Privileged structures: Synthesis and structural investigations on tricyclic sulfonamides

Article abstract of DOI:10.1039/b9nj00279k

Customized synthetic procedures for the obtainment of highly functionalizable free sulfonamide tricyclic structures having NH, O, S and SO₂ (Z group) in the central seven-membered ring are presented. Due to the role that tricycles and the sulfonamide group play in medicinal chemistry, attention have been also paid to the 3D arrangement of these molecules. Their molecular structures and conformational properties have been studied by single-crystal X-ray diffraction and quantum chemical methods (HF-SCF/6-311+G(d,p). The minimum energy geometries in the solid state (I, Z = O; II, Z = NH and III, Z = S, SO₂) have been compared with the minimum-energy geometries in the gas phase (invariably the type III geometry) and discussed with respect to the literature reference values. An effort was made to correlate the conformational differences observed in the solid-state and gas-phase with the nature of the Z grouping.

Full text of DOI:10.1039/b9nj00279k

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Privileged structures: synthesis and structural investigations
on tricyclic sulfonamidesw
Maria Altamura,a Valentina Fedi,a Danilo Giannotti,a Paola Paoli*b and
Patrizia Rossib
Received (in Gainesville, FL, USA) 22nd June 2009, Accepted 4th August 2009
First published as an Advance Article on the web 1st September 2009
DOI: 10.1039/b9nj00279k
Customized synthetic procedures for the obtainment of highly functionalizable free sulfonamide
tricyclic structures having NH, O, S and SO2 (Z group) in the central seven-membered ring are
presented. Due to the role that tricycles and the sulfonamide group play in medicinal chemistry,
attention have been also paid to the 3D arrangement of these molecules. Their molecular
structures and conformational properties have been studied by single-crystal X-ray diffraction and
quantum chemical methods (HF-SCF/6-311+G(d,p). The minimum energy geometries in
the solid state (I, Z = O; II, Z = NH and III, Z = S, SO2) have been compared with the
minimum-energy geometries in the gas phase (invariably the type III geometry) and discussed
with respect to the literature reference values. An effort was made to correlate the conformational
differences observed in the solid-state and gas-phase with the nature of the Z grouping.
terms of both chemical and pharmacological versatility since
Introduction
they contain positions which are functionalizable in different
conditions. In particular the strongly acidic, easily extractable,
SO2NH hydrogen represents an easy access to the post
functionalization, and the activation to the substitution of
some aromatic positions operated by the sulfonamide function
renders these molecules an ideal platform for new derivatives.
In the present work we describe general methods for
the obtainment of highly functionalizable free sulfonamide
tricyclic structures (B), containing respectively NH, O, S and
SO2 in the central seven-membered ring. The unsubstituted
dibenzodithiazepine 5,5-dioxide structure (Fig. 2, Z = S) and
its oxidized product (Fig. 2, Z = SO2) are described here for
the first time and complete the series represented in Fig. 2. In
addition a structural investigation of all the monomers
sketched in Fig. 3 by both experimental (by single-crystal
X-ray diffraction) and theoretical (by quantum chemical
calculations) approaches is also presented. In particular,
attention has been focused on the conformation of the sulfon-
amide moiety and the present experimental data and what is
considered the standard conformation in sulfonamides are
compared and discussed. In fact, given that several bio
logically important compounds feature this functional group,7
we think that the comprehension of the conformational
The definition of the so-called ‘‘privileged structures’’ dates by
now twenty years.1 Amongst the molecular moieties recognized
as being able to provide high affinity ligands for many types of
receptors, tricycles (Fig. 1) play an important role both in the
field of the drugs for the central nervous system and in other
very different applications as reported recently.2 It is easy to
single out tricyclic structures of the general formula depicted
in Fig. 1, either in anticancer agents such as farnesyl protein
transferase agents, or in multidrug resistance modulators,
HDAC inhibitors,3 antiviral agents such as the non nucleoside
inhibitors of HIV reverse transcriptase, antihistaminic drugs
as loratadine or cyproheptadine.2
Amongst these structures, our research group has developed
over the past years a particular interest for the skeletons
containing an amide (A) or a sulfonamide function (B)
(Fig. 2).3,4
While many general methods of synthesis are available for
the skeleton containing the amide function in the central ring
(A),5 only a few focused examples are described for the
synthesis of the analogs containing the free sulfonamide
function (B).6 The lack of a good, general and reproducible
method of synthesis for these structures could have prevented
their use in a larger number of applications. Despite this, the
sulfonamide skeletons possess some peculiar properties in
a Chemistry Department, Menarini Ricerche S.p.A., Florence, Italy
b Department of Energy Engineering ‘‘Sergio Stecco’’, University of
Florence, Italy. E-mail: paolapaoli@unifi.it
w Electronic supplementary information (ESI) available: Crystal
packing of compounds. Tables S1–S6: Intermolecular hydrogen bonds
and p–p interactions in the crystal lattices; most relevant geometrical
parameters of the theoretically calculated structures. Scheme S1:
Schematic representation of the intermolecular interactions in the
compounds. CCDC reference numbers 745476–745481. For ESI and
crystallographic data in CIF or other electronic format see DOI:
10.1039/b9nj00279k
Fig. 1 Schematic drawing of a tricyclic skeleton.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 2219–2231 | 2219

View Online
Scheme 1 Sulfonamide syntheses from previous literature.
to the leaving group, and without the need to alkylate
the sulfonamide. The synthesis of these molecules is often
complicated by the potential formation of the corresponding
dimers; this side reaction can not be suppressed by dilution of
the reaction mixture. For each of the nuclei discussed here a
customized methodology is described.
Fig. 2 Examples of tricyclic structures featuring amide and sulfon-
amide functionalities.
behaviour of this functionality is one of the key steps in order
to clarify the way these species act at the molecular level.8
In addition the solid-state crystal structures of some of the
dimers as obtained by single-crystal X-ray diffraction analysis
is also presented.
Synthesis of dibenzoxathiazepine 5,5-dioxide. Only a few
examples of dibenzoxathiazepine 5,5-dioxide compounds were
synthesized in the past with different procedures.6,10 Our
general method to prepare these compounds is outlined in
Scheme 2. Reaction of substituted 2-aminophenols 1a–j
with 2-bromobenzenesulfonyl chloride 2, in pyridine, gave sulfon-
amides 3a–j, that were cyclized in N-methylpyrrolidone
(NMP) in the presence of cesium carbonate to the corresponding
tricyclic sulfonamides 4a–j. The structure of 4a was confirmed
by single-crystal X-ray analysis. Considering the drastic
experimental conditions (140 1C and 20 h of reaction time)
we evaluated the possible use of copper to promote the
reaction, as in the Ullmann ether synthesis for the formation
of diaryl ethers.11 However, when elemental copper was added
to the reaction mixture, only dimeric products were obtained.
In particular, in the case of the cyclization of sulfonamide 3b,
the dimer 5b was obtained, whose structure was confirmed by
X-ray crystal structure analysis.
Results and discussion
Synthesis
The few synthetic methods described in the literature for the
synthesis of sulfonamides (B) usually refer to the condensation
of phenylenediamines or 2-aminophenols with 2-chlorobenzene-
sulfonyl chlorides, where the halogen is easily substituted due
to the presence of a strong activator in position 5 such as NO2
(Scheme 1).6 Otherwise some examples are reported where the
SO2NH functionality has to be alkylated before inducing the
cyclization reaction.4,9
We present here the results of our investigations on the
possibility to induce cyclization without the need of electron-
withdrawing groups located on the para-position with respect
Synthesis of dibenzothiadiazepine 5,5-dioxides. The synthesis
of unsubstituted dibenzothiadiazepine 5,5-dioxides was evaluated
in various experimental conditions such as Ullmann-type
copper-mediated procedures and Pd-catalyzed cross-coupling
reactions. Some Pd-catalyzed cross-coupling experiments,
made according to the described intramolecular palladium-
catalyzed arylations,12 are ineffective in the synthesis of these
substances. Only the copper-catalyzed reactions gave rise to
Fig. 3 Schematic drawings of the monomer molecules studied by
single-crystal X-ray diffraction (4a, 8b, 13a, 14a) and by computational
methods (4a, 8a, 13a, 14a). Inset drawing with the most relevant atom
labelling.
Scheme 2 Synthesis of dibenzoxathiazepine 5,5-dioxide.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
2220 | New J. Chem., 2009, 33, 2219–2231

View Online
Scheme 3 Synthesis of dibenzothiadiazepine 5,5-dioxides.
the formation of the desired products and the preferred solvent
was N,N-dimethylformamide (DMF).
The dibenzothiadiazepine 5,5-dioxides were obtained
according to Scheme 3: the phenylenediamines 6a,b were
reacted with 2 in pyridine to form the sulfonamides 7a–c: in
the case of the bromo-substituted phenylenediamine 6b, the
two sulfonamides 7b and 7c were formed in the ratio 3 : 1.
These products were cyclized in DMF in the presence of
Cs2CO3 and Cu at 140 1C in a few minutes. The dibenzothia-
diazepine 5,5-dioxides 8b and 8c were formed in the same ratio
3 : 1 as the open sulfonamides 7b and 7c; the structure of the
main product 8b was confirmed by X-ray analysis.
Scheme 4 Synthesis of dibenzodithiazepine 5,5-dioxides.
adopt staggered conformation about the N–S bond
a
(Scheme 5),15 while the energy minimum gas-phase conformation
predicted by high-level quantum mechanical studies has the
NH and NC bonds almost eclipsing the SO linkages.16 For
example for N-methylmethanesulfonamide, the two conformers
differ by 1.44 and 2.63 kcal molꢁ1 at the RHF/6-31G*16b and
MP2/6-31+G*16d level, respectively; while its crystal structure,
collected at 150 K, shows the staggered conformation.17 By the
way, high-level ab initio calculations have been used to develop
force field parameters for the sulfonamide derivatives.16a,b,d
It has been suggested that the gas-phase low energy
conformation found in simple sulfonamides is mainly due
to the resonance interaction between SQO bonds and the
nitrogen electron lone pair:18 the perpendicular orientation of
the latter with respect to the SQO bond allows an optimal
overlap between the nitrogen doublet and the SQO bonds.
The reason why the minimum-energy conformation in the gas
phase is not the minimum-energy conformation in the solid
state as well, can be most likely ascribed to intermolecular
hydrogen bonds and packing forces which, in the crystal,
prevail over ‘‘intramolecular effects’’. Finally as a consequence
of the uncertainty in the minimum-energy conformation,
doubts can be cast on the choice of the solution conformation
selected for modelling studies performed in solution. Because
the results of the latter are crucial for drug design conclusions
should be drawn with care.
Synthesis of dibenzodithiazepine 5,5-dioxides. The dibenzo-
dithiazepine 5,5-dioxides 13a,b and their oxidized counterparts
14a,b are new structures. Only an unsuccessful attempt to
prepare the dibenzodithiazepine 5,5 dioxide was reported by
thermolysis of diphenyl sulfide-2-sulfonyl azide.10 We instead
obtained 13a,b by Ullmann-type reaction in good yields
(Scheme 4). The sensitivity of sulfides towards oxidation holds
some difficulties and the suppression of disulfide formation
remains the main challenge for this type of reaction.11 The
desired cyclic compounds were obtained by using copper
together with a reducing agent as Cu2O, in order to inhibit
the disulfide formation, and avoiding the use of strong bases as
Cs2CO3. The use of a combination of copper and cuprous
oxide in 2-ethoxyethanol suppressed the disulfide formation
and the cyclization occurred in good yield, even in the absence
of the base. The starting 2-aminothiophenols 9a,b were
protected as benzyl thioether derivatives 10a,b according to
described procedures;13 the subsequent reaction with 2 gave
sulfonamides 11a,b. These thio-derivatives were deprotected
by treatment with a 1 M solution of BBr3 in CH2Cl2
and cyclized by treatment with copper and cuprous oxide
in refluxing 2-ethoxyethanol giving the products 13a,b,
expeditiously and in good yields. Only a little amount (5%)
of the dimeric by-product 15a was formed, whose structure
was confirmed by X-ray analysis. Finally the treatment of
derivatives 13a,b with m-chloroperbenzoic acid (mCPBA) in
CH2Cl2 gave sulfones 14a,b (Scheme 4). Structures of 13a and
14a were also confirmed by X-ray analysis.
When the sulfur and the nitrogen atoms bear a bulkier
substituent, such as an aryl moiety, as for example in the
Structural investigations
Scheme
5 Newman projections about the N–S bond for
In the solid state (data retrieved from the Cambridge Structural
Database,14 CSD hereafter) unconstrained sulfonamides
N-methylmethanesulfonamide: the experimentally solid-state observed
structure (left) and the theoretical minimum energy conformation (right).
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 2219–2231 | 2221

View Online
was also extended to the dimer species 5b (Scheme 2) and 15a
(Scheme 4).
Solid-state characterization results. The analysis and the
discussion of the structural features characterizing the tricyclic
compounds 4a, 8b, 13a and 14a (Fig. 4–7) have been focussed
on the overall shape of the molecules rather than be limited to
the geometry of the sulfonamide grouping.
Their overall structure is essentially controlled both by the
conformation of the central seven-membered ring and the
relative arrangement of the fused aromatic moieties about it.
The geometrical data as derived by X-ray measurements
listed in the last column of Tables 1–4 help to quantify
similarities and differences between the solid-state structures
of samples 4a, 8b, 13a and 14a. As far as the 7-membered ring
is concerned, the asymmetry index DCs,23,24 the bow and stern
parameters23,24 and the sum of the ring internal angles were
taken into account. In addition, the bending of the two
aromatic groups with respect to the central ring was measured
by the angles their mean planes form with the plane passing
through N1, S1 and Z (t3); whereas the relative arrangement
of the phenyls was quantified by measuring the angle between
their mean planes (+(A/B) value).4c
Scheme 6 (a) Newman projections about the N–S bond of the
solid-state lowest energy conformation for arylsulfonamide derivatives
(R = Ar; R0 = alkyl) and N-arylmethanesulfonamide (R = methyl;
R0 = Ar); (b) Newman projections about the C–S rotatable bond
of the solid-state lowest energy conformation in arylsulfonamide
derivatives (R = Ar; R0 = alkyl).
arylsulfonamides and N-arylmethanesulfonamides, respectively,
the N lone pair still bisects the OSO angle (Scheme 6(a))
in their solid-state conformation (CSD data and ref. 19).
Moreover, in the arylsulfonamides the p orbital at the ipso
carbon also bisects the OSO angle (Scheme 6(b)). It has been
hypothesized that the arylsulfonamides conformation is the
result of the stabilizing interaction between the lone pairs and
the sulfur d orbitals.20 When a non-hydrogen atom (nH) is
in ortho position, the peak of the N–S–C–CnH dihedral
distribution moves to about 701.19
Results from quantum chemical calculations studies are in
good agreement with these experimental findings. For example
the most stable conformer of benzene sulfonamide has the
S–N bond perpendicular to the benzene ring (the NH2 group is
eclipsing the SO2 moiety);21 while the presence of a methyl
ortho substituent causes a rotation of 301 of the SO2NH2
grouping (the N–S–C–CnH dihedral has a gauche conformation).22
Few crystal structures of cyclic six-, seven- and eight-
membered sulfonamides have been retrieved from the CSD
(Scheme 7 shows the molecular fragments searched for and the
number of refcodes selected for the present study). In the
smallest ring the ‘‘sulfa’’ group shows the usual 3D-arrangement
about the N–S bond: the nitrogen lone pair bisects the OSO
angle: the absolute value of the dihedral angle t1 defined by the
carbon atoms belonging to the ring is about 601. The medium
and the largest sized rings show conformations distributed in a
broader range, with mean |t1| values higher.
The geometry of the sulfonamide grouping was defined by
all the C–NH–SO2–C bond distances (see also ESIw), the sum
of the angles about the N1 atom and finally by the dihedral
angles t1 (C7–N1–S1–C1) and t2 (C6–C1–S1–N1).
A comparative inspection of the numerical data reported in
Tables 3 and 4 highlights that compounds 13a and 14a (Fig. 6
and 7) are very similar both in their resulting overall shape
(see for example t3, DCs, bow and stern values) and in the
sulfonamide arrangement (see for example t1, t2 and the
degree of pyramidalization of N1 evaluated on the basis of
the sum of the bond angles about it). In particular the
conformation about the N–S bond (Scheme 8, left) is in
keeping with the solid-state conformation observed in sulfon-
amides bearing aryl substituents (Scheme 6(a)), that is the
nitrogen lone pair bisects the OSO angle. On the contrary
the dihedral about the C–S linkage is rotated about 901
(Scheme 8, right) with respect to the literature reference value
(Scheme 6(b)). As a whole the solid-state geometry shown by
13a and 14a will be, hereafter, labelled as type III.
In both cases the hydrogen atom bound to N1 points
towards the central seven-membered ring. In 14a a weak
We report here an experimental and theoretical structural
analysis of compounds 4a, 8b (8a), 13a and 14a sketched
in Fig. 3 with the aim to elucidating their conformational
behaviour. The solid-state characterization by X-ray diffraction
Scheme 7 Molecular fragments searched for in the CSD, v5.30
(2009): (a) 34 refcodes selected; (b) 8 refcodes selected; (c) 5 refcodes
selected; X = any non-metal element; — any kind of bond.
Fig. 4 ORTEP3 view of compound 4a. The ellipsoid probability is
set to 30%.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
2222 | New J. Chem., 2009, 33, 2219–2231

View Online
Table 1 Most relevant geometrical parameters (distances (A), angles
(1)) as derived from ab initio geometry optimizations (HF/6-311+
G(d,p) level of theory) of the 4a-I, 4a-II, 4a-III modelled species.a
Those derived from X-ray diffraction data of 4a have been also
reported for comparative purposes
4a-I
4a-II
1.635
4a-III
1.633
4a
rN1–S1/A
rS1–O1b/A
rS1–O2/A
1.635
1.423
1.415
1.616(3)
1.426(2)
1.434(2)
1.423
1.415
1.422
1.419
P
(+N1)c/1 354.4
354.4
345.7
359(2)
ꢁ62.1(3)
ꢁ122.1(3)
55.7/2.3 16.3(1)/58.1(1)
t1d/1
ꢁ58.9
ꢁ59.0
ꢁ55.8
t2e/1
ꢁ121.5
ꢁ121.5
ꢁ178.8
Fig. 5 ORTEP3 view of compound 8b. The ellipsoid probability is
f
set to 30%.
t3 /1
12.5/59.9
12.6/59.9
+(A/B)g/1
DCsh/1
Bowh/1
Sternh/1
a
48.9
42.0
54.5
30.7
48.9
42.0
54.5
30.7
54.5
46.3
55.2
25.1
45.5(1)
50.0
50.6(3)
40.4(1)
b
Refer to Fig. 3 for molecular models and atom labelling. O1 is the
c
sulfone oxygen atom closest to the N-bound hydrogen. Summation
of the bond angles about the sulfonamide nitrogen atom N1.
d
e
C7–N1–S1–C1 Dihedral angle. C6–C1–S1–N1 Dihedral angle.
f
Angles between the plane through N1, S1, Y and the least-squares
g
planes through rings A and B, respectively. Interplanar angle
h
between the aromatic rings. See ref. 23 and 24 for the parameter
definitions.
Table 2 Most relevant geometrical parameters (distances (A), angles
(1)) as derived from ab initio geometry optimizations (HF/6-311+
G(d,p) level of theory) of the 8a-I, 8a-II, 8a-III modelled species.a
Those derived from X-ray diffraction data of 8b have been also
reported for comparative purposes
Fig. 6 ORTEP3 view of compound 13a. The ellipsoid probability is
set to 30%.
8a-I
8a-II
1.643
8a-III
1.630
8b
rN1–S1/A
rS1–O1b/A
rS1–O2/A
1.630
1.423
1.420
1.596(4)
1.435(3)
1.426(3)
1.424
1.423
1.423
1.420
P
(+N1)c/1 347.0
347.0
ꢁ84.0
346.6
359(3)
ꢁ81.3(3)
ꢁ132.0(3)
55.3/13.6 39.0(1)/49.1(2)
t1d/1
ꢁ65.0
ꢁ64.8
t2e/1
ꢁ168.1
ꢁ145.5
ꢁ168.3
f
t3 /1
55.3/13.8
54.3/38.6
+(A/B)g/1
DCsh/1
Bowh/1
Sternh/1
a
44.1
53.6
45.9
19.6
28.6
68.0
23.5
9.1
44.3
53.4
46.1
19.8
21.5(1)
65.7
9.7(4)
28.7(2)
b
Refer to Fig. 3 for molecular models and atom labelling. O1 is the
c
sulfone oxygen atom closest to the N-bound hydrogen. Summation
Fig. 7 ORTEP3 view of compound 14a. The ellipsoid probability
is set to 30%.
of the bond angles about the sulfonamide nitrogen atom N1.
d
e
C7–N1–S1–C1 Dihedral angle. C6–C1–S1–N1 Dihedral angle.
f
H-bond interaction (the NHꢂ ꢂ ꢂOSO distance is 2.53(8) A, the
angle 110(3)1) could contribute to further stabilize this 3D
arrangement.
Angles between the plane through N1, S1, Y and the least-squares
g
planes through rings A and B, respectively. Interplanar angle
h
between the aromatic rings. See ref. 23 and 24 for the parameter
definitions.
Fig. 6 and 7 highlight the overall shape of the molecules 13a
and 14a, which appears quite folded about the central ring
(+(A/B) value), with the N1–S2 direction acting as a sort of
hinge between the two molecular fragments which as a whole
define a convex figure. Finally the seven-membered ring is
quite asymmetric, as provided by the high value of the
asymmetry indexes.
and the NH bond, which eclipses a SO linkage, is directed
outside the molecular skeleton. The p orbital of the S-bound
aromatic carbon results nearly perpendicular to a SQO bond
in keeping with the literature concerning bulky solfonamides.
(Scheme 8 right and Tables 1–2). Besides, compounds 4a and
8b differ in their overall shape: the amine derivative (8b, Fig. 5)
is almost flat (see for example the +(A/B) angular value), and
In both the oxo (4a) and amine (8b) derivatives, and at
variance with the retrieved published data, the nitrogen atom
N1 is almost planar trigonal (Scheme 8 left and Tables 1 and 2)
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 2219–2231 | 2223

View Online
Table 3 Most relevant geometrical parameters (distances (A), angles
(1)) as derived from ab initio geometry optimizations (HF/6-311+
G(d,p) level of theory) of the 13a-I, 13a-II, 13a-III modelled species.a
Those derived from X-ray diffraction data of 13a have been also
reported for comparative purposes
13a-I
1.634
13a-II
1.634
13a-III
13a
rN1–S1/A
rS1–O1b/A
rS1–O2/A
1.632
1.422
1.420
1.610(3)
1.432(2)
1.428(2)
1.424
1.415
1.424
1.415
P
(+N1)c/1 359.3
359.4
ꢁ74.1
346.0
ꢁ53.6
167.7
346(2)
ꢁ69.4(2)
ꢁ173.0(2)
58.7(1)/7.4(1)
t1d/1
ꢁ74.6
ꢁ117.0
18.4/65.0
t2e/1
ꢁ116.9
f
t3 /1
17.9/65.0 56.9/11.7
48.3
59.0
+(A/B)g/1
DCsh/1
47.9
59.3
68.2
43.7
51.6(1)
50.6
Bowh/1
Sternh/1
a
42.5
31.9
42.7
32.1
49.2
33.5
43.2(1)
12.1(1)
b
Refer to Fig. 3 for molecular models and atom labelling. O1 is the
c
sulfone oxygen atom closest to the N-bound hydrogen. Summation
of the bond angles about the sulfonamide nitrogen atom N1.
d
e
C7–N1–S1–C1 Dihedral angle. C6–C1–S1–N1 Dihedral angle.
f
Scheme 8 Newman projections about the N–S and C–S bonds
Angles between the plane through N1, S1, Y and the least-squares
g
planes through rings A and B, respectively. Interplanar angle between
h
the aromatic rings. See ref. 23 and 24 for the parameter definitions.
showing the different conformations in the X-ray solid-state structures
of molecules 4a, 8b, 13a, 14a.
Dimeric compounds 5b and 15a are in a ‘‘sandwich-like’’
conformation (see Fig. 8 and 9) characterized by p–p inter-
actions25 involving the aromatic rings A and A0 (see Fig. 8 and 9).
The latter are almost parallel each others (the angle between
their mean planes is 6.5(2) and 10.7(3)1 for 5b and 15a,
respectively) with their centroids at 3.914(4) and 3.66(1) A in
5b and 15a, respectively, and almost perpendicular with
respect to the other two aromatic rings B and B0. The mean
planes defined by the carbon atoms of the latter form an angle
of 46.1(2) and 30.1(3)1 in 5b and 15a, respectively. The ‘‘sulfa’’
grouping shows the expected conformation about both the
S–N and C–S bonds.
Table 4 Most relevant geometrical parameters (distances (A), angles
(1)) as derived from ab initio geometry optimizations (HF/
6-311+G(d,p) level of theory) of the 14a-I, 14a-II, 14a-III modelled
species.a Those derived from X-ray diffraction data of 14a have been
also reported for comparative purposes
14a-I
1.638
14a-II
1.638
14a-III
14a
rN1–S1/A
rS1–O1b/A
rS1–O2/A
1.627
1.417
1.419
1.594(7)
1.424(4)
1.446(6)
1.421
1.412
1.420
1.412
P
(+N1)c/1 355.2
355.2
347.5
348(2)
t1d/1
ꢁ86.6
ꢁ120.4
31.5/65.5
ꢁ86.7
ꢁ60.7
ꢁ62.9(6)
178.0(5)
t2e/1
ꢁ120.5
174.8
f
In both the crystal lattices the oxygen atom O(1s) of the
crystallization Me2SO (DMSO) molecule acts as a bidentate
acceptor of H-bonds with respect to the hydrogen atoms
bound to N(1a) and N(1b) (see ESIw).
In summary the solid-state conformation about the N–S
bond found in the oxo (4a) and amine (8b) derivatives is rather
unusual with respect both to the literature data and that
t3 /1
31.6/65.5 25.8/38.4
59.1(3)/2.9(2)
60.5(2)
53.4
+(A/B)g/1
DCsh/1
39.9
45.6
39.8
45.6
64.1
50.3
Bowh/1
Sternh/1
a
32.3
27.4
32.2
27.4
48.8
28.4
45.8(3)
20.4(4)
b
Refer to Fig. 3 for molecular models and atom labelling. O1 is the
c
sulfone oxygen atom closest to the N-bound hydrogen. Summation
of the bond angles about the sulfonamide nitrogen atom N1.
d
e
C7–N1–S1–C1 Dihedral angle; C6–C1–S1–N1 Dihedral angle.
f
Angles between the plane through N1, S1, Y and the least-squares
g
planes through rings A and B, respectively. Interplanar angle between
h
the aromatic rings. See ref. 23 and 24 for the parameter definitions.
its shape results definitely more ‘‘open’’ with respect to 13a
and 14a; while compound 4a (Fig. 4) shows a concave shape
(vs. the convex one featuring 13a and 14a), folded about the
N1–O3 direction. As a whole the solid-state arrangements of
4a and 8b will be labelled as type I and type II, respectively.
Molecule 8b shows by far the most asymmetric central ring
(DCs = 65.7 vs. 50.01 (for 4a), 50.61 (for 13a) and 53.41
(for 14a).
Concerning the crystal packing, all the four tricyclic
compounds 4a, 8b, 13a and 14a are held together in the lattice
by a net of hydrogen bonds (see ESIw).
Fig. 8 ORTEP3 view of 5bꢂMe2SO. The ellipsoid probability is set
to 30%.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
2224 | New J. Chem., 2009, 33, 2219–2231

View Online
Fig.
9
ORTEP3 view of compound 15aꢂMe2SO. The ellipsoid
probability is set to 30%.
shown by the sulfur compounds (13a and 14a). The reverse
holds for the arrangement about the C–S bond which for the
lightest 4a and 8b species is in keeping with the reported data,
while in the sulfur derivatives the S-bound aromatic ring
almost eclipses the N–S bond.
Fig. 10 Newman projections about the N–S (top) and C–S (bottom)
bonds of the optimized 4a-I/4a-III models.
For compounds 4a, 13a and 14a the minimization procedure
makes geometry of type II transform into I as provided by
a comparison of the geometrical parameters featuring the
optimized models 4a-I and 4a-II (Table 1, Fig. 10), 13a-I
and 13a-II (Table 3, Fig. 12), 14a-I and 14a-II (Table 4,
Computational study results. With this in mind we set up a
modelling study aiming to finding a rationale for the solid-
state data and eventually to investigate if there exists any
correlation between the observed geometry and the nature of
the Z grouping (O, NH, S, SO2). For each molecule (4a, 8a,
13a and 14a, Fig. 3) three different starting geometries (I, II
and III) were taken into account for the subsequent fully
optimization by the quantum chemical procedures (details in
the experimental section). Tables 1–4 list the most interesting
geometrical parameters related to the optimized structures
(4a-I/4a-III, 8a-I/8a-III, 13a-I/13a-III, 14a-I/14a-III). Results
are summarized in the following.
Fig. 13). In all cases the optimized conformer I is characterized
P
by an almost planar sulfonamide nitrogen (mean (+N1) =
3561) with the bound hydrogen practically eclipsing one
SO bond and closely resembles the experimental solid-state
structure of 4a; while the optimized conformer III (the most
stable) well reproduces the experimental structure of the sulfur
(13a) and sulfone (14a) derivatives.
As to the 8a species, the geometry optimization of 8a-I
proceeds through the inversion of both the nitrogen atoms and
As a general remark, in all the optimized species the N–S
bond distances are larger than the experimental values. The
same trend is almost invariably observed for the S–O1 bond
with respect to the S–O2 one (see Tables 1–4). The most stable
conformation is always of type III (see Table 5) where the
sulfonamide nitrogen features an sp3 hybridization (mean
P
(+N1) = 3461) with the nitrogen lone pair bisecting the
SO2 angle (as in the bulky unconstrained species, Scheme 6)
and the S-bound phenyl ring parallel with respect to the S–N
bond. As a final remark, the type III conformer is increasingly
stabilized on going from NH - O - S - SO2 (0.67, 1.29, 3.3,
9.4 kcal molꢁ1, respectively, Table 5).
Table
5 HF/6-311+G(d,p)//HF/6-311G+(d,p) energy difference
(kcal molꢁ1) with respect to the most stable conformer for each
modelled compound
I
II
III
4a
8a
13a
14a
+1.29
0.0
+3.3
+9.4
+1.29
+0.67
+3.3
0.0
0.0
0.0
0.0
Fig. 11 Newman projections about the N–S (top) and C–S (bottom)
bonds of the optimized 8a-I/8a-III models.
+9.4
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 2219–2231 | 2225

View Online
In summary, while the solid-state geometry of the heavier
derivatives 13a and 14a is well reproduced by the gas phase
minimum-energy conformer III, the 3D arrangement shown
by the oxo and amine derivatives in the solid state (I and II,
respectively) appears slightly disfavoured on an energetic
ground with respect to the lowest energy gas-phase conformer III.
The latter in all cases features an sp3 hybridized amide
nitrogen, while the experimental solid-state structures of 4a
and 8b have a trigonal planar nitrogen. The Weinhold’s
Natural Bond Orbital (NBO)26 analysis highlights that in all
the optimized structures the nitrogen lone pair is partially
delocalized over the neighbouring atomic groupings. In all
cases, while the lone pair of the pyramidal nitrogen (type III
conformer) is mainly delocalized on the S1–C1 s* bond, when
the nitrogen has sp2 character, the p doublet moves in opposite
direction being largely transferred to the p* orbital localized
between C7–C8. The 3D arrangement about the S1–C1 bond
in type III conformer and in the solid-state structures of 13a
and 14a can be explained as the natural consequence of the
lone pair delocalization and seven-membered ring closure.
When the conjugation interests the N-bound phenyl (3D
arrangement of 4a and 4I-a) the dihedral about the S1–C1
linkage has the usual value as expected in case of bulky
substituents on the aryl fragment. As a final remark the
transfer of the lone pair electron density appears more efficient
when the doublet is hosted in an almost pure p orbital than in
a sp3 hybrid as implicitly suggested by the slightly larger
electronic occupancy of the latter orbital. The stabilizing effect
related to a more efficient electron density delocalization
operating in conformer I could be important in determining
the ‘‘sulfa’’ group conformation of the lightest derivative 4a
(its solid-state structure features an sp2 nitrogen and confor-
mer 4a-I is only 1.29 kcal molꢁ1 higher in energy than 4a-III).
In the sulfur species 13a and 14a, which undoubtedly prefer
conformer III (that is by far the gas phase most stable and it
is the solid-state conformer) other factors must contribute
to drive their 3D arrangement, as for example the SO2–Z
steric hindrance. In 14a the oxygen atoms of the two sulfone
groupings are 2.985 A apart in conformer I vs. 3.972 A in III
(the sum of the oxygen van der Waals radii is 3.04 A27). For
comparison 13a-I features an Sꢂ ꢂ ꢂO distance of 3.173 vs.
4.210 A in 13a-III (the sum of the corresponding van der
Waals radii is 3.32 A). Results from geometry optimizations
of two further molecular models having Z = CH2 and
C(CH3)2 could substantiate this hypothesis.28 The first
model (Z = CH2) prefers the type I geometry (D(I–III) =
ꢁ1.52 kcal molꢁ1), while for the dimethyl derivative the
reverse holds (D(I–III) = +2.99 kcal molꢁ1).
Fig. 12 Newman projections about the N–S (top) and C–S (bottom)
bonds of the optimized 13a-I/13a-III models.
Fig. 13 Newman projections about the N–S (top) and C–S (bottom)
bonds of the optimized 14a-I/14a-III models.
results into the isoenergetic 8a-III conformer (Table 2, Fig. 11).
The optimized 8a-II isomer looks like the experimental structure
8b, except for the N1 hybridization (sp3 in the former vs. sp2 in
the experimental structure) and it is 0.67 kcal molꢁ1 higher in
energy than conformer III.
Conclusions
The convergence trends outlined above may suggest that
type II geometry is somehow peculiar of the amine derivative
8a, given that the geometry optimization of the #-II (# = 4a,
13a and 14a) species leads in all cases to conformer I. As a
consequence we can suppose that for the oxo, sulfur and
sulfone derivatives there is a quite low energy barrier for the
II - I transformation. Conformer III corresponds in all cases
to the most thermodynamically stable and, most likely it is
placed in a quite deep potential well.
In conclusion, through customized synthetic procedures we
have obtained highly functionalizable sulfonamide tricyclic
structures featuring NH, O, S and SO2 groups in the central
seven-membered ring, for which general and reproducible
methods of synthesis were lacking. Furthermore, as sulfonamide-
related conformational effects have been proposed to play a
role in structure based drug design8 and tricyclics are well
know privileged structures in medicinal chemistry, attention
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
2226 | New J. Chem., 2009, 33, 2219–2231

View Online
has been paid to the 3D arrangement of these molecules.
Single-crystal X-ray diffraction data and quantum chemical
studies allowed to compare the minimum energy geometries in
the solid state (I, Z = O; II, Z = NH and III, Z = S, SO2)
with the minimum-energy geometries in the gas phase (invariably
the type III geometry). The results, which are interpreted on
the basis of both the efficiency in the amide nitrogen lone pair
delocalization and of the SO2–Z steric hindrance, possibly lead
to a better understanding of the three-dimensional arrangement
of these molecules that could be helpful in the study of their
interactions with their biological targets.
2-Bromo-N-(2-hydroxyphenyl)benzenesulfonamide 3a. To a
stirred solution of 2-aminophenol 1a (2.20 g, 20 mmol) in dry
pyridine (40 mL) at 0 1C, under an atmosphere of dry nitrogen,
2-bromobenzenesulfonyl chloride 2 (5.11 g, 20 mmol) was added
portionwise. The reaction mixture was stirred for 0.5 h at 0 1C
and 1 h at room temperature, then the solvent was removed
under reduced pressure. The residue was diluted with an
aqueous 1 M HCl solution and extracted with CH2Cl2. The
organic phase was washed with brine, dried on phase separator.
After removal of the solvent 3a (5.93 g, 90%) was obtained as
a solid.
1
HPLC: 3.65 min. H NMR (200 MHz, DMSO-d6): d 9.45
(br s, 1H), 8.00–7.89 (m, 1H), 7.88–7.77 (m, 1H), 7.55–7.44
(m, 2H), 7.11–7.02 (m, 1H), 6.98–6.86 (m, 1H), 6.80–6.60
(m, 2H). ESIꢁ [M ꢁ H]ꢁ = m/z 326.
Experimental
Synthesis
Compounds 3b–j were obtained with a similar procedure.
All reagents and solvents were used without further purification
or drying and were purchased from Sigma-Aldrich (Milan,
Italy). Melting points were measured in capillary tubes with
a digital electrothermal apparatus Buchi B-540 and were
uncorrected. TLC monitoring was performed on Merck silica
gel 60 F254 plates and detection by UV light. HPLC chromato-
grams were recorded on a Agilent HP-1100 instrument using
the following method: ZorbaxTM columm SB-18, 3.5 mm,
100 A (50 ꢃ 4.6 mm). Mobile phase: A, trifluoroacetic acid
0.1% in water; B, trifluoroacetic acid 0.1% in acetonitrile;
gradient from 5% B to 95% B in 6.5 min, then 95% B for
1 min; flow: 3 mL minꢁ1; wavelength 1: 210 nm, wavelength
2: 270 nm.
1H NMR spectra were acquired at 600 MHz on a Bruker
AC 600 spectrometer or on a Varian Gemini 200 spectrometer
and referenced against the residual solvent peak (DMSO-d6 at
2.50 ppm and CDCl3 at 7.25 ppm). All processing were
performed using the software Mestrec 4.5.9.1. Mass spectra
were obtained with a Finnigan LCQ ion trap mass spectro-
meter, operated in positive-ion or in negative-ion electrospray
ionization. The samples were analyzed by full-scan MS and
product ion MS/MS of the quasi-molecular ions, at 30%
relative collision energy, using helium as the collision gas.
The high-resolution mass spectra were obtained with a
Micromass Q-Tof micro mass spectrometer, calibrated with
0.1% phosphoric acid in 1 : 1 H2O–MeCN. This same mixture
was used also as internal reference compound during ESI-MS
accurate mass experiments, and was introduced via the
Lock-Spray channel using an infusion pump. Ionization mode:
ESI, positive or negative ion. Scan mode: full-scan MS from
m/z 100 to 1000; capillary voltage: 3300 V; capillary temperature:
6H-Dibenzo[b,f][1,4,5]oxathiazepine 5,5-dioxide 4a.
A
solution of compound 3a (1.22 g, 3.71 mmol) in 1-methyl-2-
pyrrolidone (40 mL) was treated with cesium carbonate
(3.63 g, 11.14 mmol). The mixture was heated at 140 1C for
20 h, then the solvent was removed under reduced pressure.
The residue was diluted with an aqueous 1 M HCl solution
and extracted with CH2Cl2. The organic phase was washed
with brine, dried on phase separator. After removal of the
solvent a crude oil was obtained, which was purified by
flash chromatography (cyclohexane–ethyl acetate 2 : 1) using
column STRATA SiO2 (50 g) to give 4a (590 mg, 64.3%) as a
solid, mp 132–133 1C (from ethyl ether–hexane). TLC: Rf
(ethyl acetate–cyclohexane: 1 : 2) = 0.46. HPLC: 3.42 min,
98.0% purity. 1H NMR (600 MHz, DMSO-d6): d 10.85
(s, 1H), 7.80–7.76 (m, 1H), 7.74–7.69 (m, 1H), 7.52–7.48
(m, 1H), 7.44–7.40 (m, 1H), 7.38–7.44 (m, 1H), 7.20–7.14
(m, 2H), 7.08–7.04 (m, 1H). ESIꢁ [M ꢁ H]ꢁ = m/z 246.
HRMS: m/z 246.0226 ([C12H9NO3S–H]ꢁ; calc. 246.0227).
Compounds 4b–j were obtained with a similar procedure.
2,13-Dichloro-11H,22H-5,16-dioxa-10,21-dithia-11,22-diaza-
tetrabenzo[a,d,h,k]cyclotetradecene 10,10,21,21-tetraoxide 5b.
A solution of compound 3b (0.72 g, 2 mmol) in DMF
(40 mL) was treated with cesium carbonate (1.90 g, 6 mmol)
and CuI (20 mg, 0.1 mmol). The mixture was heated at 110 1C
for 4 h. After cooling, the reaction mixture was diluted with an
aqueous 1 M NaOH solution and the insoluble substance
filtered off. The filtrated solution was acidified with 1 M HCl
and an insoluble material was recovered by filtration. A crude
solid was obtained, which was purified by flash chromato-
graphy (SiO2) and eluted using 20% of ethyl acetate in hexane
150 1C; source temperature: 80 1C; cone gas: nitrogen, 50 L hꢁ1
Sample introduction: direct infusion through the built-in syringe
pump, flow 5 mL minꢁ1
.
1
to give 5b as a solid (0.07 g, 6%). HPLC: 4.96 min. H NMR
(600 MHz, CDCl3): d 9.72 (br s, 2H), 7.78 (d, J = 2.5 Hz, 2H),
7.76–7.72 (m, 2H), 7.2 (dd, J = 2.5, 8.7 Hz, 2H), 6.94 (d, J =
8.7 Hz, 2H), 6.91–6.85 (m, 4H), 5.76 (bd, J = 5.8 Hz, 2H).
ESIꢁ [M ꢁ H]ꢁ = m/z 561.
.
EI mass spectra were obtained with an Agilent 5973
GC-MS system, using an HP-5-MS open tubular capillary
column, 15 m length, 0.25 mm internal diameter, 0.1 mm film
thickness. The oven temperature was set at 60 1C for 3 min,
then ramped at 50 1C minꢁ1 up to 310 1C, and held at the final
temperature for 2 min. 1 mL of the sample solution was
injected in split mode, with a split ratio 1 : 100. The carrier
gas was helium at 1.0 mL minꢁ1. The injector temperature
was 250 1C.
N-(2-Aminophenyl)-2-bromobenzenesulfonamide 7a. To a
stirred solution of benzene-1,2-diamine 6a (2.70 g, 25 mmol)
in dry pyridine (30 mL) at 0 1C, under an atmosphere of dry
nitrogen, 2-bromobenzenesulfonyl chloride 2 (6.38 g, 0.025 mol)
was added portionwise. The reaction mixture was stirred for
0.5 h at 0 1C and 1 h at room temperature, then the solvent was
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 2219–2231 | 2227

View Online
removed under reduced pressure. The residue was diluted with
an aqueous 1 M HCl solution and extracted with CH2Cl2
(900 mL) and again with CH2Cl2 (450 mL). The organic phase
was washed with brine, dried on phase separator. After
removal of the solvent the crude residue was washed with
CH2Cl2 and 7a (3.80 g, 46.5%) was obtained as a pure solid.
HPLC: 2.82 min. 1H NMR (200 MHz, DMSO-d6): d 7.93–7.79
(m, 2H), 7.57–7.40 (m, 2H), 6.94–6.79 (m, 1H), 6.71–6.56
(m, 2H), 6.41–6.26 (m, 1H). ESI+ [M + H]+ = m/z 327.
N-(2-Benzylsulfanylphenyl)-2-bromobenzenesulfonamide 11a.
To a stirred solution of 10a (2.54 g, 11.79 mmol) in pyridine
(25 mL), at 0 1C under an atmosphere of dry nitrogen, 2 (6.2 g,
25 mmol) was added portionwise. The reaction mixture was
stirred at 0 1C for 2 h and for 16 h at room temperature, then
the solvent was removed under reduced pressure. The residue
was diluted with an aqueous 1 M HCl solution and extracted
with CH2Cl2. The organic phase was washed with brine, an
aqueous 5% NaHCO3 solution, brine and then dried on phase
separator and concentrated in vacuo to give 11a (4.60 g, 90%)
as a solid. TLC: Rf (ethyl acetate–hexane, 1 : 4) = 0.44.
HPLC: 5.31 min. 1H NMR (200 MHz, DMSO-d6): d 9.62
(s, 1H), 7.96–7.80 (m, 2H), 7.60–7.46 (m, 2H), 7.38–6.90
(m, 9H), 4.07 (s, 2H). ESI+ [M + H]+ = m/z 434. HRMS:
m/z 433.9879 ([C19H16NO2S2Br + H]+; calc. 433.9884).
HRMS: m/z 326.9803 ([C12H11N2O2SBr
+
H]+; calc.
326.9803).
N-(2-Amino-4-bromophenyl)-2-bromobenzenesulfonamide 7b
and N-(2-amino-5-bromophenyl)-2-bromobenzenesulfonamide
7c. Compounds 7b and 7c were obtained following the
procedure described above for 7a starting from 6b (1.45 g,
7.75 mmol) and 2-bromobenzenesulfonyl chloride 2 (2.00 g,
7.84 mmol), in pyridine (15 mL) as a 3 : 1 mixture of
regioisomers, which were separated by crystallization from
MeOH in which 7b is more soluble. After evaporation of the
solvent, 7b (1.50 g, 60%) was obtained. HPLC: 4.04 min.
1H NMR (200 MHz, DMSO): d 9.62 (br s, 1 H), 7.89–7.85
(m, 2 H), 7.59–7.47 (m, 2 H), 6.82–6.80 (m, 1 H), 6.61–6.48
(m, 2 H), 5.31 (br s, 2 H).
N-(2-Benzylsulfanyl-5-chlorophenyl)-2-bromobenzenesulfo-
namide 11b. In a similar way, starting from 10b (1.07 g,
4.3 mmol), 11b was obtained (1.14 g, 56%) as a solid. HPLC:
5.68 min. 1H NMR (200 MHz, DMSO-d6): d 9.90 (s, 1H),
7.98–7.82 (m, 2H), 7.64–7.50 (m, 2H), 7.39–6.98 (m, 8H), 4.07
(s, 2H). EI+ M+ = m/z 467.
2-Bromo-N-(2-mercaptophenyl)benzenesulfonamide 12a. To
a stirred solution of compound 11a (4.30 g, 9.9 mmol) in dry
CH2Cl2 (170 mL) at 0 1C, under an atmosphere of dry
nitrogen, 1 M boron tribromide in CH2Cl2 was added, then
the cold bath was removed and the reaction mixture was
stirred for 30 min at room temperature. Brine was added to
the reaction mixture and the crude product was extracted with
CH2Cl2. The organic phase was treated with an aqueous
solution of 1 M NaOH and the product was solubilized
in the aqueous phase. Its subsequent acidification with
concentrated HCl and extraction with CH2Cl2 gave, after
washing with brine, drying on phase separator and evaporation
in vacuo, 12a as an oil (3.40 g, 100%).
The product 7c, insoluble in MeOH, was filtered off and
isolated (0.50 g, 20%). HPLC: 3.89 min. 1H NMR (200 MHz,
DMSO-d6): d 7.92–7.85 (m, 2 H), 7.59–7.52 (m, 2 H),
7.06–7.00 (m, 1 H), 6.82–6.79 (m, 1 H), 6.59 (d, J = 8.7 Hz,
1 H), 5.33 (br s, 2 H).
6,11-Dihydrodibenzo[c,f][1,2,5]thiadiazepine 5,5-dioxide 8a.
A solution of compound 7a (0.90 g, 2.76 mmol) in N,N-
dimethylformamide (40 mL) was heated at 140 1C and treated
first with elemental copper (178 mg, 2.8 mmol) and then with
cesium carbonate (0.90 g, 2.8 mmol). After 20 min at 140 1C,
the reaction mixture was cooled with a cold water bath and,
after filtration by suction, the filtered solution was evaporated
to give a residue, that was dissolved in 1 M sodium hydroxide
and washed with CH2Cl2. The aqueous solution was acidified
with HCl and extracted with CH2Cl2. The organic phase was
washed with brine, dried on a phase separator; after removal
of the solvent a crude solid was obtained, which was purified
by flash chromatography (CH2Cl2) using column STRATA
SiO2 (50 g) to give 8a as a solid (430 mg, 63%), mp 198–199 1C.
TLC: Rf (CH2Cl2–MeOH: 95/5) = 0.60. HPLC: 3.09 min,
99.7% purity. 1H NMR (600 MHz, DMSO-d6): d 9.82 (s, 1H),
9.00 (s, 1H), 7.63–7.59 (m, 1H), 7.40–7.35 (m, 1H), 7.20–7.12
(m, 2H), 7.10–7.00 (m, 2H), 6.90–6.80 (m, 2H). ESI+
[M + H]+ = m/z 247. HRMS: m/z 247.0541 ([C12H10N2O2S
+ H]+; calc. 247.0541).
HPLC: 4.18 min. 1H NMR (200 MHz, CDCl3): d 8.20–8.12
(m, 1H), 7.97 (s, 1H), 7.75–7.76 (m, 1H), 7.50–6.90 (m, 6H),
3.24 (s, 1H).
2-Bromo-N-(5-chloro-2-mercaptophenyl)benzenesulfonamide
12b. In a similar way, starting from 11b, compound 12b was
obtained as a solid (1.18 g, 96.8%).
HPLC: 4.60 min. 1H NMR (200 MHz, CDCl3): d 8.26–8.17
(m, 1H), 8.11 (s, 1H), 7.76–7.68 (m, 1H), 7.60–6.82 (m, 5H),
3.15 (s, 1H).
Both compounds were used for the following synthesis of
13a and 13b without further purification.
6H-Dibenzo[b,f][1,5,2]dithiazepine 5,5-dioxide 13a. A solu-
tion of compound 12a (1.52 g, 4.4 mmol) in 2-ethoxyethanol
(60 mL) was treated with elemental copper (0.56 g, 8.8 mmol),
heated at reflux and then treated with cuprous oxide (0.31 g,
2.2 mmol) and refluxed for 20 min. The hot bath was removed.
The reaction mixture was cooled with a cold water bath and,
after filtration by suction, the filtered solution was evaporated
to give an oil that was treated with boiling ethyl acetate.
The insoluble material was recovered by filtration, and was
characterized as the dimer compound 15a (see data below).
The filtered solution was treated with hexane and the desired
9-Bromo-6,11-dihydrodibenzo[c,f][1,2,5]thiadiazepine 5,5-di-
oxide 8b. Compound 8b was prepared following the procedure
described above for 8a starting from 7b (1.50 g, 3.70 mmol),
Cu (0.60 g, 9.44 mmol) and Cs2CO3 (3.00 g, 9.23 mmol).
The product 8b (0.90 g, 75%) was obtained as a solid. HPLC:
3.68 min. 1H NMR (200 MHz, DMSO-d6): d 9.99 (s, 1 H), 9.17
(s, 1 H), 7.62 (d, J = 7.9 Hz, 1 H), 7.46–7.38 (m, 1 H),
7.29–7.28 (m, 1 H), 7.17 (d, J = 8.2 Hz, 1 H), 7.04–6.84
(m, 3 H). ESI+ [M + H]+ = m/z 325.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
2228 | New J. Chem., 2009, 33, 2219–2231

View Online
1
product crystallized to give 13a (0.926 g, 80%), mp 118–119 1C.
HPLC: 3.67 min, 499% purity. 1H NMR (600 MHz,
DMSO-d6): d 10.78 (s, 1H), 7.92–7.88 (m, 1H), 7.67–7.64
(m, 1H), 7.61–7.56 (m, 1H), 7.53–7.44 (m, 2H), 7.37–7.33
(m, 1H), 7.25–7.20 (m, 2H). ESIꢁ [M ꢁ H]ꢁ = m/z 262
HRMS: m/z 261.9996 ([C12H9NO2S2 ꢁ H]ꢁ; calc. 261.9996).
purity. H NMR (600 MHz, DMSO-d6): d 11.71 (br s, 1H),
8.32–8.28 (m, 1H), 8.14–8.06 (m, 2H), 8.00–7.87 (m, 3H),
7.73–7.65 (m, 2H). ESIꢁ [M ꢁ Hꢁ] = m/z 294. HRMS: m/z
293.9898 ([C12H9NO4S2 ꢁ H]ꢁ; calc. 293.9895).
8-Chloro-6H-dibenzo[b,f][1,5,2]dithiazepine 5,5,11,11-tetra-
oxide 14b. In a similar way, starting from compound 13b
(0.36 g, 1.2 mmol), 14b was obtained as a crude solid which
was purified in boiling water. The insoluble substance was
filtered from the hot suspension, obtaining the pure product
14b (0.33 g, 83%).
HPLC: 3.63 min, 499% purity. 1H NMR (600 MHz,
DMSO-d6): d 11.86 (br s, 1H), 8.32–8.27 (m, 1H), 8.13 (d,
J = 8.5 Hz, 1H), 8.10–8.07 (m, 1H), 8.01–7.97 (m, 1H),
7.93–7.90 (m, 1H), 7.79 (d, J = 8.5 Hz, 1H), 7.76 (dd, J =
8.5, 1.9 Hz, 1H). ESIꢁ [M ꢁ Hꢁ] = m/z 328. HRMS: m/z
327.9502 ([C12H8NO4S2Cl ꢁ H]ꢁ; calc. 327.9505).
11H,22H-5,10,16,21-Tetrathia-11,22-diazatetrabenzo[a,d,h,k]-
cyclotetradecene 10,10,21,21-tetraoxide 15a. mp 380–382 1C
decomp. (DMSO). HPLC: 5.16 min. 1H NMR (600 MHz,
DMSO-d6): d 9.58 (s, 2H), 7.97–7.93 (m, 2H), 7.70–7.66 (m, 2H),
7.58–7.55 (m, 2H), 7.43–7.36 (m, 4H), 6.97–6.92 (m, 2H),
6.89–6.84 (m, 2H), 5.76–5.72 (m, 2H). ESI+ [M + H]+
=
m/z 527. HRMS: m/z 527.0213 ([C24H18N2O4S4 + H]+; calc.
527.0228).
8-Chloro-6H-dibenzo[b,f][1,5,2]dithiazepine 5,5-dioxide 13b.
In a similar way as for 13a, starting from compound 12b (0.49 g,
1.3 mmol), 13b was obtained as solid which was crystallized
from hexane–ethyl acetate (0.75 g, 26%).
Crystal data and refinement
HPLC: 4.16 min, 4 99% purity. 1H NMR (600 MHz,
DMSO-d6): d 11.03 (br s, 1H), 7.93–7.91 (m, 1H), 7.69–7.66
(m, 1H), 7.65–7.61 (m, 1H), 7.56–7.53 (m, 1H), 7.51 (d, J =
8.5 Hz, 1H), 7.28 (dd, J = 8.5, 2.3 Hz, 1H), 7.24 (d, J = 2.3 Hz,
1H). ESIꢁ [M ꢁ H]ꢁ = m/z 296. HRMS: m/z 295.9612
([C12H8NO2S2Cl ꢁ H]ꢁ; calc. 295.9607).
Intensity data were collected on an Oxford Diffraction
Xcalibur diffractometer equipped with a CCD area detector,
using Mo-Ka radiation (0.71073 A) monochromated with a
graphite prism for compounds 4a, 8b, 13a, 14a, 15aꢂMe2SO,
while for compound 5bꢂMe2SO, Cu-Ka radiation (1.54018 A)
was used. All the data, except for compound 15aꢂMe2SO
(T = 150 K), were collected at room temperature. Data were
collected through the program CrysAlis CCD29 and the
reduction was carried on with the program CrysAlis RED.30
Absorption correction was performed with the program
ABSPACK in CrysAlis RED. Structures were solved with
the direct methods of the SIR9731 package and refined by full-
matrix least squares against F2 with the program SHELX97.32
Geometrical calculations were performed by PARST9733 and
molecular plots were produced by the program ORTEP3.34
In all the six structures all the non-hydrogen atoms were
refined anisotropically. Concerning the hydrogen atoms in
compounds 4a, 8b and 13a they were all found in the Fourier
6H-Dibenzo[b,f][1,5,2]dithiazepine 5,5,11,11-tetraoxide 14a.
A solution of compound 13a (0.66 g, 2.5 mmol) in CH2Cl2
(40 mL) was treated, under stirring, with 3-chloroperbenzoic
acid 77% (1.12 g, 5 mmol). The mixture was stirred at room
temperature overnight and then it was diluted with CH2Cl2
(250 mL). The solution was washed with NaHCO3 (5%
aqueous solution, 100 mL) containing sodium thiosulfate
(10% aqueous solution, 25 mL). Drying the organic phase
with phase separator and removal of the solvent under
reduced pressure gave 14a (0.40 g, 55% yield), mp 229–230 1C
decomp. (from ethyl acetate–hexane). HPLC: 3.16 min, 97.2%
Table 6 Crystallographic data and refinement parameters for 4a, 8b, 13a, 14a, 5bꢂMe2SO and 15aꢂMe2SO (refinement method: full-matrix
least-squares on F2)
14a
5bꢂMe2SO
15aꢂMe2SO
4a
8b
13a
Chemical formula
M
T/K
C12H9NO3S
247.26
298
C12H9N2O2SBr
325.18
298
C12H9NO2S2
263.32
298
C12H9NO4S2
295.32
298
C26H22N2O7S3Cl2
641.54
298
C26H24N2O5S5
604.77
150
l/A
Crystal system
Space group
a/A
b/A
c/A
a/1
0.71073
Orthorhombic
Pna21
8.342(1)
9.678(2)
13.675(2)
0.71073
Triclinic
0.71073
Monoclinic
P21/a
7.1412(5)
22.255(2)
7.9556(8)
0.71073
Triclinic
1.54018
Monoclinic
P21/a
8.4934(4)
40.567(2)112.706(6)
8.9646(7)
0.71073
Triclinic
ꢀ
P1
9.261(2)
10.064(4)
15.947(5)
99.54(3)
ꢀ
P1
ꢀ
P1
7.333(2)
8.131(3)
10.314(2)
97.41(2)
90.15(2)
91.61(2)
609.5(3)
2, 1.772
3.537
2963/2394
2394/199
R1 = 0.0441
wR2 = 0.1141
R1 = 0.0690
wR2 = 0.1309
8.050(3)
8.123(3)
11.287(6)
101.816(9)
97.338(9)
119.11(2)
608.5(5)
2, 1.612
0.446
3621/1709
1709/176
R1 = 0.0567
wR2 = 0.1042
R1 = 0.1564
wR2 = 0.1189
b/1
109.175(8)
90.98(2)
g/1
112.46(3)
1349.3(7)
2, 1.489
0.471
8850/4149
4149/343
R1 = 0.0883
wR2 = 0.2886
R1 = 0.1040
wR2 = 0.2933
V/A3
1104.0(3)
4, 1.488
0.287
6149/2174
2174/190
R1 = 0.0354
wR2 = 0.0629
R1 = 0.0712
wR2 = 0.0698
1194.2(2)
4, 1.465
0.433
8323/2569
2569/190
R1 = 0.0428
wR2 = 0.0795
R1 = 0.0972
wR2 = 0.1000
2849.4(3)
4, 1.495
4.522
11 997/4254
4254/371
R1 = 0.0598
wR2 = 0.1633
R1 = 0.0820
wR2 = 0.1715
Z, Dc/g cmꢁ3
m/mmꢁ1
Reflections collected/unique
Data/parameters
Final R indices
[I 4 2s(I)]
R indices
(all data)
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 2219–2231 | 2229

View Online
synthesis and refined isotropically. In compounds 14a and
5bꢂMe2SO the hydrogen atoms bonded to the nitrogen atoms
were found in the Fourier synthesis and refined isotropically,
all the other hydrogens were introduced in calculated position
and refined in agreement with the coordinates of the atoms to
which they are bound.
8 S. Senger, C. Chan, M. A. Convery, J. A. Hubbard, G. P. Shah,
N. S. Watson and R. J. Young, Bioorg. Med. Chem. Lett., 2007, 17,
2931.
9 (a) A. Weber and J. Frossard, Ann. Pharm. Fr., 1966, 6, 445;
(b) N. Lebegue, S. Gallet, N. Flouquet, P. Carato, S. Giraudet and
P. Berthelot, Heterocycles, 2004, 63, 2457; (c) N. Lebegue,
S. Gallet, N. Flouquet, P. Carato, B. Pfeiffer, P. Renard,
´ ´
S. Leonce, A. Pierre, P. Chavatte and P. Berthelot, J. Med. Chem.,
2005, 48, 7363.
10 R. A. Abramovitch, C. I. Azogu, I. T. McMaster and
D. P. Vanderpool, J. Org. Chem., 1978, 43, 1218.
11 K. Kunz, U. Scholz and D. Ganzer, Synlett, 2003, 2428.
12 (a) M. Carril, R. SanMartin, F. Churruca, I. Tellitu and
E. Dominguez, Org. Lett., 2005, 7, 4787; (b) M. Carril,
R. SanMartin, E. Dominguez and I. Tellitu, Tetrahedron, 2007,
63, 690.
Finally in compound 15aꢂMe2SO all the hydrogen atoms
were introduced in calculated positions and refined in
agreement with the coordinates of the atoms to which they
are bound. Fig. 4–9 report ORTEP3 views of the six structures.
Crystallographic data and refinement parameters for the six
structures are reported in Table 6.
13 N. Hucher, B. Decroix and A. Daich, J. Org. Chem., 2001, 66,
4695.
Computational study
14 F. H. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58, 380.
15 (a) H. Bock, N. Nagel and C. Naether, Z. Naturforsch., Teil B,
1998, 53, 1389; (b) A. Parkin, A. Collins, C. J. Gilmore and
C. C. Wilson, Acta Crystallogr., Sect. B: Struct. Sci., 2008, 64, 66.
16 (a) R. D. Bindal, J. T. Golab and J. A. Katzenellenbogen, J. Am.
Chem. Soc., 1990, 112, 7861; (b) J. B. Nicholas, R. Vance,
E. Martin, B. J. Burke and A. J. Hopfinger, J. Phys. Chem.,
1991, 95, 9803; (c) J. Heyd, W. Thiel and W. Weber, J. Mol.
Struct. (THEOCHEM), 1997, 391, 125; (d) G. Liang, J. P.
Bays and J. P. Bowen, J. Mol. Struct. (THEOCHEM), 1997,
401, 165.
17 T. C. Higgs, A. Parkin, S. Parsons and P. A. Tasker, Acta
Crystallogr., Sect. E: Struct. Rep. Online, 2002, 58, o523.
18 K. B. Wiberg, Advances in Molecular Modeling, ed. D. Liotta, Jai
Press, London, 1988.
19 K. A. Brameld, B. Kuhn, D. C. Reuter and M. Stahl, J. Chem. Inf.
Model., 2008, 48, 1.
The GAUSSIAN03 (Revision B.05)35 package implemented
on a personal computer was used for all the computational
studies concerning the monomers (see Fig. 3, Scheme 8 and
ref. 28). In all cases the level of theory was HF-SCF, the basis
set was 6-311+G(d,p).36 The Berny algorithm37 was used for
the optimization procedure. The reliability of the stationary
points was assessed by the evaluation of the vibrational
frequencies. NBO analyses were performed using the NBO
5.0 program.26 Given that the oxo (4a), amine (8b) and sulfur
derivatives (13a and 14a) show in the solid state three distinct
overall geometries, namely I, II and III (see Scheme 8 and the
X-ray structure discussion in the Solid state characterization
results paragraph), for each modelled molecule sketched in
Fig. 3, three starting geometries were considered and fully
optimized (4a-I/4a-III; 8a-I/8a-III; 13a-I/13a-III; 14a-I/14a-III).
Tables 1–4 list the most interesting geometrical optimized para-
meters, Table 5 reports the relative energy contents of each
optimized species.
20 M. C. Menziani, M. Cocchi and P. G. De Benedetti, J. Mol. Struct.
(THEOCHEM), 1992, 256, 217.
21 V. Petrov, V. Petrova, G. V. Girichev, H. Oberhammer,
N. I. Giricheva and S. Ivanov, J. Org. Chem., 2006, 71, 2952.
22 V. M. Petrov, G. V. Girichev, H. Oberhammer, V. N. Petrova,
N. I. Giricheva, A. V. Bardina and S. N. Ivanov, J. Phys. Chem. A,
2008, 112, 2969.
23 W. L. Duax, C. M. Weeks and D. C. Rohrer, Top. Stereochem.,
1976, 9, 271.
24 V. Bertolasi, V. Ferretti, G. Gilli and P. A. Borea, Cryst. Struct.
Commun., 1982, 11, 1481.
25 C. A. Hunter and J. K. M. Sanders, J. Am. Chem. Soc., 1990, 112,
5525.
26 E. D. Glendening, J. K. Badenhoop, A. E. Reed, J. E. Carpenter,
J. A. Bohmann, C. M. Morales and F. Weinhold, NBO 5.0.,
Theoretical Chemical Institute, University of Wisconsin, Madison,
WI, 2001.
References
1 (a) B. E. Evans, K. E. Rittle, M. G. Bock, R. M. DiPardo,
R. M. Freidinger, W. L. Whittle, G. F. Lundell, D. F. Veber,
P. S. Anderson, R. S. L. Chang, V. J. Lotti, D. J. Cerino,
T. B. Chen, P. J. Kling, K. A. Kunkel, J. P. Springer and
J. Hirshfield, J. Med. Chem., 1988, 31, 2235; (b) A. A. Patchett
and R. P. Nargund, Annu. Rep. Med. Chem., 2000, 35, 289.
2 V. Fedi, A. Guidi and M. Altamura, Mini-Rev. Med. Chem., 2008,
8, 1464.
27 A. Bondi, J. Phys. Chem., 1964, 68, 441.
28 To assess the role of steric effects in driving the tricyclic arrange-
ment, geometry optimizations were performed of two further
molecular models having Z = CH2 and C(CH3)2 (CH2-I, CH2-III,
C(CH3)2-I and C(CH3)2-III). Assuming that the steric hindrance
exerted by the methylene group is negligible when Z = CH2, the
electronic factors should prevail in driving the molecular geometry,
while in the bulkier derivative (Z = C(CH3)2) the steric repulsions
3 M. Paris, M. Porcelloni, M. Binaschi and D. Fattori, J. Med.
Chem., 2008, 51, 1505.
4 (a) D. Giannotti, G. Viti, P. Sbraci, V. Pestellini, G. Volterra,
F. Borsini, A. Lecci, A. Meli, P. Dapporto and P. Paoli, J. Med.
Chem., 1991, 34, 1356; (b) D. Giannotti, G. Viti, R. Nannicini,
V. Pestellini and D. Bellarosa, Bioorg. Med. Chem. Lett., 1995, 5,
1461; (c) M. Altamura, P. Dapporto, A. Guidi, N. J. S. Harmat,
L. Jierry, E. Libralesso, P. Paoli and P. Rossi, New J. Chem., 2008,
32, 1617.
5 E. J. Warawa, B. M. Migler, C. J. Ohnmacht, A. L. Needles,
G. C. Gatos, F. M. McLaren, C. L. Nelson and K. M. Kirkland,
J. Med. Chem., 2001, 44, 372.
6 (a) K. Hiroshi and M. Mitsuru, Jap. Pat., JP 45015983, 1970;
(b) E. Enders, F. Muth and H. Herlinger, GB Pat., GB 1013046,
1963; (c) K. Nagarajan, V. Ranga, A. Venkarlu and R. K. Shah,
Indian J. Chem., 1974, 12, 252.
7 (a) J. Jabicky, The Chemistry of Amides, Interscience, London,
1970; (b) Y. Hiroshi, U. Norihiro, N. Jun, S. Hiroyuki,
K. Yoshihiko, K. Nozomu, Y. Kentaro, A. Makoto and
W. Tatsuo, J. Med. Chem., 1992, 35, 2496.
should play the leading role. As
a matter of fact CH2-I
is more stable than CH2-III. As for the bulkier derivative
(Z = C(CH3)2), the steric effects appear important as provided
by the re-arrangement of the both the I and III geometries during
the minimization processes, which indicate the C(CH3)2-III as the
most stable conformer. Finally the electronic picture of each
model, provided by the NBO analysis, is in keeping with that
already presented and discussed.
29 CrysAlis CCD, Oxford Diffraction Ltd., Version 1.171.29.2
(release 20.01.2006 CrysAlis171.NET).
30 CrysAlis RED, Oxford Diffraction Ltd., Version 1.171.29.2
(release 20.01.2006 CrysAlis171. NET).
31 A. Altomare, G. L. Cascarano, C. Giacovazzo, A. Guagliardi,
A. G. Moliterni, M. C. Burla, G. Polidori, M. Camalli and
R. Spagna, J. Appl. Crystallogr., 1999, 32, 115.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009
2230 | New J. Chem., 2009, 33, 2219–2231

View Online
32 G. M. Sheldrick, SHELX 97, University of Gottingen, Germany,
¨
1997.
33 M. Nardelli, J. Appl. Crystallogr., 1995, 28, 659.
K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain,
O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford,
J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill,
B. G. Johnson, W. Chen, M. W. Wong, C. Gonzalez and
J. A. Pople, GAUSSIAN 03 (Revision B.5), Gaussian, Inc.,
Wallingford, CT, 2004.
34 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
35 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr.,
T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam,
S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi,
G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada,
M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida,
T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li,
J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. Ochterski, P. Y. Ayala,
36 T. Clark, J. Chandrasekhar, G. W. Spitznagel and P. v.
R. Schleyer, J. Comput. Chem., 1983, 4, 294.
37 C. Peng, P. Y. Ayala, H. B. Schlegel and M. J. Frisch, J. Comput.
Chem., 1996, 17, 49.
ꢀc
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 2219–2231 | 2231