Sulfonamide-pyridine-N-oxide cocrystals

Article abstract of DOI:10.1021/cg200094x

Hydrogen-bond motifs for the supramolecular synthesis of sulfonamide-pyridine-N-oxide complexes were derived from the analysis of cocrystal structures of aryl sulfonamides and pyridine-N-oxides. The sulfonamide NH donor persistently interacts with the N-oxide acceptor via a N-H ··· O hydrogen bond, though the exact geometry of the motif, such as discrete (D), infinite chain C(4), cyclic motifs R ₄² (8) or R₂² (4), varies from structure to structure. Surprisingly, the two-point R₂² (8) cyclic motif of N-H···O and C- H···O hydrogen bonds, analogous to the heterosynthon in carboxamide-pyridine-N-oxide cocrystals, was not observed in sulfonamide-N-oxide cocrystals. The utility of this model study is demonstrated by making cocrystals of Furosemide with 4,4′-bipyridine-N,N′-dioxide and isonicotinamide-N-oxide.

Full text of DOI:10.1021/cg200094x

ARTICLE
pubs.acs.org/crystal
SulfonamideꢀPyridine-N-oxide Cocrystals
N. Rajesh Goud, N. Jagadeesh Babu, and Ashwini Nangia*
School of Chemistry, University of Hyderabad, Prof. C. R. Rao Road, Gachibowli, Hyderabad 500 046, India
S Supporting Information
b
ABSTRACT: Hydrogen-bond motifs for the supramolecular
synthesis of sulfonamideꢀpyridine-N-oxide complexes were
derived from the analysis of cocrystal structures of aryl sulfo-
namides and pyridine-N-oxides. The sulfonamide NH donor
persistently interacts with the N-oxide acceptor via a NꢀH
O
3 3 3
hydrogen bond, though the exact geometry of the motif, such as
2
2
discrete (D), infinite chain C(4), cyclic motifs R4 (8) or R2 (4),
varies from structure to structure. Surprisingly, the two-point
2
R2 (8) cyclic motif of NꢀH O and CꢀH O hydrogen
3 3 3
3 3 3
bonds, analogous to the heterosynthon in carboxamideꢀ
pyridine-N-oxide cocrystals, was not observed in sulfon-
amideꢀN-oxide cocrystals. The utility of this model study is
demonstrated by making cocrystals of Furosemide with 4,40-bipyridine-N,N0-dioxide and isonicotinamide-N-oxide.
Scheme 1. Strong Hydrogen-Bond Homosynthons and
Heterosynthons with Their Occurrence Probability Indicated
’ INTRODUCTION
The identification of new and robust supramolecular synthons
is the first step in the crystal engineering of organic solids.
Following up on Desiraju’s1 supramolecular synthon concept,
Zaworotko2 subclassified synthons as heterosynthons (those
between complementary unlike functional groups) and homo-
synthons (between the same functional group). Some common
examples of homo- and heterosynthons and their probability
statistics3 in the Cambridge Structural Database4 are given in
Scheme 1. Ideally, one would like to find out the likelihood of a
particular synthon for retrosynthetic planning.5 Cyclic synthons,
or hydrogen bond ring motifs, can be classified according to the
graph set notation6 as R42(8), R22(7), etc. Etter’s7 hydrogen bond
rules guide the pairing-off of donors and acceptors in a multi-
functional system: the best donor in the molecule preferentially
interacts with the best acceptor in the system, the second best
donorꢀacceptor group hydrogen bond next, and so on. For
example, in carboxylic acidꢀcarboxamide heterosynthon, the
strongest donor (COOH) hydrogen bonds to the best acceptor
(amide CdO) and then the next best donorꢀacceptor, amide
NH and acid CdO, pair up to give the R22(8) motif of 47%
occurrence probability.8 Similarly, acidꢀpyridine heterosynthon
of R22(7) graph set is the most popular supramolecular synthon in
the crystal engineer’s toolkit with >90% success rate.9
molecules.11 The sulfonamide group occupies a place of promi-
nence in drug molecules, being the common functional moiety of
sulfa drugs.12 Whereas a few isolated cases of sulfonamide
hydrogen bonding to other functional groups, such as COOH,
CONH2, urea, etc., are reported13 there is no specific hetero-
synthon associated with the SO2NH2 group. Sulfoxide as a
potent cocrystal former for the NH functional group via a
Some time ago we noted that the carboxamide functional
group does not have an equivalent high-probability motif, and
developed the novel carboxamideꢀpyridine-N-oxide heterosyn-
thon of R22(8) geometry.10 The strong amide NH to N-oxide
NꢀH OdS hydrogen bond was recently reported.14 Our
3 3 3
NꢀH O motif has a good occurrence probability of 70%
3 3 3
compared to 4% for amideꢀpyridine. Carboxylic acids and
carboxamide are the most popular organic functional groups
for crystal design and they are often found in pharmaceutical
Received: January 22, 2011
Revised:
March 5, 2011
Published: March 10, 2011
r
2011 American Chemical Society
1930
dx.doi.org/10.1021/cg200094x Cryst. Growth Des. 2011, 11, 1930–1939
|

Crystal Growth & Design
ARTICLE
Scheme 2. Homo- vs Heterosynthon Competition for the Sulfonamide Groupa
a (a) In sulfonamideꢀCOOH (or CONH2), the hydrogen bonds are of the strong OꢀH O/NꢀH O type in the target heterosynthon and the
3 3 3
3 3 3
competing homosynthons. (b) In sulfonamideꢀpyridine-N-oxide, the target sulfonamideꢀpyridine-N-oxide has strongest donorꢀstrongest acceptor
pairing, whereas the competing homosynthons are not so strong. This should give a higher success rate for the sulfonamide group to hydrogen bond with
pyridine-N-oxide in co-crystallization experiments.
objective was to use pyridine-N-oxide as complementary hydro-
gen bond acceptor for the SO2NH2 group. An advantage with
pyridine-N-oxide compared to COOH and CONH2 as a com-
plementary functional group is that homosynthon competition is
minimal because the pyridine-N-oxide lacks strong donor groups
of its own (Scheme 2).
A stable heterosynthon assembly is based on the fact that the
dissimilar functional groups will aggregate better compared to
identical ones. Observations such as (1) the carboxamideꢀ
pyridine-N-oxide heterosynthon has hydrogen bond strength
greater than the constituent homosynthons by about 3 kcal/
mol,10 and (2) that pyridine-N-oxides are able to disrupt the
strong urea R-network15 suggested that the pyridine-
N-oxide group will be a good complementary partner for
the sulfonamide group to prepare heteromeric cocrystals
(Scheme 3).
’ RESULTS
A search of the CSD4 (August 2010 update) for the sulfon-
amide group with the filters ‘3D coordinates determined,
R-factor <0.1, organic molecules only’ gave 2090 hits, which
included primary and secondary sulfonamide structures. This
sulfonamide subdatabase contained only 34 cocrystals without
any solvent/water inclusion, of which 6 are with primary
sulfonamides. There are 75 multicomponent complexes such
as salt hydrates and cocrystal solvates, 165 solvates, and 61 salt
structures. The propensity of sulfonamide to form cocrystals is
1.63%, molecular complexes 3.59%, solvates 7.89%, and salts
2.91%. This means that the sulfonamide functional group is
relatively less explored for cocrystal engineering. Moreover, the
current version of the Cambridge database does not contain
even a single hit of a primary sulfonamide and pyridine N-oxide
1931
dx.doi.org/10.1021/cg200094x |Cryst. Growth Des. 2011, 11, 1930–1939

Crystal Growth & Design
ARTICLE
Scheme 3. Primary Sulfonamide and Pyridine-N-oxide coformers Gave Five Cocrystals and a Quarter Hydrate As Listed above
(abbreviations are used throughout the paper)
functional group in the same crystal structure, suggesting that
this class of heteromeric cocrystals is completely unexplored.
We report six novel cocrystals of a few model sulfonamides and
a well-known drug molecule Furosemide (Lasix) with pyridine-
N-oxides as coformers. The objective was to understand the
nature of hydrogen bonding in cocrystal structures and to know
if a particular motif is strong and recurring to be reliably used for
crystal design.
The most common method of obtaining cocrystals is to
dissolve the components in a suitable solvent for single crystals
of the binary phase to appear after slow evaporation of solvent(s).
Gentle warming is necessary to dissolve the solids, and an
antisolvent is added to accelerate crystallization. This empirical,
trial-and-error method, referred to as solution crystallization, was
repeated with several solvents until there was evidence of a new
chemical complex or cocrystal based on DSC, IR, powder X-ray
diffraction, and single-crystal X-ray structure. Precipitation of the
individual components instead of the desired cocrystal and
formation of undesired solvates/hydrates are common difficul-
ties in the solution-based approach. Solid-state grinding16 over-
comes the problem of solvent inclusion, but the product is
usually microcrystalline for routine single-crystal X-ray structure
determination. Adding a few drops of solvent during grinding/
kneading, referred to as liquid-assisted grinding,17 accelerates
cocrystal formation due to lubrication.
Cocrystals of 2-CBS, 4-MBS, 2-BBS, and PTS were obtained
by the traditional solution crystallization method (see Scheme 3
for chemical diagrams). However, this method was not successful
with FUROS. Liquid-assisted grinding with acetonitrile solvent
gave 0.25 hydrate with BPNO and a cocrystal with INICNO (see
Scheme 3 for chemical diagrams and Experimental Section for
cocrystallization details). The molecular stoichiometry and com-
position of each cocrystal was confirmed from its X-ray crystal
structure. Attempts with increasing the temperature of the solid-
state reaction18 or melt cocrystallization19 did not yield any new
adduct. Crystallographic parameters and hydrogen bond metrics
are listed in Table 1 and Table 2. Since both SdO and N-oxide
acceptors are present in the system, the O atom is subscripted to
identify the functional group as Osulfonamide or ONoxide whenever
there is ambiguity.
Crystal Structure Analysis. 2-Chlorobenzene Sulfonamideꢀ
Bipyridine-N,N0-dioxide (2-CBS:BPNO 1:1). This crystal structure
in tetragonal space group P43212 contains one molecule each
of 2-CBS and BPNO in the asymmetric unit. The 2-CBS and
BPNO molecules form a zigzag tape along the crystallographic
[001] axis via a NꢀH ONoxide hydrogen bond (1.80 Å, 173°).
3 3 3
The normal sulfonamide dimer of NꢀH Osulfonamide is
3 3 3
absent; one of the sulfonamide O is involved in intramolecular
CꢀCl O contact of 3.06 Å.20 The cocrystal units are
3 3 3
connected through CꢀH ONoxide interactions of BPNO
3 3 3
(Figure 1).
2-Bromobenzene SulfonamideꢀBipyridine-N,N0-dioxide (2-
BBS:BPNO 1:0.5). 2-BBS and BPNO molecules form a two-
dimensional array of tetrameric ring motif R24(8) of NꢀH
3 3 3
ONoxide hydrogen bonds. There are auxiliary O Br interac-
3 3 3
tions in a 10-member ring as shown in Figure 2. Thus while
2-CBS has single NꢀH ONoxide hydrogen bond, BBS makes a
3 3 3
ring motif with the coformer.
4-Methoxybenzene Sulfonamideꢀbipyridine-N,N0-dioxide
(4-MBS:BPNO 1:0.5). 4-MBS molecules form a linear tape along
the [010] axis through one point NꢀH ONoxide and Nꢀ
3 3 3
H
Osulfonamide hydrogen bond (1.99 Å, 166°; 1.74 Å, 167°).
3 3 3
Four 4-MBS and 2 BPNO molecules make large R66(24) rings
with the p-anisyl groups projecting above and below the ring
plane (Figure 3). The commonly occurring sulfonamide dimer is
absent in this structure. There is also an NꢀH Osulfonamide
3 3 3
C(4)chain that can be discerned in the structure.
4-Toluene Sulfonamideꢀbipyridine-N,N0-dioxide (PTS:BPNO
2:1). PTS and BPNO molecules form tetrameric ring motif
R24(8)of NꢀH ONoxide hydrogen bonds through two sym-
3 3 3
metry independent TS molecules and one BPNO (1.85 Å, 173°,
1.97 Å, 164°, 1.86 Å, 164°, 1.80 Å, 171°) as shown in Figure 5.
The sulfonamide O atoms are engaged in R22(9) motif of Cꢀ
H
O interactions (Figure 6).
3 3 3
1932
dx.doi.org/10.1021/cg200094x |Cryst. Growth Des. 2011, 11, 1930–1939

Crystal Growth & Design
ARTICLE
Table 1. X-ray Crystal Structure Data
2-CBSꢀBPNO
2-BBSꢀBPNO
4-MBSꢀBPNO
PTSꢀBPNO
FUROSꢀBPNOꢀH2O
FUROSꢀINICNO
(1:1)
(1:0.5)
(1:0.5)
(2:1)
(1:1:0.25)
(1:1)
empirical formula
(C6H6O2NSCl)
(C6H6O2NSBr)
(C7H9O3NS)
2(C7H9O2NS)
(C12H11O5N2ClS)
(C12H11O5N2ClS)
3
3
3
3
3
3
(C10H8N2O2)
379.81
0.5(C10H8N2O2)
660.36
0.5(C10H8N2O2)
(C10H8N2O2)
530.61
(C10H8N2O2) 0.5(O)a
(C6H6O2N2)
468.87
3
fw
562.61
526.92
cryst syst
space group
T (K)
a (Å)
tetragonal
P43212
100(2)
7.8871(5)
7.8871(5)
51.100(7)
90.00
triclinic
P1
monoclinic
P21/c
monoclinic
P21/c
triclinic
P1
triclinic
P1
298(2)
8.181(3)
8.421(3)
10.250(4)
98.774(6)
95.521(6)
114.520(6)
1
100(2)
100(2)
10.3170(9)
12.3913(10)
20.0553(17)
90.00
100(2)
8.0230(6)
8.3201(6)
17.5479(12)
77.7340(10)
77.5200(10)
81.2860(10)
2
100(2)
8.7426(8)
5.2061(5)
27.7269(19)
90.00
5.2214(4)
9.5860(8)
20.1540(17)
85.2220(10)
88.7590(10)
74.8260(10)
2
b (Å)
c (Å)
R (deg)
β (deg)
γ (deg)
Z
90.00
102.935(2)
90.00
103.7280(10)
90.00
90.00
8
2
4
V (Å3)
Dcalcd (g cmꢀ3
3178.7(5)
1.587
624.9(4)
1.755
1229.96(18)
1.519
2490.6(4)
1.415
1110.71(14)
1.576
970.20(14)
1.605
)
no. of reflns collected 32933
6340
11917
25031
11547
10099
no. of obsd reflns
no. of unique reflns
R1 [ I > 2σ(I)]
wR2 [all]
3142
3134
0.0325
0.0781
234
2406
2422
4856
4339
3808
2207
2290
4493
4043
3566
0.0299
0.0764
163
0.0397
0.0335
0.0376
0.0317
0.1036
0.0900
0.0973
0.0849
no. of params
173
343
341
338
GOF
1.150
1.110
1.082
1.017
1.059
1.038
a Water was assigned 0.25 sof in crystal structure. H atoms are omitted. SHELX refinement gave a value of 0.5(O).
FurosemideꢀBipyridine-N,N0-dioxideꢀ0.25Hydrate (FUROSꢀ
BPNOꢀH2O 1:1:0.25). The presence of partial water occupancy
in the crystal structure was initially suggested by unaccounted for
electron density of about 1.67 Å3 in the structure during
refinement and checking cycles.21 Placing a partial water at this
site of 0.25 atom occupancy gave better structure convergence
and R-factor. Hence the quarter hydrate stoichiometry is fixed for
this cocrystal. The crystal structure contains a prominent tetra-
meric ring motif of graph set R22(4) connected through strong
Vibrational Spectroscopic Characterization. All com-
pounds were characterized by IR, Raman, and NIR spectros-
copy and by changes in the signature peaks associated with the
transformation of the sulfonamide and the coformer to the new
cocrystal structure. The NꢀH stretching vibrations of the
sulfonamide group showed the expected red shift of 10ꢀ
40 cmꢀ1 due the formation of the stronger NꢀH
ONoxide
3 3 3
hydrogen bond in the cocrystal compared to NꢀH
3 3 3
Osulfonamide, i.e., the NꢀH bond is weaker in the cocrystal. This
bathochromic shift is observed for all cocrystals (see Figure S2
in the Supporting Information). There are clear differences in
the solid-state spectra of the starting components compared to
that of the product cocrystals (see Table S1 in the Supporting
Information).
NꢀH ONoxide hydrogen bond (2.04 Å, 140.2°; Figure 7). The
3 3 3
FUROS molecular conformation is frozen by an intramolecular
NꢀH O bond (1.91 Å, 132.3°) from the amine NH to the
3 3 3
carboxylic acid CdO of S(6) notation. Two partial occupancy
water molecules act as bridges to two N-oxide groups, which is
chemically reasonable interaction geometry. Further, the loss of
water could be detected in TGA (see Figure S1 in the Supporting
Information). Incidentally, the structure contains the sulfonamide
Similar to the IR spectra, the Raman spectra showed bath-
ochromic shift in the NꢀH stretching peaks due to the weaken-
ing of the NꢀH bond after stronger hydrogen bonding with the
pyridine-N-oxide acceptor. However, due to the weak intensities
of the Raman resonances the symmetric and asymmetric NꢀH
peaks merged together to give a single broad band between 3400
and 3200 cmꢀ1 in all the cocrystals and the hydrate (see Figure
S3 in the Supporting Information). Attempts to resolve the
NꢀH peaks by increasing the laser intensity invariably caused
sample burn. The changes in the NꢀH stretch, bend and SdO
stretch vibrations are listed in Table S2 in the Supporting
Information.
NꢀH O dimer (1.95 Å, 153.2°) common in sulfonamide
3 3 3
crystal structures (Figure 8). The R22(4) ring motifs are connected
through OꢀH ONoxide (1.62 Å, 157.9°) shown in Figure 9.
3 3 3
FurosemideꢀIsonicotinamide-N-oxide
(FUROSꢀINICNO
1:1). The FUROS molecule is disordered at the furan ring. The
R22(8) ring motif between the carboxylic group of FUROS and
the carboxamide group of INICNO is the main heterosynthon
(NꢀH O 1.87 Å, 167°; OꢀH O 1.60 Å, 168°). The
3 3 3
3 3 3
sulfonamide group connects FUROS to the N-oxide moiety on
the other side (NꢀH ONoxide 1.80 Å, 162°) as shown in the
NIR spectra were analyzed in the 4000ꢀ7000 cmꢀ1 range.
The NꢀH overtone vibrations at 6400ꢀ6200 cmꢀ1 (see
Figure S3 in the Supporting Information) are consistent with
the fundamental NꢀH stretch resonances in the IR and
Raman spectra wherein the bathochromic shift of about
3 3 3
Figure 10.
The FUROS and INICNO molecules extend to make a larger
R44(32) [R22(8)] ring motif, which in turn is connected through
CꢀH O R22(8) dimer as shown in Figure 11.
3 3 3
1933
dx.doi.org/10.1021/cg200094x |Cryst. Growth Des. 2011, 11, 1930–1939

Crystal Growth & Design
ARTICLE
Table 2. Hydrogen Bond Distance and Angle (neutron-normalized NꢀH, OꢀH, and CꢀH distance)
DꢀH
A
D
A (Å)
H
A (Å)
DꢀH A (deg)
symmetry code
3 3 3
3 3 3
3 3 3
3 3 3
2-CBSꢀBPNO (1:1)
N1ꢀH1A O4
2.806(3)
2.839(3)
3.388(3)
3.497(3)
2.825(3)
3.428(3)
3.360(3)
3.383(3)
3.233(3)
1.80
1.87
2.51
2.42
2.36
2.35
2.29
2.40
2.23
172.9
158.0
136.5
168.4
103.6
173.0
169.2
149.1
153.1
1/2 þ y, 1/2 ꢀ x, ꢀ3/4 þ z
1 þ x, ꢀ1 þ y, ꢀ1 þ z
ꢀ1 þ x, y, z
3 3 3
N1ꢀH1B O3
3 3 3
C3ꢀH3 N1
3 3 3
C5ꢀH5 O1
ꢀ1/2 þ x, 1/2 ꢀ y, 1/4 ꢀ z
3 3 3
a
C6ꢀH6 O1
3 3 3
C8ꢀH8 O3
y, 1 þ x, 2 ꢀ z
3 3 3
C12ꢀH12 O4
3/2 ꢀ x, 1/2 þ y, 7/4 ꢀ z
y, 1 þ x, 2 ꢀ z
3 3 3
C13ꢀH13 O3
3 3 3
C16ꢀH16 O4
3/2 ꢀ x, ꢀ1/2 þ y, 7/4 ꢀ z
3 3 3
2-BBSꢀBPNO (1:0.5)
2.00
1.80
2.40
N1ꢀH1A O3
2.812(3)
2.806(3)
3.108(3)
135.6
172.5
121.3
1 ꢀ x, 1 ꢀ y, ꢀz
x, ꢀ1 þ y, z
ꢀ1 ꢀ x, 1 ꢀ y, 2 ꢀ z
3 3 3
N1ꢀH1B O3
3 3 3
C2ꢀH2 O2
3 3 3
4-MBSꢀBPNO (1:0.5)
N1ꢀH1A O3
2.737(2)
2.988(2)
3.321(2)
1.74
166.7
165.7
157.4
x, ꢀ1 þ y, z
x, 1 þ y, z
3 3 3
N1ꢀH2A O2
1.99
3 3 3
C12ꢀH12 O1
2.29
ꢀ1 þ x, 1 þ y, z
3 3 3
PTSꢀBPNO (2:1)
N2ꢀH1A O5
2.850(2)
2.809(2)
2.858(2)
2.961(2)
2.893(2)
2.903(2)
3.180(2)
3.324(2)
1.86
1.80
1.85
1.97
2.48
2.49
2.21
2.44
164.0
170.7
172.5
164.1
101.0
101.0
146.5
137.3
ꢀ1 þ x, y, z
3 3 3
N2ꢀH2A O6
x, 3/2 ꢀ y, 1/2 þ z
x, 3/2 ꢀ y, 1/2 þ z
3 3 3
N1ꢀH3A O6
3 3 3
N1ꢀH4A O5
ꢀ1 þ x, y, z
3 3 3
a
C2ꢀH2 O1
3 3 3
a
C9ꢀH9 O4
3 3 3
C21ꢀH21 O1
1 þ x, 3/2 ꢀ y, ꢀ1/2 þ z
1 þ x, y, z
3 3 3
C23ꢀH23 O4
3 3 3
FUROSꢀBPNO-H2O (1:1:0.25)
N1ꢀH1A O6
2.897(2)
2.886(2)
2.698(2)
2.564(2)
2.819(2)
3.197(2)
3.239(2)
3.239(3)
3.178(2)
3.703(2)
3.330(2)
3.256(3)
2.04
140.2
153.2
132.3
157.9
104.1
124.0
151.7
154.1
165.5
160.7
155.2
156.7
ꢀ2 þ x, y, 1 þ z
3 3 3
N1ꢀH1B O2
1.95
ꢀ1 ꢀ x, 1 ꢀ y, 2 ꢀ z
3 3 3
a
N2ꢀH2 O3
1.91
3 3 3
O4ꢀH4 O6
1.62
2 ꢀ x, 2 ꢀ y, 1 ꢀ z
3 3 3
a
C3ꢀH3 O1
2.35
3 3 3
C3ꢀH3 O2
2.46
ꢀx, 1 ꢀ y, 2 ꢀ z
1 ꢀ x, 2 ꢀ y, 1 ꢀ z
x, 1 þ y, z
1 þ x, y, ꢀ1 þ z
ꢀx, 1 ꢀ y, 1 ꢀ z
1 þ x, y, ꢀ1 þ z
1 ꢀ x, 1 ꢀ y, 1 ꢀ z
3 3 3
C11ꢀH11 O3
2.24
3 3 3
C12ꢀH12 O7
2.22
3 3 3
C16ꢀH16 O1
2.11
3 3 3
C18ꢀH18 Cl1
2.66
3 3 3
C19ꢀH19 O1
2.31
3 3 3
C22ꢀH22 O7
2.23
3 3 3
FUROSꢀINICNO (1:1)
N2ꢀH2A O3
2.866(2)
3.074(2)
2.694(2)
2.784(2)
2.569(2)
3.277 (2)
3.414(2)
2.828(2)
1.87
2.12
1.87
1.80
1.60
2.35
2.33
2.36
167.1
156.1
135.7
162.4
167.7
142.0
176.5
103.8
ꢀ2 þ x, y, z
3 3 3
N2ꢀH2B N4
x, ꢀ1 þ y, z
3 3 3
a
N3ꢀH3A O3
3 3 3
N4ꢀH4A O1
1 ꢀ x, 2 ꢀ y, ꢀz
2 þ x, y, z
3 3 3
O4ꢀH4C O2
3 3 3
C4ꢀH4 O5
x, ꢀ1 þ y, z
3 3 3
C5ꢀH5 O1
2 ꢀ x, 1 ꢀ y, ꢀz
3 3 3
a
C12ꢀH12 O5
3 3 3
a Intramolecular hydrogen bond.
50ꢀ70 cmꢀ1 were clearly noted in the cocrystals due to the
weakening of the NꢀH bond. A summary of the major
overtone bands are listed in Table S3 in the Supporting
Information.
1934
dx.doi.org/10.1021/cg200094x |Cryst. Growth Des. 2011, 11, 1930–1939

Crystal Growth & Design
ARTICLE
Figure 1. 2-CBS and BPNO molecules are connected via a strong
NꢀH ONoxide hydrogen bond and such binary units extend via a
3 3 3
dimeric CꢀH ONoxide motif.
3 3 3
Figure 2. Tetrameric R24(8)of NꢀH ONoxide hydrogen bonds ring
3 3 3
motifs assemble the binary adduct.
Figure 4. Helix formation through alternating 4-MBS and BPNO
molecules.
Figure 5. Tetrameric R24(8) rings of NꢀH O hydrogen bonds
3 3 3
between PTS and BPNO molecules. Symmetry-independent PTS
molecules are shown as capped-stick and ball-stick models.
Figure 3. Hydrogen-bonded macrocyclic rings sustain the layer struc-
ture motif of 4-MBSꢀBPNO. A part of this crystal structure in the space
group P21/c may be discerned to contain a helix made up of NꢀH
O
3 3 3
and CꢀH O interactions along [001] (see Figure 4).
3 3 3
’ DISCUSSION
The formation of the sulfonamideꢀpyridine-N-oxide hetero-
synthon appears to be very reliable due to the enhanced hydro-
gen bond donor and acceptor ability of the sulfonamide NH and
N-oxide, respectively. The formation of the dominant NꢀH
3 3 3
ON-oxide hydrogen bond in the crystal structure is based on the
strongest donor (sulfonamide NH) interacting with the strongest
acceptor (N-oxide).6 The exact geometry of the sulfonamide motif,
i.e., discrete NꢀH O (D), infinite chain C(4), cyclic motifs
3 3 3
R24(8) or R22(4) varies from one structure to another and it is
difficult to anticipate their presence in a particular structure.
Surprisingly, the two-point NꢀH O þ CꢀH O hetero-
Figure 6. CꢀH O synthon of R22(9) motif in the crystal structure.
3 3 3
3 3 3
3 3 3
synthon of R22(8) geometry, similar to the carboxamideꢀ
structure. The nonplanarity of the SO2NH2 group compared
to the planar CONH2 perhaps prevents the two-point motif
pyridine-N-oxide heterosynthon was not observed in any
1935
dx.doi.org/10.1021/cg200094x |Cryst. Growth Des. 2011, 11, 1930–1939

Crystal Growth & Design
ARTICLE
Figure 7. BPNO molecules are connected to FUROS through tetrameric R22(4) ring motif on one side (right) and to another BPNO molecule through
water bridges on the other side (left).
Figure 8. FUROS molecules connected through strong sulfonamide
dimer R22(8) motif.
Figure 11. R44(32) [R22(8)] motif of NꢀH O and OꢀH
O
3 3 3
3 3 3
hydrogen bonds and CꢀH O R22(8) dimer in FUROSꢀINICNO
3 3 3
structure.
mediated between molecules (say drug and coformer) con-
taining these two functional groups via a NꢀH
ONoxide
3 3 3
Figure 9. Tetrameric R22(4) ring motifs are connected through OꢀH
hydrogen bond or ring motif. A second preliminary observa-
tion is the lack of polymorphism in the cocrystals studied
(based on the crystallization experiments listed in Table S4
in the Supporting Information), compared to relatively
promiscuous multiple forms in some sulfonamides studied
by us recently,22 and for sulfa drugs in general.12
3 3 3
O
Noxide hydrogen bond.
The newly identified sulfonamideꢀN-oxide heterosynthon
is illustrated with the loop diuretic Furosemide.23 The recur-
ring sulfonamide NꢀH
O R22(8) dimer in three poly-
3 3 3
morphs of this drug24 is present in the hydrate with BPNO.
The anti NH makes an R22(4) ring motif with the N-oxide. The
situation in FUROSꢀINICNO cocrystal is quite predictable:
the strongest SO2NH2 donor bonds to the N-oxide acceptor,
and the COOH and CONH2 groups of the drug and coformer
make the acidꢀamide heterosynthon at the next level of
hierarchy.
Figure 10. INICNO coformer is connected to FUROS by amideꢀacid
’ CONCLUSIONS
R22(8) motif and NꢀH N hydrogen bond.
3 3 3
The ability of aromatic primary sulfonamides to make a strong
hydrogen bond with pyridine-N-oxides is demonstrated for the
first time and then this synthon is used to make cocrystals of the
loop diuretic drug Furosemide. The scope of this model study
will be easily extendable to sulfa drugs. The fact that pyridine-N-
oxide functional group is present in some drug molecules and
discovery candidates25 leads to the possibility of making two-
drug cocrystals.
with an aromatic partner group. Whereas the six heavy atoms
in the SO2NH2 dimer adopt the chair cyclohexane confor-
mation, the arrangement is planar in the amide dimer. Even
though a dominant sulfonamideꢀpyridine-N-oxide motif
could not be identified in this subset of cocrystal structures,
it is clear that bimolecular recognition can be reproducibly
1936
dx.doi.org/10.1021/cg200094x |Cryst. Growth Des. 2011, 11, 1930–1939

Crystal Growth & Design
ARTICLE
Table 3. Summary of Experimental Conditions and Results
S. no.
1
experimental conditions
product obtained
2-chlorobenzene sulfonamide (2-CBS) (0.1 mmol) þ
Bipyridine-N,N’-dioxide BPNO (0.1 mmol)
MeOH, 4 mL, room temperature (RT), solution crystallization (SC)
EtOH, 4 mL, RT, SC
no cocrystal observed
no cocrystal observed
BPNO hydrate
i-PrOH, 4 mL, RT, SC
n-PrOH, 4 mL, RT, SC
1:1 cocrystal after 4ꢀ5 d
2
3
2 bromobenzene sulfonamide (2-BBS) (0.1 mmol) þ
bipyridine-N,N’-dioxide BPNO (0.1 mmol)
MeOH, RT, 3 mL, SC
no cocrystal observed
1:0.5 cocrystal after 2ꢀ3 d
BPNO hydrate
EtOH, RT, 4 mL, SC
n-PrOH, RT, 4 mL, SC
DMF, RT, 3 mL, SC
no cocrystal observed
4-methoxybenzene sulfonamide (4-MBS) (0.1 mmol) þ
bipyridine-N,N’-dioxide BPNO (0.1 mmol)
MeOH, RT, 4 mL, SC
BPNO hydrate
EtOH, RT, 4 mL, SC
4-MBS crystals
i-PrOH, RT, 5 mL, SC
4-MBS crystals
n-PrOH, RT, 4 mL, SC
1:0.5 cocrystal after 4ꢀ5 d
no cocrystal observed
no cocrystal observed
DMF, RT, 3 mL,SC
DMSO, RT, 3 mL, SC
4
5
p-toluene sulfonamide (PTS) (0.1 mmol) þ
bipyridine-N,N’-dioxide BPNO (0.1 mmol)
MeOH, 5 mL, RT, SC
no cocrystal observed
no cocrystal observed
PTS
EtOH, 5 mL, RT, SC
n-PrOH, 4 mL, RT, SC
i-buOH, 7 mL, RT, SC
2:1 cocrystal after 4ꢀ5 d
furosemide (FUROS) (0.1 mmol) þ bipyridine-N,N’-dioxide
(BPNO) (0.1 mmol)
MeOH, 9 mL, RT, SC
FUROS Form 1
Abs. EtOH, 6 mL, RT, SC
BPNO hydrate
DMSO, 3 mL, RT, SC
no cocrystal observed
no cocrystal observed
1:1:0.25 hydrate after 4 d
DMF, 3 mL, RT, SC
4ꢀ5 drops CH3CN, liquid-assisted grinding (LAG)
MeOH, 9 mL, RT, SC
4ꢀ5 drops CH3CN, LAG
EtOH, 6 mL, SC, RT
FUROS Form 1
4ꢀ5 drops CH3CN, LAG
no cocrystal observed
FUROS Form 1
i-PrOH, 6 mL, RT, SC
4ꢀ5 drops CH3CN, LAG
n-PrOH, 6 mL, RT, SC
6
furosemide (FUROS) (0.1 mmol) þ Isonicotinamide N-oxide
(INICNO) (0.1 mmol)
4ꢀ5 drops CH3CN, LAG
no cocrystal observed
1:1 cocrystal after 4 d
no cocrystal observed
no cocrystal observed
no cocrystal observed
no cocrystal observed
MeOH, 6 mL, RT, SC
4ꢀ5 drops CH3CN, LAG
EtOH, 6 mL, RT, SC
4ꢀ5drops CH3CN, LAG
n-PrOH, 5 mL, RT, SC
4ꢀ5 drops CH3CN, LAG
i-PrOH, 6 mL, RT, SC
4ꢀ5 drops CH3CN, LAG
i-BuOH, 6 mL, RT, SC
4ꢀ5 drops CH3CN, LAG
n-BuOH, 6 mL, RT, SC
1937
dx.doi.org/10.1021/cg200094x |Cryst. Growth Des. 2011, 11, 1930–1939

Crystal Growth & Design
ARTICLE
Table 4. IR Stretching Frequencies of NꢀH Group in Co-
crystallographic CIF files. This material is available free of charge
via the Internet at http://pubs.acs.org.
crystal Compared to the Starting Sulfonamide
primary sulfonamide
NꢀH stretch in
’ AUTHOR INFORMATION
NꢀH stretch
N-oxide cocrystal
Corresponding Author
*E-mail: ashwini.nangia@gmail.com.
FUROS
3400.0, 3351.0
3364.3, 3261.0
3347.2, 3270.1
3356.0, 3254.0
3325.7, 3243.0
3400.0, 3351.0
FUROSꢀBPNO
3392.7, 3326.0
3286.6, 3194.2
3297.2
2-BBS
4-MBS
2-CBS
PTS
2-BBSꢀBPNO
4-MBSꢀBPNO
2-CBSꢀBPNO
PTSꢀBPNO
3332.7, 3240.3
3290.0, 3238.1
3390.2, 3316.1
’ ACKNOWLEDGMENT
N.R.G. and N.J.B. thank the CSIR and UGC for fellowship.
We thank the DST (SR/S1/RFOC-01/2007 and SR/S1/OC-
67/2006) and CSIR (01(2079)/06/EMRII) for research fund-
ing, and DST (IRPHA) and UGC (PURSE grant) for instru-
mentation and infrastructure facilities.
FUROS
FUROSꢀINICNO
’ EXPERIMENTAL SECTION
Cocrystal Preparation. Starting materials were purchased from
commercial suppliers. Pyridine N-oxides were prepared by the oxidation
of the corresponding pyridines with m-CPBA. The crystallization scale,
number of experiments carried out, and the conditions used are
summarized in Table 3. A full list of experiments, crystallization
conditions and results are given in Table S4 (Supporting Information).
Broadly, two methods were adopted: (1) the components were dis-
solved in a suitable solvent and crystallization of new crystals was
examined, and (2) the components were ground with a few drops of
solvent added, and then the material was dissolved in a solvent for single
crystals to appear. The formation of a new crystalline phase was
monitored by changes in sulfonamide NH IR bands from starting
material to the product cocrystal followed by diffraction and spectro-
scopic confirmation as shown in Table 4.
X-ray Crystal Structure. Reflections were collected on a Bruker
SMART APEX-CCD diffractometer. MoꢀKR (λ = 0.71073 Å) radia-
tion was used to collect X-ray reflections on the single crystal. Data
reduction was performed using Bruker SAINT software.26 Intensities for
absorption were corrected using SADABS.27 RLATT3 and CELL_
NOW programs28 were used to generate a new reindexed cell for the
structure MBSꢀBPNO, which was used during initialization and mer-
ging of raw data. Crystal structures were solved by direct methods and
refined on F2 with SHELXS-97 and SHELXL-97 programs29 to give
satisfactory R factor. Hydrogen atoms on O and N were experimentally
located in difference electron density maps. All CꢀH atoms were fixed
geometrically using HFIX command in SHELX-TL. In the modeled
structure FUROSꢀINICNO the disorder was modeled for all the heavy
atoms with isotropic displacement parameters using the PART com-
mand by assigning sof (site occupancy factor) of 0.5 and 0.5 for the two
parts using the FVAR command. A check of the final CIF file using
PLATON30 did not show any missed symmetry. X-Seed31 was used to
prepare packing diagrams.
’ REFERENCES
(1) Desiraju, G. R. Angew. Chem., Int. Ed. 1995, 34, 2311.
(2) Walsh, R. D. B.; Bradner, M. W.; Fleischman, S. G.; Morales,
L. A.; Moulton, B.; Rodriguez-Hornedo, N.; Zaworotko, M. J. Chem.
Commun. 2003, 186.
(3) Allen, F. H.; Motherwell, W. D. S.; Raithby, P. R.; Shields, G. P.;
Taylor, R. New. J. Chem. 1999, 23, 25.
(4) Cambridge Crystallographic Data Center, Cambridge, U.K.,
www.ccdc.cam.ac.uk.
(5) (a) Nangia, A.; Desiraju, G. R. Top. Curr. Chem. 1998, 198, 57.
(b) Nangia, A.; Desiraju, G. R. Acta Crystallogr., Sect. A 1998, 54, 934.
(6) (a) Etter, M. C.; Macdonald, J. C.; Bernstein, J. Acta Crystallogr.,
Sect. B 1990, 46, 256. (b) Bernstein, J.; Davis, R. E.; Shimoni, L.; Chang,
N. L. Angew. Chem., Int. Ed. 1995, 34, 1555.
(7) (a) Etter, M. C. J. Phys. Chem. 1991, 95, 4601. (b) Etter, M. C.
Acc. Chem. Res. 1990, 23, 120.
(8) (a) Fleischman, S. G.; Kuduva, S. S.; McMahon, J. A.; Moulton,
B.; Walsh, R. D. B.; Rodriguez-Hornedo, N.; Zaworotko, M. J. Cryst.
Growth Des. 2003, 3, 909. (b) Vishweshwar, P.; McMahon, J. A.; Bis,
J. A.; Zaworotko, M. J. J. Pharm. Sci. 2006, 95, 499.
(9) (a) Bhogala, B. R.; Vishweshwar, P.; Nangia, A. Cryst. Growth
€
Des. 2002, 2, 325. (b) Almarsson, O.; Zaworotko, M. J. Chem. Commun.
2004, 1889. (c) Aaker€oy, C. B.; Salmon, D. J. CrystEngComm 2005,
7, 439. (d) Bhogala, B. R.; Nangia, A. New J. Chem. 2008, 32, 800.
(e) Aaker€oy, C. B.; Beatty, A. M.; Helfrich, B. A. Angew. Chem., Int. Ed.
2001, 40, 3240.
(10) (a) Reddy, L. S.; Babu, N. J.; Nangia, A. Chem. Commun.
2006, 1369. (b) Babu, N. J.; Reddy, L. S.; Nangia, A. Mol. Pharmaceutics
2007, 4, 417. (c) Babu, N. J.; Reddy, L. S.; Aitipamula, S.; Nangia, A.
Chem. ꢀ An Asian J. 2008, 3, 1122.
(11) Nangia, A. J. Chem. Sci. 2010, 122, 295.
Vibrational Spectroscopy. Nicolet 6700 FT-IR spectrometer
with a NXR FT-Raman and NIR Module was used to record IR and
Raman and NIR spectra. IR and NIR spectra were recorded on samples
dispersed in KBr pellets. Raman spectra were recorded on samples
contained in standard NMR diameter tubes or on compressed samples
contained in a gold-coated sample holder.
Thermogravimetric Analysis. TGA was performed on a Mettler
Toledo TGA/SDTA 851e module. The sample 8ꢀ12 mg was placed in
open alumina pan and the temperature increased from 30 to 400 °C at
heating rate of 5 °C/min. The sample was purged by a stream of dry
nitrogen at flow rate of 50 mL/min.
(12) (a) Yang, S. S.; Guillory, J. K. J. Pharm. Sci. 1972, 61, 26. (b) Ko,
S.-K.; Jin, H. J.; Jung, D.-W.; Tian, X.; Shin, I. Angew. Chem., Int. Ed.
2009, 48, 7809. (c) Caira, M. R. Mol. Pharmaceutics 2007, 4, 310.
(d) Byrn, S. R.; Pfeiffer, R. R.; Stowell, J. G. Solid-State Chemistry of
Drugs; SSCI, Inc.: West Lafayette, IN, 1999; p 160.
(13) (a) Ishida, T.; In, Y.; Doi, M.; Inoue, M.; Yanagisawa, I. Acta
Crystallogr., Sect. B 1989, 45, 505. (b) Leger, J. M.; Alberola, S.; Carpy, A.
Acta Crystallogr. 1977, B33, 1455. (c) Alberola, S.; Sabon, F.; Jaud et, J.;
Galy, J. Acta Crystallogr. 1977, B33, 3337. (d) Caira, M. R. J. Chem.
Crystallogr. 1994, 24, 695. (e) Caira, M. R.; Nassimbeni, L. R.; Wild-
erwanck, A. F. J. Chem. Soc., Perkin Trans. 1995, 2, 2213. (f) Caira, M. R.
J. Crystallogr. Spectrosc. Res. 1992, 22, 193.
(14) Eccles, K. S.; Elcoate, C. J.; Stokes, S. P.; Maguire, A. R.;
Lawrence, S. E. Cryst. Growth Des. 2010, 10, 4243.
(15) (a) Reddy, L. S.; Basavoju, S.; Vangala, V. R.; Nangia, A. Cryst.
Growth Des. 2006, 6, 161. (b) Videnova-Adrabiꢀnska, V.; Janeczko, E.
Chem. Commun. 1999, 1527. (c) Custelcean, R.; Moyer, B. A.; Bryantsev,
V. S.; Hay, B. P. Cryst. Growth Des. 2006, 6, 555. (d) Custelcean, R.
’ ASSOCIATED CONTENT
S
Supporting Information. IR, NIR, and Raman spectra;
b
table of crystallization experiments; TGA curve (PDF);
1938
dx.doi.org/10.1021/cg200094x |Cryst. Growth Des. 2011, 11, 1930–1939

Crystal Growth & Design
ARTICLE
Chem. Commun. 2008, 295. (e) Custelcean, R.; Sellin, B.; Moyer, B. A.
Chem. Commun. 2007, 1541.
(16) (a) Shan, N.; Toda, F.; Jones, W. Chem. Commun. 2002, 2372.
(b) Trask, A. V.; Jones, W. Top. Curr. Chem. 2005, 254, 131.
(17) (a) Trask, A. V.; Motherwell, W. D. S.; Jones, W. Chem.
Commun. 2004, 890. (b) Friꢁsꢁciꢀc, T.; Trask, A. V.; Jones, W.; Motherwell,
W. D. S. Angew. Chem., Int. Ed. 2006, 45, 7546.
(18) Reddy, L. S.; Bhatt, P.; Banerjee, R.; Nangia, A.; Kruger, G. J.
Chem. Asian J. 2007, 2, 505.
(19) (a) Berry, D. J.; Seaton, C. C.; Clegg, W.; Harrington, R. W.;
Coles, S. J.; Horton, P. N.; Hursthouse, M. B.; Storey, R.; Jones, W.;
Friꢁsꢁciꢀc, T.; Blagden, N. Cryst. Growth Des. 2008, 8, 1697. (b) Bettinetti,
G.; Caira, M. R.; Callegari, A.; Merli, M.; Sorrenti, M.; Tadini, C.
J. Pharm. Sci. 2000, 89, 478.
(20) (a) Saha, B. K.; Nangia, A.; Jaskꢀolski, M. CrystEngComm 2005,
7, 355. (b) Thaimattam, R.; Sharma, C. V. K.; Clearfield, A.; Desiraju,
G. R. Cryst. Growth Des. 2001, 1, 103.
(21) International Union of Crystallography, Chester, U.K., checkcif.
iucr.org.
(22) Sanphui, P.; Sarma, B.; Nangia, A. Cryst. Growth Des. 2010,
10, 4550.
(23) (a) Friedman, P. A.; Berndt, W. O. In Modern Pharmacology
with Clinical Applications, 5th ed.; Craig, C. R., Stitzel, R. E., Eds.; Little
Brown & Company: Boston, 1997; pp 239ꢀ255. (b) Khan, M. G.
Encyclopedia of Heart Diseases; Elsevier: New York, 2006.
(24) Babu, N. J.; Cherukuvada, S.; Thakuria, R.; Nangia, A. Cryst.
Growth Des. 2010, 10, 1979.
(25) (a) Feng, S.; Yang, M.; Zhang, Z.; Wang, Z.; Hong, D.; Richter,
H.; Benson, G. M.; Bleicher, B.; Grether, U.; Martin, R. E; Plancher,
J.-M.; Kuhn, B.; Rudolph, M. G.; Chen, L. Biorg. Med. Chem. Lett. 2009,
19, 2595. (b) McKeown, S. R.; Hejmadi, M. V.; McIntyre, I. A.; McAleer,
J. J.; Patterson, L. H. Br. J. Cancer 1995, 72, 76. (c) Nishida, C. R.; Lee,
M.; Ortiz de Montellano, P. R. Mol. Pharm. 2010, 78, 497.
(26) SAINT-Plus, version 6.45; Bruker AXS: Madison, WI, 2003.
(27) Sheldrick, G. M. SADABS, Program for Empirical Absorption
Correction of Area Detector Data; University of G€ottingen: G€ottingen,
Germany, 1997.
(28) (a) RLATT, Reciprocal Lattice Viewer, version 3.0; Bruker AXS:
Madison, WI, 2000. (b) Sheldrick, G. M. CELL_NOW; University of
G€ottingen: G€ottingen, Germany, 2004.
(29) (a) SMART (Version 5.625) and SHELX-TL (Version 6.12);
Bruker-AXS: Madison, WI, 2000. (b) Sheldrick, G. M. SHELXS-97and
SHELXL-97; University of G€ottingen:, G€ottingen, Germany, 1997.
(30) Spek, A. L. PLATON: A Multipurpose Crystallographic Tool;
Utrecht University: Utrecht, The Netherlands, 2002. Spek, A. L. J. Appl.
Crystallogr. 2003, 36, 7.
(31) Barbour, L. J. X-Seed, Graphical Interface to SHELX-97 and
POV-Ray; University of MissouriꢀColumbia: Columbus, MO, 1999.
1939
dx.doi.org/10.1021/cg200094x |Cryst. Growth Des. 2011, 11, 1930–1939