.
Found: C, 49.73; H, 3.82; N, 11.31. [
a
]
D = ꢀ5 (c 1, CHCl3) ee
7. Pellissier, H. Tetrahedron 2007, 63, 1297–1330.
8. Waldner, A. Tetrahedron Lett. 1989, 30, 3061–3064.
9. Davis, F. A.; Sheppard, A. C. Tetrahedron 1989, 45, 5703–5742.
10. Davis, F. A.; Reddy, R. T.; Han, W.; Carrol, P. J. J. Am. Chem. Soc. 1992, 114, 1428–
1437.
5.53% HPLC separation of (R)- and (S)-2g: 20
l
L loop on Chiralcel
ODH, (isopropanol/hexane 30:70), flow: 1 mL minꢀ1, retention
time (R)-2g 7.08 min; (S)-2g 9.72 min.
11. Davis, F. A.; Kumar, A.; Chen, B.-C. J. Org. Chem. 1991, 56, 1143–1145.
12. Davis, F. A.; Billmers, J. M.; Gosciniak, D. J.; Towson, J. C.; Bach, R. D. J. Org.
Chem. 1986, 51, 4240–4245.
13. Davis, F. A.; Weismiller, M. C.; Murphy, C. K.; Reddy, R. T.; Chen, B.-C. J. Org.
Chem. 1992, 57, 7274–7285.
14. Bach, R. D.; Wolber, G. J. Am. Chem. Soc. 1984, 106, 1410–1415.
15. Bach, R. D.; Coddens, B. A.; Mc Dovall, J. J.; Schlegel, H. B.; Davies, F. A. J. Org.
Chem. 1990, 55, 3325–3330.
16. Bell, A. S.; Fishwick, C. W. G.; Reed, J. E. Tetrahedron 1999, 55, 12313–12330.
17. Lewis, S. N.; Miller, G. A.; Hausman, M.; Szamborski, E. C. J. Heterocycl. Chem.
1971, 8, 571–580.
4.4.6. N-Benzyl-4-chloro-isothiazol-3-amine S-oxide 2d
Mp 143–144 °C (powder from CH2Cl2/Et2O). IR 3247 cmꢀ1 (NH).
1H NMR (DMSO-d6): d = 4.50–4.55 (m, 2H; CH2), 7.20–7.40 (5H;
ArH), 8.52 (s, 1H; H-5), 9.02 (br s, 1H; NH) 13C NMR (DMSO-d6):
d = 46.7 (CH2), 127.9, 128.2, 129.0 (ArCH), 131.0 (C-4), 138.4
(ArC), 152.3 (C-5), 160.9 (C-3) MS (EI) m/z = 241 (M+, 100). Anal.
Calcd for C10H9ClN2OS (240.45): C, 49.90; H, 3.77; N, 11.64. Found:
C, 50.23; H, 3.53; N, 11.31. HPLC separation of (R)- and (S)-2d:
18. Weiler, E. D.; Petigara, R. B.; Wolfersberger, M. H.; Miller, G. A. J. Heterocycl.
Chem. 1977, 14, 627–630.
19. Bunch, L.; Krogsgaard-Larsen, P.; Madsen, U. J. Org. Chem. 2002, 67, 2375–
2377.
20 lL loop on Chiralcel OD, (isopropanol/hexane 3:6), flow:
1 mL minꢀ1, retention time (R)-2d 10.9 min; (S)-2d 15.0.
4.4.7. N-Benzyl-5-bromo-isothiazol-3-amine S-oxide 2f
20. Katritzky, A. R.; Laurenzo, K.; Relyea, D. Can. J. Chem. 1988, 66, 1617–1624.
21. (a) Polavarapu, P. L. Mol. Phys. 1997, 91, 551–554; (b) Polavarapu, P. L.
Tetrahedron: Asymmetry 1997, 8, 3397–3401; (c) Polavarapu, P. L. Chirality
2002, 14, 768–781.
22. (a) Kondru, R. K.; Wipf, P.; Beratan, D. N. J. Am. Chem. Soc. 1998, 120, 2204–
2205; (b) Kondru, R. K.; Lim, S.; Wipf, P.; Baratan, D. N. Chirality 1997, 9, 469–
477.
23. (a) Pecul, M.; Ruud, K. Adv. Quantum. Chem. 2006, 50, 185–227; (b) Crawford, T.
D. Theor. Chem. Acc. 2006, 115, 227–245.
24. Very recent examples of assignment of absolute configuration by the ab initio
Mp 123–124 °C (powder from CH2Cl2/Et2O). IR 3225 cmꢀ1 (NH).
1H NMR (CDCl3): d = 2.07 (br s, 1H; NH), 4.55–4.62 (m, 2H; CH2),
6.89 (s, 1H; H-4), 7.19–7.32 (m, 5H; ArH) 13C NMR (CDCl3):
d = 46.8 (CH2), 127.7 (C-4), 127.2, 128.3, 129.0 (ArCH), 136.4
(ArC), 151.3 (C-5), 166.4 (C-3). MS (EI) m/z = 285.3 (M+, 100). Anal.
Calcd for C10H9BrN2SO (285.05): C, 42.12; H, 3.18; N, 9.82. Found:
C, 41.94; H, 3.28; N, 9.68. [a]D = +6.0 (c 1, CHCl3) ee 6.4% HPLC sep-
calculation of [
a] and ECD spectra are: (a) Stephens, P. J.; McCann, D. M.;
D
aration of (R)- and (S)-2f: 20 lL loop on Chiralcell OD, (isopropanol/
Butkus, E.; Stoncius, S.; Cheeseman, J. R.; Frisch, M. J. J. Org. Chem. 2004, 69,
1948–1958; (b) Stephens, P. J.; McCann, D. M.; Devlin, F. J.; Cheeseman, J. R.;
Frisch, M. J. J. Am. Chem. Soc. 2004, 126, 7514–7521; (c) Sebert, S.; Konig, G. M.;
Voloshina, E.; Raabe, G.; Fleischhauer, J. Chirality 2006, 18, 413–418; (d)
Stephens, P. J.; Pan, J. J.; Devlin, F. J.; Cheesman, J. R. J. Nat. Prod. 2008, 71, 285–
2888; (e) Mennucci, B.; Claps, M.; Evidente, A.; Rosini, C. J. Org. Chem. 2007, 72,
6680–6691; (f) Hussain, H.; Akhtar, N.; Draeger, S.; Schulz, B.; Pescitelli, G.;
Salvadori, P.; Antus, S.; Kurtan, T.; Krohn, K. Eur. J. Org. Chem. 2009, 749–756.
and references cited therein.
hexane 3:7), flow: 1 mL minꢀ1, retention time (R)-2f 6.92 min; (S)-
2f 8.34 min.
4.4.8. N-Benzyl-4-bromo-isothiazol-3-amine S-oxide 2e
Mp 158 °C (after recrystallization from MeOH/H2O: 98% purity).
IR 3220 cmꢀ1 (NH). 1H NMR (CD3COCD3): d = 4.73 (d, J = 5.1 Hz, 2H;
CH2), 7.30–7.48 (m, 5H; ArH), 7.98 (br s, 1H; NH), 8.40 (s, 1H; H-4).
13C NMR (CD3COCD3): d = 47.5 (CH2), 120.1 (C-4), 128.2, 128.7,
129.5 (ArCH), 138.6 (ArC), 156.8 (C-5), 161.7 (C-3). MS (EI) m/
z = 287 (M+). Anal. Calcd for C10H9BrN2SO (285.05): C, 42.12; H,
25. (a) Stephens, P. J.; Devlin, F. J.; Cheeseman, J. R.; Frisch, M. J. J. Phys. Chem. A
2001, 105, 5356–5371; (b) Stephens, P. J.; McCann, D. M.; Cheeseman, J. R.;
Frisch, M. J. Chirality 2005, 17, S52–S64. and references cited therein.
26. Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105, 2999–3093. and
references cited therein.
3.18; N, 9.82. Found: C, 42.34; H, 3.25; N, 9.57. [
(c 1, CHCl3, calculated on a sample with ee 55.76%); [
a
a
]
D = +130.95
D = +152.6
27. Autschbach, J.; Jensen, L.; Schatz, G. C.; Electra Tse, Y. C.; Krykunov, M. J. Phys.
Chem. A 2006, 110, 2461–2473.
]

2256
A. Casoni et al. / Tetrahedron: Asymmetry 20 (2009) 2247–2256
28. (a) Desiraju, G. R. Chem. Commun. 2005, 2995–3001; (b) Muthuraman, M.; Le
Fur, Y.; Bagieu-Beucher, M.; Masse, R.; Nicoud, J. F.; Gorge, S.; Nangia, A.;
Desiraju, G. R. J. Solid State Chem. 2000, 152, 221–228.
347–354; (c) Aversa, M. C.; Barattucci, A.; Bonaccorsi, P.; Contini, A. J. Phys. Org.
Chem. 2009. doi: 10.1002/poc.1557.
37. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann, R. E.;
Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M.
C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.;
Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.;
Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.;
Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.;
Replogle, E. S.; Pople, J. A. GAUSSIAN 03; Gaussian: Pittsburgh, PA, USA, 2003,
http://www.gaussian.com/.
29. Nishio, M. Tetrahedron 2005, 61, 6923–6950.
30. (a) Contini, A.; Leone, S.; Menichetti, S.; Viglianisi, C.; Trimarco, P. J. Org. Chem.
2006, 71, 5507–5514; (b) Borsini, E.; Broggini, G.; Contini, A.; Zecchi, G. Eur. J.
Org. Chem. 2008, 2808–2816.
31. SPARTAN ‘02, Wavefunction Inc., 18401 Von Karman Avenue, Suite 370, Irvine, CA
92612, 2002, http://www.wavefun.com/.
32. (a) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785–789; (b) Becke, A. D. J.
Chem. Phys. 1993, 98, 5648–5652.
33. Frisch, M. J.; Pople, J. A.; Binkley, J. S. J. Chem. Phys. 1984, 80, 3265–3269.
34. Woon, D. E.; Dunning, T. H., Jr. J. Chem. Phys. 1993, 98, 1358–1371.
35. Giorgio, E.; Tanaka, K.; Ding, W.; Krishnamurty, G.; Pitts, K.; Ellestad, G. E.;
Rosini, C.; Berova, N. Bioorg. Med. Chem. 2005, 13, 5072–5079.
36. (a) Roux, M. V.; Temprado, M.; Jimènez, P.; Dávalos, J. Z.; Notario, R.; Martín-
Valcárel, G.; Garrido, L.; Guzmán-Mejía, R.; Juaristi, E. J. Org. Chem. 2004, 69,
5454–5459; (b) Juaristi, E.; Notario, R.; Roux, M. V. Chem. Soc. Rev. 2005, 34,
38. Clerici, F.; Contini, A.; Gelmi, M. L.; Pocar, D. Tetrahedron 2003, 59, 9399–9408.
39. Davis, F. A.; Stringer, O. D. J. Org. Chem. 1982, 47, 1174–1175.
40. Basing on our results we assumed that the major diastereoisomer obtained by
using (+)-6b as the oxidant has an (S) configuration at the sulfur atom.